Research article Special Issues

A full convolutional network based on DenseNet for remote sensing scene classification

  • The convolutional neural networks (CNN) applied in remote sensing scene classification have two common problems. One is that these models have large number of parameters, which causes over-fitting easily. The other is that the network is not deep enough, thus more abstract semantic information cannot be extracted. To solve these two problems, we propose a simple and efficient full convolutional network based on DenseNet for remote sensing scene classification. We construct a small number of convolutional kernels to generate a large number of reusable feature maps by dense connections, which makes the network deeper, but does not increase the number of parameters significantly. Our network is so deep that it has more than 100 layers. However, it has only about 7 million parameters, which is far less than the number of VGGos parameters. Then we incorporate an adaptive average 3D pooling operation in our network. This operation fixes feature maps of size 7 × 7 from the last DenseBlock to 1 × 1 and decreases the number of channels from 1024 to 512, thus the whole network can accept input images with different sizes. Furthermore, we design the convolutional layer instead of the fully connected layer that is used as a classifier usually, so that the output features of the network can be classified without flattening operation, which simplifies the classification operation. Finally, a good model is trained by exploiting pre-trained weights and data augmentation technology. Compared with several state-of-the-art algorithms, our algorithm improves classification performance significantly on UCM dataset, AID dataset, OPTIMAL-31 dataset and NWPU-RESISC45 dataset.

    Citation: Jianming Zhang , Chaoquan Lu , Xudong Li , Hye-Jin Kim, Jin Wang. A full convolutional network based on DenseNet for remote sensing scene classification[J]. Mathematical Biosciences and Engineering, 2019, 16(5): 3345-3367. doi: 10.3934/mbe.2019167

    Related Papers:

    [1] Yufeng Li, Chengcheng Liu, Weiping Zhao, Yufeng Huang . Multi-spectral remote sensing images feature coverage classification based on improved convolutional neural network. Mathematical Biosciences and Engineering, 2020, 17(5): 4443-4456. doi: 10.3934/mbe.2020245
    [2] Haifeng Song, Weiwei Yang, Songsong Dai, Haiyan Yuan . Multi-source remote sensing image classification based on two-channel densely connected convolutional networks. Mathematical Biosciences and Engineering, 2020, 17(6): 7353-7377. doi: 10.3934/mbe.2020376
    [3] Ansheng Ye, Xiangbing Zhou, Kai Weng, Yu Gong, Fang Miao, Huimin Zhao . Image classification of hyperspectral remote sensing using semi-supervised learning algorithm. Mathematical Biosciences and Engineering, 2023, 20(6): 11502-11527. doi: 10.3934/mbe.2023510
    [4] Eric Ke Wang, Nie Zhe, Yueping Li, Zuodong Liang, Xun Zhang, Juntao Yu, Yunming Ye . A sparse deep learning model for privacy attack on remote sensing images. Mathematical Biosciences and Engineering, 2019, 16(3): 1300-1312. doi: 10.3934/mbe.2019063
    [5] Jing Wang, Jiaohua Qin, Xuyu Xiang, Yun Tan, Nan Pan . CAPTCHA recognition based on deep convolutional neural network. Mathematical Biosciences and Engineering, 2019, 16(5): 5851-5861. doi: 10.3934/mbe.2019292
    [6] Eric Ke Wang, Fan Wang, Ruipei Sun, Xi Liu . A new privacy attack network for remote sensing images classification with small training samples. Mathematical Biosciences and Engineering, 2019, 16(5): 4456-4476. doi: 10.3934/mbe.2019222
    [7] Song Yang, Huibin Wang, Hongmin Gao, Lili Zhang . Few-shot remote sensing scene classification based on multi subband deep feature fusion. Mathematical Biosciences and Engineering, 2023, 20(7): 12889-12907. doi: 10.3934/mbe.2023575
    [8] Liguo Zhang, Liangyu Zhao, Yongtao Yan . A hybrid neural network-based intelligent body posture estimation system in sports scenes. Mathematical Biosciences and Engineering, 2024, 21(1): 1017-1037. doi: 10.3934/mbe.2024042
    [9] Liyong Ma, Wei Xie, Haibin Huang . Convolutional neural network based obstacle detection for unmanned surface vehicle. Mathematical Biosciences and Engineering, 2020, 17(1): 845-861. doi: 10.3934/mbe.2020045
    [10] Wenjuan Gu, Hongcheng Wang, Xiaobao Liu, Yanchao Yin, Biao Xu . Urban scene segmentation model based on multi-scale shuffle features. Mathematical Biosciences and Engineering, 2023, 20(7): 11763-11784. doi: 10.3934/mbe.2023523
  • The convolutional neural networks (CNN) applied in remote sensing scene classification have two common problems. One is that these models have large number of parameters, which causes over-fitting easily. The other is that the network is not deep enough, thus more abstract semantic information cannot be extracted. To solve these two problems, we propose a simple and efficient full convolutional network based on DenseNet for remote sensing scene classification. We construct a small number of convolutional kernels to generate a large number of reusable feature maps by dense connections, which makes the network deeper, but does not increase the number of parameters significantly. Our network is so deep that it has more than 100 layers. However, it has only about 7 million parameters, which is far less than the number of VGGos parameters. Then we incorporate an adaptive average 3D pooling operation in our network. This operation fixes feature maps of size 7 × 7 from the last DenseBlock to 1 × 1 and decreases the number of channels from 1024 to 512, thus the whole network can accept input images with different sizes. Furthermore, we design the convolutional layer instead of the fully connected layer that is used as a classifier usually, so that the output features of the network can be classified without flattening operation, which simplifies the classification operation. Finally, a good model is trained by exploiting pre-trained weights and data augmentation technology. Compared with several state-of-the-art algorithms, our algorithm improves classification performance significantly on UCM dataset, AID dataset, OPTIMAL-31 dataset and NWPU-RESISC45 dataset.


    Remote sensing images has been applied to a broad range of fields such as land use classification [1,2], geographic image retrieval [3,4] and remote sensing scene classification. Remote sensing scene classification is an important work, which can provide some decision-making basis for the follow-up work of urban planning and environmental monitoring. Remote sensing scene classification divides remote sensing images into various categories such as forests, cities and rivers according to the semantic information of remote sensing images. Remote sensing images we focus on are obtained by aerial photography. These remote sensing images have a high spatial resolution, highly complex geometric structures and spatial patterns. However, there is high similarity between categories of remote sensing while images in the same scene category may have great differences. This makes it difficult to classify remote sensing scenes.

    Multimedia information processing and communication are popular in our daily lives, and security and privacy protection techniques are becoming more and more important [5]. Remote sensing images are almost saved in the digital form, therefore they are possible to be tampered with. It should be noted that for some remote sensing scene images, modifying a small number of pixels can affect the classification results greatly, which may cause some risks. This has become another challenge in the classification of remote sensing scenes. Because of these challenges mentioned above, the classification models should be supposed to extract the semantic information from the image very well to achieve the better robustness.

    Researchers pay more attention to the design of handcrafted features that are used for remote sensing scene classification before the development of deep learning. Those low-level features designed by the human is hard to represent semantic information of remote sensing scenes. The classification accuracy is unsatisfactory in the methods using artificial features. With the development of deep learning, deep learning has been applied to many research field [6,7,8,9]. The convolution neural network (CNN), as the most popular technology of deep learning, is introduced to remote sensing scene classification naturally. In general, the neural networks can be trained very well with a large amount of training data while they are often over-fitting and poor at generalization with a small amount of training data. To avoid or alleviate this situation, on the one hand, data enhancement technology can be used to expand the training data [10,11], on the other hand, a specially designed network also can make the training more efficient [12].

    Some neural network architectures are applied to remote sensing scene classification, such as CaffeNet, GoogleNet, and VGGNet. Although many methods based on those networks improve classification accuracy significantly, they still have a bottle-neck. For these networks, we observe that:

    1) These networks that are trained with small amounts of data have a large number of parameters. Note that this often results in over-fitting in convolutional neural networks.

    2) These networks are shallow. Because of the complexity of remote sensing images, to better classify remote sensing scenes, classification network is supposed to extract high-level semantic features of images well. It is worth noting that deeper networks often can extract more abstract semantic information.

    In view of the above observations, we propose a deeper network with fewer parameters to improve the classification accuracy of remote sensing scene. Based on the DenseNet, we built a network with deeper layers but fewer parameters than those networks mentioned above. Table 1 shows the comparison of parameters and layers of different networks. To increase the diversity of training data, we use data augmentation technology on training dataset, such as brightness, color, contrast, sharpness and rotation when training the network. All of the parameters of those augmentation transformations are generated randomly. In addition, we utilize pre-training parameters trained on ImageNet [13], which can accelerate training and overcome over-fitting. Moreover, we fix the length of the final feature vector that is used for classification by an adaptive average 3D pooling operation. Thus, the network can accept images of various sizes. To avoid the flattening operation and make the network operation more concise, we design a convolutional layer instead of fully connected layer to classify the features. Unlike many methods that use Support Vector Machines (SVM) as a classifier, we use the softmax as the probability mapping function, so our network is an end-to-end process without additional training on SVM.

    Table 1.  Comparison of different network parameters and layers.
    Network Parameters (millions) Convolution layers Fully connected layers
    CaffeNet 61.1 5 3
    GoogleNet 6.7 22 1
    VGG16 138.3 13 3
    VGG19 143.6 16 3
    ours 7.5 121 0

     | Show Table
    DownLoad: CSV

    The methods of remote sensing scene classification are mainly based on three aspects at present: low-level visual features, middle-level visual features and high-level visual features [14].

    Low-level visual features are used in many tasks, such as object tracking [15]. There are some frequently-used low-level visual features in the field of classification of remote sensing scenes, such as spectral feature, texture feature and structure feature. Those methods based on low-level visual features either extract feature vector from these low-level visual features or use a combination of feature vectors extracted from different low-level visual features to describe remote sensing images [16,17,18,19]. This means that such methods are based on the premise that remote sensing images can be described by a feature vector or a combination of different kinds of feature vectors. For some images whose scene are simple, one kind of feature can be used as a descriptor of remote sensing images [16,20]. Unlike using only one kind of feature, literatures [17,18,19] show that the combinations of different kind of features can improve the results. Specifically, [17] combine six different kinds of feature to form a multiple-feature representation, which describes the scene better. However, low-level visual features are only applicable to simple scenarios that have no complex structure. For complex scenarios, their expressive ability is limited [21].

    To overcome the limitations of low-level visual features, many methods based on middle-level visual features are proposed. The main idea of such methods is to encode the extracted features such as SIFT, LBP and color histograms. Then further extract the more discriminant feature representation. Bag of Word (BoVW) model [22] and its related variants [23,24,25,26,27] are typical algorithms based on the middle-level visual features. The BoVW regards an image as a collection of visual words, and to construct new features that are used for image classification with the frequency of visual words. Specific steps are as follows: it first extracts SIFT features from all images, next constructs a dictionary using a clustering algorithm, and then counts the occurrence frequency of each word of the dictionary in the image. The features used as a descriptor of an image for classification is composed of these occurrence frequencies of each word of the dictionary. The features obtained by such methods are a further abstraction of low-level visual features, lacking flexibility and adaptability for different scenarios [14].

    The third category of methods are based on high-level visual features. Such methods achieve the best accuracy comparing with the above two kinds methods, even far than algorithms based on low-level visual features [28,29,30,31,32,33,34,35]. As mentioned above, neither low-level visual features nor middle-level visual features can represent remote sensing images well and flexibly, which means that the features obtained by these two methods lack abstract semantic information. As we all know, CNN can extract abstract semantic information from images. Thus, those features extracted by CNN are made to be a good descriptor of remote sensing images. In addition, CNN can learn how to extract the high-level semantic information of an image by itself, which is very flexible and avoids complex design of artificial features.

    At present, CNN models used to the classification of remote sensing scene mainly include CaffeNet, GoogleNet and VGGNet [33,36,37,38,39]. Usually, these networks are shallow with no more than 30 layers, and the number of their parameters is large, such as VGG16 that has about 138 million parameters. Chaib et al. [37] use a VGGNet to extract the features from remote sensing images, then the features from different layers are transformed by discriminant correlation analysis (DCA), the transformed features are fused to construct a new feature that is classified by SVM that is used as a classifier in other fields [40]. The main idea of literature [39] is also features fusion. Different from [37], Yu et al. [39] proposed two schemes. One is that using the original RGB network and the mapped LBP coded network extracts the features from RGB images and LBP feature maps respectively. The features from different network are fused by concatenation layer, and then the obtained features are classified after being processed by the fully connected layers. The steps of the other one are similar. The difference is that a saliency coded network is used instead of mapped LBP coded network. Wang Q et al. [38] introduced the attention mechanism into the classification of remote sensing scene by constructing a recurrent attention structure with recurrent neural networks (RNN). Different from the above methods, Cheng G et al. [36] focus on the loss function, and propose a new loss function by applying metric learning to CNN features. By combining proposed metric learning loss with cross-entropy loss, features extracted from the same scene class are as close as possible, while features extracted from different categories are as far as possible.

    As mentioned above, the CNN model generally used in remote sensing scene classification have huge number of parameters, or the depth of network layer is too shallow, which is easy to cause over-fitting of the model and inadequate feature extraction ability of the model. Aiming at the problem of network applied in remote sensing scene classification, we apply the DenseNet [41] to our architecture and construct a full convolution network for remote sensing scene classification. Based on cross-layer connectivity, our full convolutional network is a deep network architecture with only a few parameters. In addition, we introduce an adaptive average 3D pooling operation, which fixes the sizes of the output feature maps from 7 × 7 to 1 × 1 and decrease the number of channels from 1024 to 512. Our method can accept images that are different sizes. Furthermore, our network exploits a convolutional layer instead of full connection layer as a classification layer. Thus, features can be classified without flattening, which makes the network simpler and more efficient.

    In 2017, Huang et al. [41] proposed DenseNet derived from [42]. In general, the network structures are progressively hierarchical. In such a network structure, the feature maps from the (i1)th layer are the input of the ith layer. Experiments [42] have shown that the input of the ith layer can be not only the output of the (i1)th layer, but also the (i2)th layer, or the (in)th layer (n less than the number of layers). This may lead to a more generalizable network. The basic idea of DenseNet is that each layer in the network is directly connected to the front layers. The connection strategy in DenseNet is shown in Figure 1. It is worth noting that in order to ensure that the feature maps can be concatenated, the sizes of feature maps need to be consistent. This means that the outputs of the convolutional layer are of the same size as the input.

    Figure 1.  Densely concatenated convolution. The input of each layer comes from the output of all previous layers, and the final output of the network fuses all feature maps produced by front layers.

    The input of each layer is the output of all front layers in the densely concatenated convolutions:

    Xl=Hl([X0,X1,...,Xl1]) (1)

    Hl represents the lth layer, Xl represents the output of the lth layer. In DenseNet, Batch Normalizing (BN) [43] is used to normalize the input of layer, which reduces the absolute difference between data and considers the relative difference more. The BN algorithm is described as the Algorithm 1.

    Table Algorithm 1.  Batch Normalizing Transform.
    Input: Values of x over a batch: B={x1...m}; Parameters to be learned: γ, β
    Output: {yi=BNγ,β(xi)}
    (1) μB1mmi=1xi
    (2) σ2B1mmi=1(xiμB)2
    (3) xixiμBσ2B+ε
    (4) yiγxi+βBNγ,β(xi)

     | Show Table
    DownLoad: CSV

    The more details about BN you can get from [43]. In order to increase the nonlinearity of network, ReLU is used as the activation function, which is described as:

    ReLU(x)={xx>00x0 (2)

    Compared with other networks, the network composed of Densenet has the following advantages:

    1) Fewer parameters. The input of current layer of the network is composed of the feature maps from front layers, which results in the continuous accumulation of the number of feature maps. Thus, many reusable feature maps can be learned with very few convolutional kernels.

    2) Strong ability to prevent over-fitting. Short paths are created from early to late layers because of the dense connection. Each layer receives additional supervision from loss function, which greatly reduces the gradient disappearance. Therefore, dense connection has a very good ability to resist over-fitting, which makes it especially suitable for applications where training data is relatively scarce.

    3) Deeper layers. The network can be designed very deep by the dense connection, which only have a few parameters. For example, a network with 96 convolution layers can be designed while it only has 5.34 million parameters.

    Experiments on ImageNet [13] demonstrate the efficiency of the dense connection [41]. Therefore, densely concatenated convolution can extract high-level features of remote sensing images. Based on this idea, we construct a full convolution network based on DenseNet [41] for remote sensing scene classification.

    Our full convolutional network based on DenseNet is shown in Figure 2. In the whole network, it is divided into an initial layer, three transition layers and four densely connected blocks (DenseBlock). After the last DenseBlock, we exploit an adaptive average 3D pooling operation to fix number and size of feature maps and then classify them by a convolution with kernel size of 1 × 1. We describe our architecture in detail in session 3.3.

    Figure 2.  Our full convolutional network for remote sensing scene classification. In our network, each densely connected block (DenseBlock) consists of several Tuple Operations (TO). Each TO is composed of Batch Normalizing (BN), ReLu, Conv(1 × 1) and Conv(3 × 3). Conv(1 × 1) represents a convolutional operation with kernel size of 1 × 1 while Conv(3 × 3) represents a convolutional operation with kernel size of 3 × 3. DenseBlock1 has 6 TOs, DenseBlock2 has 12 TOs, DenseBlock3 has 24 TOs and DenseBlock4 has 16 TOs.

    As shown in Figure 2, the whole network architecture does not strictly follow the rule of dense connection, it mainly consists of densely connected blocks instead. The reason for this is that the whole network has several down-sampling operations on feature maps, which results in the inconsistency of sizes between feature maps in different stages, so that feature maps with different sizes cannot be spliced in the dimension of channel. Moreover, concatenation operation requires feature maps to be kept in memory. With the increase of network layers, feature maps kept in memory are superimposed, which causes huge memory consumption. If every layer of the network strictly follows the dense connection theory, the network will consume many resources, which makes the network stop working.

    In this section, we detail the various parts of our network architecture.

    Initial layer. We first extract 64 features using 64 convolutional kernels with kernel size of 7 × 7. We set the convolutional stride as 2, which reduces the sizes of the feature maps and reduces the computational complexity of subsequent operations. After the convolution, the operations of BN, Relu and MaxPool are performed in turn. For example, when the size of the input image is 224 × 224, the size of the image after padding is 230 × 230 and convolution kernels of size 7 × 7 are used for convolution operation, then we will get feature maps of size 112 × 112. Subsequently, these feature maps will go through the operations of BN, Relu and MaxPool in turn. Finally, the obtained feature maps are size of 56 × 56. The generated feature maps are used as input of the subsequent module of the network.

    Densely connected block. Our full convolutional network mainly consists of densely connected blocks (DenseBlock). Every DenseBlock acts as one stage for extracting features and consists of several Tuple Operations (TO). As shown in the Figure 2, a TO is composed of Batch Normalizing (BN), ReLU, Conv(1 × 1) and Conv(3 × 3). In particular, in all densely connected blocks, the stride of Conv(1 × 1) is 1 and the number of the convolutional kernels is 128. Moreover, the stride of Conv(3 × 3) is 1 and the number of convolutional kernels is 32. To ensure the same size of the output and input, the padding of Conv(3 × 3) is set to 1. It is worth mentioning that the dense connection operations are strictly enforced between every TO in DenseBlock. The structure of dense connection is shown as Figure 1.

    Transition layer. It is observed that, this layer consists of BN, Relu, Conv(1 × 1) and AvgPool. Feature maps pass through these operations in turn. After a DenseBlock, the output of each layer in this DenseBlock is kept in memory, resulting in a large number of feature maps. Note that if the number of these feature maps does not decrease, the network will be difficult to deepen and may stop working due to too much computation. Therefore we reduce the number of the feature maps to half by using a convolutional layer with a convolution kernel size of 1 × 1 after a DenseBlock, which can not only reduce the number of feature maps but also increase the depth and improves fitting ability of the network.

    Adaptive average 3D pooling. After 3 transition layers and 4 densely connected blocks, the obtained feature map size is 1/32 of the original image, and the number of feature maps is 1024. To further reduce the computational consumption, we need to reduce the number and size of the feature map. Through the adaptive average 3D pooling operation, the number of feature maps is reduced to 512 and the size becomes 1 × 1 without increasing the network parameters. As shown in Figure 2. In addition, this operation ensures that the size of output feature maps is 1 × 1 and the number is 512, which provides a fixed size feature map for the later convolutional classifier and makes the network accept images with different sizes. As shown in Figure 3.

    Figure 3.  Our network can accept images of different sizes. After extracting the features from input images of sizes 3 × 224 × 224, 3 × 280 × 280, 3 × 300 × 300 by previous layers, the obtained feature map sizes are 1024 × 7 × 7, 1024 × 8 × 8, 1024 × 9 × 9 respectively. The sizes of the feature maps will be 512 × 1 × 1 after the adaptive average 3D pooling operation. Adaptive average 3D pooling operation produces a fixed size feature map for the later convolution classifier.

    Convolution classifier. After adaptive average 3D pooling, a 512-dim feature representation is obtained. Then we exploit a 1 × 1 convolution to classify the 512-dim features and get the final category feature representation. Finally, we exploit the softmax function to perform the probability mapping.

    Softmax function. This function maps the final category feature representation into a probability for every class. The function is defined as:

    P(c|v)=exp(Xc)nj=1exp(Xj) (3)

    Xcv, Xc represents the cth number of vector v that comes from convolution classifier.

    In many classification tasks, the loss function is designed as:

    L(x;θ)=J(θ)=1nni=1log(P(c|hθ(xi))) (4)

    In equation (4), hθ is defined as mapping function of the network with parameters θ, P(c|hθ(xi)) indicates the probability of real category for the ith sample, and n is the number of samples used for one epoch. In order to prevent over-fitting, we normalize the network parameters by L2-norm.

    L2 - norm(θ)=12nθλniθi2 (5)

    In equation (5), nθ represents the number of parameters θ of the network. Our ultimate loss function with L2-norm becomes:

    L(x;θ) = J(θ)=1nni=1log(P(c|hθ(xi)))+12nθλniθi2 (6)

    In equation (6), λ is defined as weight decay rate. In our experiment, we set λ to 5×10-4.

    There are many algorithms that can be used to optimize the parameters of neural network, such as genetic algorithm [44] and gradient descent algorithm. It is worth noting that the gradient descent algorithm, the most commonly used algorithm in deep learning, is used to optimize the parameters of the network in our experiments. For gradient descent algorithm, there are various optimizers can be selected, such as Adam and SGD. According to our experience, SGD usually takes much more time to train, but the results are more reliable in the case of a good initialization and learning rate scheduling scheme. Because we use pre-training parameters, which satisfies the condition having a good initialization, so we use SGD as the optimization strategy for training network. For general SGD, the parameters θ of the network are updated as follows:

    θ = θ - ηθJ(θ) (7)

    In equation (7), the gradient θJ(θ) is defined as:

    θJ(θ) = J(θ)θ=L(x,θ)θ (8)

    In equation (7), η is defined as the learning rate. One disadvantage of this type of SGD is that its update is very unstable because its update direction is completely dependent on the current batch. A simple way to solve this problem is to introduce the momentum. Momentum simulates the inertia of an object when it is moving, that is, it retains the direction of the previous update to a certain extent when updating, and uses the gradient of the current batch to fine tune the final update direction. In this way, stability can be increased to a certain extent, so that the learning is faster, and there is a certain ability to get rid of the local optimum. For SGD with momentum, the parameters θ of the network are updated as follows:

    vm×v+θJ(θ) (9)
    θθηv (10)

    In equation (8) and (9), m and v denote momentum and velocity respectively. In our experiment, we set m to 0.9.

    We adopt confusion matrix and overall classification accuracy, which are widely used in classification tasks to evaluate the effectiveness of our algorithm.

    In the confusion matrix, each row represents the actual category, and each column represents the predicted category. The number of classes correctly classified and the number of classes misclassified can be clearly seen by the confusion matrix.

    Overall accuracy is the ratio of the number of models correctly predicted on all test sets to the total number of datasets. The overall accuracy can generally characterize the classification accuracy of model.

    OA=TP+TNTP+FN+FP+TN (11)

    TP represents true samples that are correctly classified by the model. FN represents positive samples that are misclassified by the model. FP represents negative samples that are misclassified by the model. TN represents negative samples that are correctly classified by the model.

    In order to verify the effectiveness of our algorithm, we have done a lot of experiments. We experiment on four data sets: UC Merced Land Use Dataset [22], Aerial Image Dataset [14], OPTIMAL-31 Dataset [38] and NWPU-RESISC45 [45] Dataset. These datasets contain a lot of images and cover a lot of categories. In addition, images of these datasets are pretty complex and have some similarities among different scenes, which make these datasets challenging. Some pictures of these datasets are shown in Figure 4.

    Figure 4.  Some samples of AID, UCM, OPTIMAL-31, NWPU-RESISC45 datasets. From these samples, we can see that remote sensing scene image scene is complex and contains a lot of interference information, so remote sensing scene classification is a challenging task.

    UC Merced Land-Use Dataset (UCM) is a scene classification dataset containing 21 categories. This dataset includes agricultural, golf course, medium density residential, airplane, baseball diamond, beach, buildings, chaparral, storage tanks, dense residential, overpass, freeway, harbor, intersection, mobile home park, parking lot, river, runway, forest, sparse residential, and tennis courts. There are 100 RGB color images of the size 256 × 256 and 30 cm spatial resolution for each category.

    Aerial Image Dataset (AID) includes bare land, dense residential, baseball field, bridge, center, church, railway station, commercial, viaduct, desert, storage tanks, farmland, forest, industrial, meadow, medium residential, beach, mountain, park, parking, playground, pond, port, resort, river, school, square, stadium, sparse residential and airport. This data set has 30 categories and 10000 images in all. Pixel resolution of each picture is 600 × 600.

    A remote sensing scene classification dataset named OPTIMAL-31 constructed by Qi Wang et al [38]. The dataset consists of 31 classes, each of which consists of 60 images with 256 × 256 pixels. It has 1860 pictures in total. The categories of the whole dataset are playground, airplane, airport, crossroads, runway, basketball court, bridge, parking lot, lake, bushes, church, beach, mountain, round farmland, business district, roundabout, dense houses, desert, forest, freeway, golf field, harbor, factory, island, meadow, medium houses, mobile house area, overpass, railway, square farmland, and baseball field. The whole dataset contains many categories, but the number of each category is small, so the dataset is more challenging.

    The NWPU-RESISC45 dataset covers 45 scene classes, and each category has 700 images. The 45 scene classes include chaparral, airplane, airport, baseball, diamond, golf course, basketball court, beach, freeway, bridge, church, circular farmland, cloud, commercial area, rectangular farmland, river, dense residential, roundabout, desert, forest, palace, ground track field, harbor, sparse residential, industrial area, thermal power station, intersection, sea ice, island, lake, meadow, medium residential, mobile home park, storage tank, mountain, overpass, parking lot, railway, railway station, runway, ship, snowberg, stadium, tennis court, terrace and wetland.

    In order to evaluate the effectiveness of our network, we train and test our method on UCM, AID, OPTIMAL-31 and NWPU-RESISC45. Each dataset is divided into two parts in our experiments, one for training and the other for testing. We design seven experiments in all to evaluate our network. The ratios of training and testing on each data set are as shown in Table 2. In each experiment, we randomly generate training set and testing set to ensure the validity of the experiment. Note that the training set is only used for training, while the testing set is only used for testing.

    Table 2.  The setting of training and testing scales on each data.
    Dataset Train Test
    UCM 80% 20%
    50% 50%
    AID 20% 80%
    50% 50%
    NWPU-RESISC45 10% 90%
    20% 80%
    OPTIMAL-31 80% 20%

     | Show Table
    DownLoad: CSV

    In our experiments, we set train epoch to 400, each epoch will train all images of a training set. The batch size of each epoch is set to 80. After each two epochs, we will evaluate our network on the testing set once. To reduce memory usage, we resize the images used for training to the size of 224 × 224. Ones more, we initialize our network with pre-trained parameters trained on the ImageNet. Further, we use data augmentation technology to train a better model. All experiments are performed on a GPU server. The configurations of server are as follow: CPU Intel (R) Xeon (R) E5-2670 v3,128 GB of memory, GeForce GTX TITAN X, 12GB of display memory. The experimental framework is pytorch0.4.1.

    The following four obfuscation matrices are obtained by testing our network on AID, UCM, OPTIMAL and NWPU-RESISC45, respectively.

    From Figure 58, we can see that most images can be correctly classified in our experiments. Especially in UCM dataset, only two images are misclassified. Among these experiments, the overall accuracy of NWPU-RESISC is the worst, only 94.98%. This is because there are many categories in the dataset while samples for training are few. For the case of misclassification, there are three main reasons.

    Figure 5.  The confusion matrix with AID, 50% of dataset for training and 50% of the dataset for testing. The overall accuracy is 97.44%.
    Figure 6.  The confusion matrix with UCM, 80% of the dataset for training and 20% of the dataset for testing. The overall accuracy is 99.50%.
    Figure 7.  The confusion matrix with OPTIMAL, 80% of the dataset for training and 20% of the dataset for testing. The overall accuracy is 95.89%.
    Figure 8.  The confusion matrix with NWPU-RESISC45, 20% of the dataset for training and 80% of the dataset for testing. The overall accuracy is 94.98%.

    1) The appearance is similar. The appearance of the two images shown in Figure 9(a) is very similar.

    Figure 9.  The cases of misclassification. In our experiment, the palace is misclassified into church easily, rectangular farmland is misclassified into terrace and wetland is misclassified.

    2) Texture is similar. As shown in Figure 9(b), two images are divided by roads, which causes a similar texture structure.

    3) One category may contain information of other categories. As shown in Figure 9(c), there is a lake-like area in the wetland, it is disturbed when wetland image is classified.

    To evaluate the performance of our algorithm, we compare it with some algorithms in recent years. From Table 3 to 6, we report the comparison results on UCM, AID, OPTIMAL-31 and NWPU-RESISC45 respectively.

    Table 3.  Overall accuracies and standard deviations (%) of different feature fusion methods on the UCM dataset.
    Method Training Ratios (%)
    50% 80%
    MS-CLBP + FV [46] 88.76 ± 0.76 93.00 ± 1.20
    GoogLeNet [14] 92.70 ± 0.60 94.31 ± 0.89
    CaffeNet [14] 93.98 ± 0.67 95.02 ± 0.81
    VGG-VD-16 [14] 94.14 ± 0.69 95.21 ± 1.20
    SalM3LBPCLM [47] 94.21 ± 0.75 95.75 ± 0.80
    TEX-TS-Net(feature fusion model) [39] 97.55 ± 0.46 98.40 ± 0.76
    SAL-TS-Net(feature fusion model) [39] 97.79 ± 0.56 98.90 ± 0.95
    ARCNet-VGG16 [38] 96.81 ± 0.14 99.12 ± 0.40
    ours 98.67 ± 0.42 99.50 ± 0.25

     | Show Table
    DownLoad: CSV
    Table 4.  Overall accuracies and standard deviations (%) of different feature fusion methods on the AID dataset.
    Method Training Ratios (%)
    20% 50%
    CaffeNet [14] 86.86 ± 0.47 89.53 ± 0.31
    VGG-VD-16 [14] 86.59 ± 0.29 89.64 ± 0.36
    GoogLeNet [14] 83.44 ± 0.40 86.39 ± 0.55
    ARCNet-VGG16 [38] 88.75 ± 0.40 93.10 ± 0.55
    D-CNN with GoogLeNet [36] 88.79 ± 0.10 96.22 ± 0.10
    D-CNN with VGGNet-16 [36] 90.82 ± 0.16 96.89 ± 0.10
    TEX-TS-Net(feature fusion model) [39] 93.31 ± 0.11 95.17 ± 0.21
    SAL-TS-Net(feature fusion model) [39] 94.09 ± 0.34 95.99 ± 0.35
    ours 95.37 ± 0.32 97.19 ± 0.23

     | Show Table
    DownLoad: CSV
    Table 5.  Overall accuracies and standard deviations (%) of different feature fusion methods on the OPTIMAL-31 dataset.
    Method Training Ratios (%)
    80%
    Fine-tuning AlexNet [38] 81.22 ± 0.19
    Fine-tuning GoogLeNet [38] 82.57 ± 0.12
    Fine-tuning VGGNet16 [38] 87.45 ± 0.45
    VGG-VD-16 [14] 89.12 ± 0.35
    ARCNet-Alexnet [38] 85.75 ± 0.35
    ARCNet-ResNet34 [38] 91.28 ± 0.45
    ARCNet-VGGNet16 [38] 92.70 ± 0.35
    ours 95.41 ± 0.61

     | Show Table
    DownLoad: CSV
    Table 6.  Overall accuracies and standard deviations (%) of different feature fusion methods on the NWPU-RESISC45 dataset.
    Method Training Ratios (%)
    10% 20%
    TEX-TS-Net(feature fusion model) [39] 84.77 ± 0.24 86.36 ± 0.19
    SAL-TS-Net(feature fusion model) [39] 85.02 ± 0.25 87.01 ± 0.19
    D-CNN with AlexNet [36] 85.56 ± 0.20 87.24 ± 0.12
    D-CNN with GoogLeNet [36] 86.89 ± 0.10 90.49 ± 0.15
    D-CNN with VGGNet-16 [36] 89.22 ± 0.50 91.89 ± 0.22
    ours 92.90 ± 0.17 94.95 ± 0.04

     | Show Table
    DownLoad: CSV

    From the Table 36, we can get the following conclusions.

    1) CNN regard as an advanced feature extractor is suitable for classification of remote sensing scenes, and it has achieved a good accuracy. As shown in Table 3, with 80% of dataset for training, the accuracy of ARCNet-VGG16 [38] is 99.12 ± 0.40, while our algorithm achieves 99.50 ± 0.25, which improves by about 0.38. It is worth noting that the accuracy of those methods based on CNN are very close to 100% in many datasets. This fully demonstrates the efficiency of convolutional neural networks.

    2) In contrast to various algorithms, our algorithm gets a state-of-art recognition rate. As shown in Table 3, we are 0.88 and 0.38 higher than the best algorithms at the training ratios 50% and 80% respectively. As shown in Table 4, we are 1.28 and 0.3 higher than the best algorithm at the training ratios 20% and 50% respectively. As shown in Table 5, we are 2.71 higher than the best algorithm at the training ratio 80%. As shown in Table 6, we are 3.68 and 3.06 higher than the best algorithm at the training ratios 10% and 20% respectively. This fully demonstrates the superiority of our algorithm and the excellent ability to extract features by dense connections.

    3) For CNN, the more data used for training, the better results for test. The results are reported in Table 3 show that whether it is our method or other deep learning models, the training ratio used is different, and the accuracy of the test is different. It is worth noting that as the training ratio increases, the test accuracy is also improved.

    Literature [33] shows that making full use of the pre-training model can improve the classification accuracy of the network. In our work, one of the keys to improve the classification accuracy of our network is to initialize network with the pre-training parameters trained on ImageNet [13]. In our experiments, the parameters of the whole network (except the parameters of the convolution classifier) are trained on ImageNet first, and then are further fine-tuned on the remote sensing dataset. The other key point for our network to classify remote sensing images well is using data augmentation technology. As we all know, big data is the key to the success of deep learning. However, in the field of remote sensing, the number of images in the dataset is often small, because the images are difficult to obtain, and it is time-consuming to manually mark the corresponding scene categories. To solve this problem, data augmentation technology is introduced. In our experiments, we performed color transformation, brightness transformation, contrast transformation, and sharpness transformation on images. Besides those transformations, we rotated the image at random angles. The effect of data augmentation is shown in Figure 11. To demonstrate the effectiveness of pre-training models and data augmentation technique, we conduct experiments on various datasets. The report is shown in Figure 10. Note that when the network does not be initialized with pre-training parameters and does not use the data augmentation technology the classification accuracy of the network is very poor. But the accuracy will be greatly improved when using the pre-training parameters and the data augmentation technology. Thus, using pre-training parameters and data augmentation techniques is effective for classification and may lead to improved classification accuracy.

    Figure 10.  The experimental results of whether to use the pre-training parameters and data augmentation technology on dataset. PMN_DEN denotes that pre-training parameters and data augmentation technique are not both used. PMN_DEY denotes that pre-training parameters is not used but a data augmentation technique is used. PMY_DEN denotes that pre-training parameters is used but a data augmentation technique is not used. PMY_DEY denotes that pre-training parameters and data augmentation technique are both used. The datasets are split as follows: For OPTIMAL, 80% of dataset are treated as training set, while 20% are treated as testing set. For AID 50% of dataset are treated as training set, while the 50% are treated as testing set. For NWPU-RESISC45, 10% of dataset are treated as training set, while the 90% are treated as testing set. For UCM, 80% of dataset are treated as training set, while the 20% are treated as testing set.
    Figure 11.  The category of the original picture is a forest. The pictures A, B, C, D and E obtained are changed by brightness, color, contrast, rotate and sharpness, respectively.

    In this paper, we exploit the advantages of densely concatenated convolution to propose a full convolutional network for remote sensing scene classification. Our network is deep, but the parameters are few. As shown in Figure 2, these TOs directly contribute to network depth. The numbers of convolutional kernels of Conv(3 × 3) and Conv(1 × 1) are 32 and 128 respectively in DenseBlock, which produces few parameters, this is the reason that our network has few parameters. We use an adaptive average 3D pooling operation to down-sample the feature maps of the network output. The number of channel decrease to 512, which reduces the calculated consumption for the latter operations. And the size of the feature map is fixed, so that network can accept images with different sizes.

    The experiment results on the several common datasets show that our fully convolutional network gets a state-of-art recognition rate.

    The research work was funded by the National Natural Science Foundation of China grants nos. 61772454, 61811540410 and 61811530332, the Scientific Research Fund of Hunan Provincial Education Department Grant no. 16A008, the Industry-University Cooperation and Collaborative Education Project of Department of Higher Education of Ministry of Education Grant no. 201702137008, the Postgraduate Scientific Research Innovation Fund of Hunan Province Grant no. CX2018B565, the Undergraduate Inquiry Learning and Innovative Experimental Fund of CSUST Grant no. 2018-6-119, and the Postgraduate Training Innovation Base Construction Project of Hunan Province Grant no. 2017-451-30.

    The authors declare no conflict of interest.



    [1] X. W. Yao, J. W. Han, G. Cheng, et al., Semantic annotation ofhigh-resolution satellite images via weakly supervised learning, IEEE Trans. Geosci. Remote Sens., 54 (2016), 3660–3671.
    [2] S. Y.Cui and M. H. Datcu, Comparison of approximation methods to Kullback–Leiblerdivergence between Gaussian mixture models for satellite image retrieval, Remote Sens. Lett., 7 (2016), 651–660.
    [3] Y. B.Wang, L. Q. Zhang, X.H. Tong, et al., A three-layered graph-basedlearning approach for remote sensing image retrieval, IEEE Trans. Geosci. Remote Sens., 54 (2016), 6020–6034.
    [4] J.Muñoz-Marí, F. Bovolo, L.Gómez-Chova, et al., Semisupervised one-class support vector machines forclassification of remote sensing data, IEEETrans. Geosci. Remote Sens., 48 (2010), 3188–3197.
    [5] L. Y.Xiang, Y. Li, W.Hao, et al., Reversible natural language watermarking using synonymsubstitution and arithmetic coding, Comput. Mater. Continua, 55 (2018), 541–559.
    [6] Y. Tu,Y. Lin, J. Wang, et al., Semisupervised learningwith generative adversarial networks on digital signal modulationclassification. Comput. Mater. Continua, 55 (2018), 243–254.
    [7] D. J.Zeng, Y. Dai, F.Li, et al., Adversarial learning for distant supervised relation extraction, Comput. Mater. Continua, 55 (2018), 121–136.
    [8] J. M.Zhang, X. K. Jin, J.Sun, et al., Spatial and semantic convolutionalfeatures for robust visual object tracking, MultimediaTools Appl., Forthcoming 2018. Available at https://doi.org/ 10.1007/s11042-018-6562-8.
    [9] S. R.Zhou, W. L. Liang, J.G. Li, et al., Improved VGG model for road traffic sign recognition, Comput. Mater. Continua, 57 (2018), 11–24.
    [10] S. Karimpouli and P. Tahmasebi, Image-basedvelocity estimation of rock using convolutionalneural networks, Neural Netw., 111 (2019), 89–97.
    [11] S. Karimpouli and P. Tahmesbi, Segmentationof digital rock images using deep convolutional autoencoder networks, Comput. Geosci-UK, 126 (2019), 142–150.
    [12] P. Tahmasebi and A. Hezarkhani, Applicationof a modular feedforward neural network for grade estimation, Nat. Resour. Res., 20 (2011), 25–32.
    [13] O. Russakovsky, J. Deng, H. Su, et al., ImageNet large scale visualrecognition challenge, Int. J. Comput.Vision, 115 (2015), 211–252.
    [14] G. S.Xia, J. W. Hu, F.Hu, et al., AID: a benchmark data set for performance evaluation of aerial sceneclassification, IEEE Trans. Geosci.Remote Sens., 55 (2017), 3965–3981.
    [15] J. M.Zhang, Y. Wu, X.K. Jin, et al., A fast object tracker based on integrated multiple features anddynamic learning rate, Math. Probl. Eng.,2018 (2018), Article ID 5986062, 14 pages.
    [16] Y.Yang and N. Shawn, Comparing sift descriptors and gabor texture features forclassification of remote sensed imagery, 15thIEEE International Conference on Image Processing,(2008), 1852–1855.
    [17] B.Luo, S. J. Jiang and L. P. Zhang, Indexing of remote sensing images withdifferent resolutions by multiple features, IEEEJ. Sel. Top. Appl. Earth Obs. Remote Sens., 6 (2013), 1899–1912.
    [18] A.Avramović and V. Risojević, Block-based semantic classification ofhigh-resolution multispectral aerial images, Signal Image Video Process., 10 (2016), 75–84.
    [19] X.Chen, T. Fang, H.Huo, et al., Measuring the effectiveness of various features for thematicinformation extraction from very high resolution remote sensing imagery, IEEE Trans. Geosci. Remote Sens., 53 (2015), 4837–4851.
    [20] J. A. dos Santos, O. A. B. Penatti and R.da Silva Torres, Evaluating the potential of texture and color descriptors forremote sensing image retrieval and classification, 5th International Conference on Computer Vision Theory and Applications,(2010), 203–208.
    [21] Y.Yang and N. Shawn, Geographic image retrieval using local invariant features, IEEE Trans. Geosci. Remote Sens., 5 (2013), 818–832.
    [22] Y.Yang and N. Shawn, Bag-of-visual-words and spatial extensions for land-useclassification, 18th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems,(2010), 270–279.
    [23] Y.Yang and N. Shawn, Spatial pyramid co-occurrence for image classification, IEEE International Conference on ComputerVision, (2011), 1465–1472.
    [24] W. Shao, W. Yang, G.S. Xia, et al., A hierarchical scheme of multiple feature fusion forhigh-resolution satellite scene categorization, IEEE InternationalConference on Computer Vision Systems, (2013), 324–333.
    [25] W. Shao, W. Yang and G. S. Xia, Extremevalue theory-based calibration for the fusion of multiple features inhigh-resolution satellite scene classification, Int. J. Remote Sens., 34 (2013), 8588–8602.
    [26] N. Romain, P. David and G. Philippe-Henri,Evaluation of second-order visual features for land-use classification, 12th International Workshop on Content-BasedMultimedia Indexing, (2014), 1–5.
    [27] L. J.Chen, W. Yang, K.Xu, et al., Evaluation of local features for scene classification using VHRsatellite images, 2011 Joint Urban RemoteSensing Event, (2011), 385–388.
    [28] F. Hu,G. S. Xia, J. W. Hu, et al., Transferring deepconvolutional neural networks for the scene classification of high-resolutionremote sensing imagery, Remote Sens., 7 (2015), 14680–14707.
    [29] M. Castelluccio, G. Poggi, C. Sansone, et al., Land useclassification in remotesensing imagesby convolutional neuralnetworks, preprint, arXiv:1508.00092.
    [30] O. A. B. Penatti, K. Nogueira and J. A. dos Santos, Do deep features generalize from everydayobjects to remote sensing and aerial scenes domains?, IEEE Conference on Computer Vision and Pattern Recognition Workshops, (2015), 44–51.
    [31] F. P.S. Luus, B. P. Salmon, F.Van den Bergh, et al.,Multiview deep learning for land-use classification, IEEE Geosci. Remote Sens. Lett., 12 (2015), 2448–2452.
    [32] F. Zhang, B. Du and L. P. Zhang, Sceneclassification via a gradient boosting random convolutional network framework, IEEE Trans. Geosci. Remote Sens., 54 (2016), 1793–1802.
    [33] K.Nogueira, O. A. B. Penatti and J. A. dos Santos, Towards better exploitingconvolutional neural networks for remote sensing scene classification, Pattern Recogn., 61 (2017), 539–556.
    [34] G. Cheng, P. C. Zhou and J. W. Han,Learning rotation-invariant convolutional neural networks for objectdetection in VHR optical remote sensing images, IEEE Trans. Geosci. Remote Sens., 54 (2016), 7405–7415.
    [35] X. W.Yao, J. W. Han, G. Cheng, et al., Semantic annotation of high-resolutionsatellite images via weakly supervised learning, IEEE Trans. Geosci. Remote Sens., 54 (2016), 3660–3671.
    [36] G.Cheng, C. Y. Yang, X.W. Yao, et al., When deep learning meets metric learning: remote sensing imagescene classification via learning discriminative CNNs, IEEE Trans. Geosci. Remote Sens., 56 (2018), 2811–2821.
    [37] S.Chaib, H. Liu,Y. F. Gu, et al., Deep feature fusion for VHR remote sensing sceneclassification, IEEE Trans. Geosci.Remote Sens., 55 (2017), 4775–4784.
    [38] Q.Wang, S. T. Liu,J. Chanussot, et al., Scene classification with recurrent attention of VHR remotesensing images, IEEE Trans. Geosci.Remote Sens., 99 (2018), 1–13.
    [39] Y. L.Yu and F. X. Liu, Dense connectivity based two-stream deep feature fusion frameworkfor aerial scene classification, RemoteSens., 10 (2018), 1158.
    [40] Y. T. Chen, W. H. Xu, J. W. Zuo, et al., The fire recognition algorithmusing dynamic feature fusion and IV-SVM classifier, Cluster Comput., Forthcoming 2018. Available at https://doi.org/10.1007/s10586-018-2368-8.
    [41] G.Huang, Z. Liu, L.van derMaaten, et al., Densely connected convolutional networks, IEEE Conferenceon Computer Vision and Pattern Recognition,(2017), 4700–4708.
    [42] G.Huang, Y. Sun,Z. Liu, et al., Deep networks with stochastic depth, European Conferenceon Computer Vision,(2016), 646–661.
    [43] S.loffe and C. Szegedy, Batch normalization: acceleratingdeep network training by reducing internal covariate shift, 32nd International Conference on Machine Learning,(2015), 448–456.
    [44] P. Tahmasebi, F. Javadpour and M. Sahimi,Data mining and machine learning for identifying sweet spots in shalereservoirs, Expert Sys. Appl., 88 (2017), 435–447.
    [45] G.Cheng, J. W. Han and X. Q. Lu, Remote sensing image scene classification: benchmark and state of the art, Proc. IEEE, 105 (2017), 1865–1883.
    [46] L. H.Huang, C. Chen,W. Li, et al., Remote sensing image scene classification using multi-scalecompleted local binary patterns and fisher vectors, Remote Sens., 8 (2016),483.
    [47] X. Y.Bian, C. Chen,L. Tian, et al., Fusing local and global features for high-resolution sceneclassification, IEEE J. Sel. Top. Appl.Earth Obs. Remote Sens., 10 (2017), 2889–2901.
  • This article has been cited by:

    1. Jianming Zhang, Chaoquan Lu, Jin Wang, Xiao-Guang Yue, Se-Jung Lim, Zafer Al-Makhadmeh, Amr Tolba, Training Convolutional Neural Networks with Multi-Size Images and Triplet Loss for Remote Sensing Scene Classification, 2020, 20, 1424-8220, 1188, 10.3390/s20041188
    2. Jianming Zhang, Wei Wang, Chaoquan Lu, Jin Wang, Arun Kumar Sangaiah, Lightweight deep network for traffic sign classification, 2020, 75, 0003-4347, 369, 10.1007/s12243-019-00731-9
    3. Fan Wu, Xiong Li, Lili Xu, Saru Kumari, Dingbao Lin, Joel J. P. C. Rodrigues, An anonymous and identity-trackable data transmission scheme for smart grid under smart city notion, 2020, 75, 0003-4347, 307, 10.1007/s12243-020-00765-4
    4. M. Cheng, W. Chen, L. -M. Fang, M. -Z. Lou, H. Jin, B. Zhang, L. -L. Zang, Synthesis, Crystal Structure, and Anti-Gastric Cancer Activity of a Heterometallic Coordination Polymer Based on Flexible 6,6-Dithiodinicotinic Acid, 2020, 61, 0022-4766, 987, 10.1134/S0022476620060207
    5. Jianming Zhang, Chaoquan Lu, Jin Wang, Lei Wang, Xiao-Guang Yue, Concrete Cracks Detection Based on FCN with Dilated Convolution, 2019, 9, 2076-3417, 2686, 10.3390/app9132686
    6. Tao Li, Wencong Jiao, Li-Na Wang, Guoqiang Zhong, Automatic DenseNet Sparsification, 2020, 8, 2169-3536, 62561, 10.1109/ACCESS.2020.2984130
    7. Luogen Yao, Global exponential stability on anti-periodic solutions in proportional delayed HIHNNs, 2021, 33, 0952-813X, 47, 10.1080/0952813X.2020.1721571
    8. Non-Cooperative Spectrum Access Strategy Based on Impatient Behavior of Secondary Users in Cognitive Radio Networks, 2019, 8, 2079-9292, 995, 10.3390/electronics8090995
    9. Jiaohua Qin, Jianhua Chen, Xuyu Xiang, Yun Tan, Wentao Ma, Jing Wang, A privacy-preserving image retrieval method based on deep learning and adaptive weighted fusion, 2020, 17, 1861-8200, 161, 10.1007/s11554-019-00909-3
    10. Hong Zhang, Chaofan Qian, Convergence analysis on inertial proportional delayed neural networks, 2020, 2020, 1687-1847, 10.1186/s13662-020-02737-3
    11. Jianming Zhang, Yang Liu, Hehua Liu, Jin Wang, Learning Local–Global Multiple Correlation Filters for Robust Visual Tracking with Kalman Filter Redetection, 2021, 21, 1424-8220, 1129, 10.3390/s21041129
    12. Yirui Wu, Pengfei Han, Zhan Zheng, Instant water body variation detection via analysis on remote sensing imagery, 2021, 1861-8200, 10.1007/s11554-020-01062-y
    13. Khalid A. AlAfandy, Hicham Hicham, Mohamed Lazaar, Mohammed Al Achhab, Investment of Classic Deep CNNs and SVM for Classifying Remote Sensing Images, 2020, 5, 24156698, 652, 10.25046/aj050580
    14. Jianming Zhang, Zhipeng Xie, Juan Sun, Xin Zou, Jin Wang, A Cascaded R-CNN With Multiscale Attention and Imbalanced Samples for Traffic Sign Detection, 2020, 8, 2169-3536, 29742, 10.1109/ACCESS.2020.2972338
    15. Adaptive Leader-Follower Formation Control of Under-actuated Surface Vessels with Model Uncertainties and Input Constraints, 2019, 9, 2076-3417, 3901, 10.3390/app9183901
    16. Hao Gao, Lin Cao, Dingfeng Yu, Xuejun Xiong, Maoyong Cao, Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism, 2020, 8, 2169-3536, 142483, 10.1109/ACCESS.2020.3013898
    17. Yu Quan, Zhixin Li, Canlong Zhang, Huifang Ma, Object Detection Model Based on Deep Dilated Convolutional Networks by Fusing Transfer Learning, 2019, 7, 2169-3536, 178699, 10.1109/ACCESS.2019.2958817
    18. Yuxiang Yang, Weiwei Xing, Shunli Zhang, Limin Gao, Qi Yu, Xiaoping Che, Wei Lu, Visual Tracking With Long-Short Term Based Correlation Filter, 2020, 8, 2169-3536, 20257, 10.1109/ACCESS.2020.2968125
    19. Jian Zhang, Chuangxia Huang, Dynamics analysis on a class of delayed neural networks involving inertial terms, 2020, 2020, 1687-1847, 10.1186/s13662-020-02566-4
    20. Jian Zhang, Caihong Liu, A study of a clothing image segmentation method in complex conditions using a features fusion model, 2020, 61, 0005-1144, 150, 10.1080/00051144.2019.1691835
    21. Sowmya Natesan, Costas Armenakis, Udayalakshmi Vepakomma, Individual tree species identification using Dense Convolutional Network (DenseNet) on multitemporal RGB images from UAV, 2020, 8, 2291-3467, 310, 10.1139/juvs-2020-0014
    22. Yanli Xu, Qian Cao, Global asymptotic stability for a nonlinear density-dependent mortality Nicholson’s blowflies system involving multiple pairs of time-varying delays, 2020, 2020, 1687-1847, 10.1186/s13662-020-02569-1
    23. Yifei Tian, Wei Song, Long Chen, Yunsick Sung, Jeonghoon Kwak, Su Sun, A Fast Spatial Clustering Method for Sparse LiDAR Point Clouds Using GPU Programming, 2020, 20, 1424-8220, 2309, 10.3390/s20082309
    24. Attoumane Loukmane, Manuel Grana, Mohammed Mestari, 2020, A Model for Classification of Traffic Signs Using Improved Convolutional Neural Network and Image Enhancement, 978-1-7281-8084-7, 1, 10.1109/ICDS50568.2020.9268761
    25. Yuanjing Luo, Jiaohua Qin, Xuyu Xiang, Yun Tan, Qiang Liu, Lingyun Xiang, Coverless real-time image information hiding based on image block matching and dense convolutional network, 2020, 17, 1861-8200, 125, 10.1007/s11554-019-00917-3
    26. Xinliang Liu, Tao Ren, Hongfeng Chen, Yufeng Chen, Classification of tectonic and non-tectonic seismicity based on convolutional neural network, 2020, 224, 0956-540X, 191, 10.1093/gji/ggaa444
    27. Yuantao Chen, Jin Wang, Songjie Liu, Xi Chen, Jie Xiong, Jingbo Xie, Kai Yang, Multiscale fast correlation filtering tracking algorithm based on a feature fusion model, 2019, 1532-0626, 10.1002/cpe.5533
    28. Osama A. Shawky, Ahmed Hagag, El-Sayed A. El-Dahshan, Manal A. Ismail, A very high-resolution scene classification model using transfer deep CNNs based on saliency features, 2020, 1863-1703, 10.1007/s11760-020-01801-5
    29. Hirofumi Noguchi, Shigeki Sugano, Ephemeral-Cyber-Physical System: A Cloud-Like CPS Using Shared Devices in Open IoT, 2020, 14, 1932-8184, 5176, 10.1109/JSYST.2019.2950719
    30. Xiaowei Chen, Yiqi Tao, Lei Tong, Ran Su, 2020, GIS guided remote sensing image data for rural industry and space interaction mechanism modeling, 978-1-7281-2813-9, 837, 10.1109/ICISC47916.2020.9171164
    31. Qian Cao, Guoqiu Wang, Chaofan Qian, New results on global exponential stability for a periodic Nicholson’s blowflies model involving time-varying delays, 2020, 2020, 1687-1847, 10.1186/s13662-020-2495-4
    32. Jianming Zhang, You Wu, Wenjun Feng, Jin Wang, Spatially Attentive Visual Tracking Using Multi-Model Adaptive Response Fusion, 2019, 7, 2169-3536, 83873, 10.1109/ACCESS.2019.2924944
    33. Mohammad Javad Salehpour, Hamid Radmanesh, Seyyed Mohammad Hosseini Rostami, Jin Wang, Hye-Jin Kim, Effect of Load Priority Modeling on the Size of Fuel Cell as an Emergency Power Unit in a More-Electric Aircraft, 2019, 9, 2076-3417, 3241, 10.3390/app9163241
    34. Khalid A. AlAfandy, Hicham Omara, Mohamed Lazaar, Mohammed Al Achhab, Using Classic Networks for Classifying Remote Sensing Images: Comparative Study, 2020, 5, 24156698, 770, 10.25046/aj050594
    35. Seksan Mathulaprangsan, Kitsana Lanthong, Duangpen Jetpipattanapong, Siwadol Sateanpattanakul, Sujin Patarapuwadol, 2020, Rice Diseases Recognition Using Effective Deep Learning Models, 978-1-7281-6398-7, 386, 10.1109/ECTIDAMTNCON48261.2020.9090709
    36. Chaofan Qian, Yuhui Hu, Novel stability criteria on nonlinear density-dependent mortality Nicholson’s blowflies systems in asymptotically almost periodic environments, 2020, 2020, 1029-242X, 10.1186/s13660-019-2275-4
    37. Wei Song, Lingfeng Zhang, Yifei Tian, Simon Fong, Jinming Liu, Amanda Gozho, CNN-based 3D object classification using Hough space of LiDAR point clouds, 2020, 10, 2192-1962, 10.1186/s13673-020-00228-8
    38. Hong Zhang, Yawei Li, Yujie Wu, Zeyu Zhang, BID: An Effective Blind Image Deblurring Scheme to Estimate the Blur Kernel for Various Scenarios, 2020, 8, 2169-3536, 9185, 10.1109/ACCESS.2020.2964621
    39. Dan Tang, Liu Tang, Rui Dai, Jingwen Chen, Xiong Li, Joel J.P.C. Rodrigues, MF-Adaboost: LDoS attack detection based on multi-features and improved Adaboost, 2020, 106, 0167739X, 347, 10.1016/j.future.2019.12.034
    40. Anrui Fu, Bo Wang, 2020, Portfolio Optimization based on LSTM Neural Network Prediction, 978-1-7281-6853-1, 1, 10.1109/ICNSC48988.2020.9238089
    41. Dengyong Zhang, Shanshan Wang, Jin Wang, Arun Kumar Sangaiah, Feng Li, Victor S. Sheng, Detection of Tampering by Image Resizing Using Local Tchebichef Moments, 2019, 9, 2076-3417, 3007, 10.3390/app9153007
    42. Qian Zhang, Yeqi Liu, Chuanyang Gong, Yingyi Chen, Huihui Yu, Applications of Deep Learning for Dense Scenes Analysis in Agriculture: A Review, 2020, 20, 1424-8220, 1520, 10.3390/s20051520
    43. Weiwei Cai, Botao Liu, Zhanguo Wei, Meilin Li, Jiangming Kan, TARDB-Net: triple-attention guided residual dense and BiLSTM networks for hyperspectral image classification, 2021, 80, 1380-7501, 11291, 10.1007/s11042-020-10188-x
    44. Feng Yang, Wentong Li, Haiwei Hu, Wanyi Li, Peng Wang, Multi-Scale Feature Integrated Attention-Based Rotation Network for Object Detection in VHR Aerial Images, 2020, 20, 1424-8220, 1686, 10.3390/s20061686
    45. Shanpeng Liu, Xiong Li, Fan Wu, Junguo Liao, Jin Wang, Dingbao Lin, A Novel Authentication Protocol with Strong Security for Roaming Service in Global Mobile Networks, 2019, 8, 2079-9292, 939, 10.3390/electronics8090939
    46. Lingyan Zhang, Mifang Li, Yujia Zhou, Guangming Lu, Quan Zhou, Deep Learning Approach for Anterior Cruciate Ligament Lesion Detection: Evaluation of Diagnostic Performance Using Arthroscopy as the Reference Standard, 2020, 52, 1053-1807, 1745, 10.1002/jmri.27266
    47. Abtin Riasatian, Morteza Babaie, Danial Maleki, Shivam Kalra, Mojtaba Valipour, Sobhan Hemati, Manit Zaveri, Amir Safarpoor, Sobhan Shafiei, Mehdi Afshari, Maral Rasoolijaberi, Milad Sikaroudi, Mohd Adnan, Sultaan Shah, Charles Choi, Savvas Damaskinos, Clinton JV Campbell, Phedias Diamandis, Liron Pantanowitz, Hany Kashani, Ali Ghodsi, H.R. Tizhoosh, Fine-Tuning and training of densenet for histopathology image representation using TCGA diagnostic slides, 2021, 70, 13618415, 102032, 10.1016/j.media.2021.102032
    48. Yuantao Chen, Jiajun Tao, Jin Wang, Xi Chen, Jingbo Xie, Jie Xiong, Kai Yang, The Novel Sensor Network Structure for Classification Processing Based on the Machine Learning Method of the ACGAN, 2019, 19, 1424-8220, 3145, 10.3390/s19143145
    49. Biserka Petrovska, Tatjana Atanasova-Pacemska, Roberto Corizzo, Paolo Mignone, Petre Lameski, Eftim Zdravevski, Aerial Scene Classification through Fine-Tuning with Adaptive Learning Rates and Label Smoothing, 2020, 10, 2076-3417, 5792, 10.3390/app10175792
    50. Qian Cao, Guoqiu Wang, Dynamic analysis on a delayed nonlinear density-dependent mortality Nicholson's blowflies model, 2020, 0020-7179, 1, 10.1080/00207179.2020.1725134
    51. Chunyong Yin, Biao Zhou, Zhichao Yin, Jin Wang, Local privacy protection classification based on human-centric computing, 2019, 9, 2192-1962, 10.1186/s13673-019-0195-4
    52. Hong Zhang, Qian Cao, Hedi Yang, Asymptotically almost periodic dynamics on delayed Nicholson-type system involving patch structure, 2020, 2020, 1029-242X, 10.1186/s13660-020-02366-0
    53. JinWoo Yoo, JaeWook Shin, PooGyeon Park, A bias-compensated proportionate NLMS algorithm with noisy input signals, 2019, 32, 10745351, e4167, 10.1002/dac.4167
    54. Min Beom Lee, Yu Hwan Kim, Kang Ryoung Park, Conditional Generative Adversarial Network- Based Data Augmentation for Enhancement of Iris Recognition Accuracy, 2019, 7, 2169-3536, 122134, 10.1109/ACCESS.2019.2937809
    55. Md. Tarek Habib, Md. Ariful Islam Arif, Sumaita Binte Shorif, Mohammad Shorif Uddin, Farruk Ahmed, 2021, Chapter 10, 978-981-33-6423-3, 143, 10.1007/978-981-33-6424-0_10
    56. Chaofan Qian, New periodic stability for a class of Nicholson's blowflies models with multiple different delays, 2020, 0020-7179, 1, 10.1080/00207179.2020.1766118
    57. Jiajia Zhao, Wangyang Cai, Hairong He, Wei Wang, Ming Tang, Jin Wang, Honeycomb pure-silica-core fiber array with air-hole cladding for image transmission, 2019, 2, 2578-7519, 2470, 10.1364/OSAC.2.002470
    58. B. H. Shekar, Habtu Hailu, 2021, Chapter 15, 978-981-16-1085-1, 161, 10.1007/978-981-16-1086-8_15
    59. Sk Maidul Islam, Subhankar Joardar, Arif Ahmed Sekh, 2021, Chapter 9, 978-981-16-1091-2, 100, 10.1007/978-981-16-1092-9_9
    60. Sikha Das, Samarjit Kar, Arif Ahmed Sekh, 2021, Chapter 38, 978-981-16-1091-2, 455, 10.1007/978-981-16-1092-9_38
    61. Emilio Rapuano, Gabriele Meoni, Tommaso Pacini, Gianmarco Dinelli, Gianluca Furano, Gianluca Giuffrida, Luca Fanucci, An FPGA-Based Hardware Accelerator for CNNs Inference on Board Satellites: Benchmarking with Myriad 2-Based Solution for the CloudScout Case Study, 2021, 13, 2072-4292, 1518, 10.3390/rs13081518
    62. Weiwei Cai, Zhanguo Wei, Runmin Liu, Yuan Zhuang, Yan Wang, Xin Ning, Remote Sensing Image Recognition Based on Multi-attention Residual Fusion Networks, 2021, 1, 2788-6743, 1, 10.52810/TPRIS.2021.100005
    63. Sk Maidul Islam, Subhankar Joardar, Arif Ahmed Sekh, 2022, Chapter 17, 978-3-031-22484-3, 180, 10.1007/978-3-031-22485-0_17
    64. Renfei Yang, Fang Luo, Fu Ren, Wenli Huang, Qianyi Li, Kaixuan Du, Dingdi Yuan, Identifying Urban Wetlands through Remote Sensing Scene Classification Using Deep Learning: A Case Study of Shenzhen, China, 2022, 11, 2220-9964, 131, 10.3390/ijgi11020131
    65. Rishabh Kapoor, William Sleeman, Jatinder Palta, Elisabeth Weiss, 3D deep convolution neural network for radiation pneumonitis prediction following stereotactic body radiotherapy, 2022, 1526-9914, 10.1002/acm2.13875
    66. Ayşe CENGİZ, Derya AVCI, Uydu İmgelerine Derin Öğrenme Tabanlı Süper Çözünürlük Yöntemlerinin Uygulanması, 2021, 21, 2149-3367, 1069, 10.35414/akufemubid.829644
    67. Yang Zhang, Ruohan Zong, Lanyu Shang, Dong Wang, On Coupling Classification and Super-Resolution in Remote Urban Sensing: An Integrated Deep Learning Approach, 2022, 60, 0196-2892, 1, 10.1109/TGRS.2022.3169703
    68. Mayank Jha, Om Amrit, Harsh Jain, Mayank Sharma, Achal Kaushik, 2022, Chapter 53, 978-981-16-2122-2, 681, 10.1007/978-981-16-2123-9_53
    69. Jianming Zhang, Wenjun Feng, Tingyu Yuan, Jin Wang, Arun Kumar Sangaiah, SCSTCF: Spatial-Channel Selection and Temporal Regularized Correlation Filters for visual tracking, 2022, 118, 15684946, 108485, 10.1016/j.asoc.2022.108485
    70. Zhangdong Wang, Jiaohua Qin, Xuyu Xiang, Yun Tan, Neal N. Xiong, Criss-Cross Attentional Siamese Networks for Object Tracking, 2022, 73, 1546-2226, 2931, 10.32604/cmc.2022.028896
    71. Sk Maidul Islam, Subhankar Joardar, Debi Prosad Dogra, Arif Ahmed Sekh, Ornament Image Retrieval Using Multimodal Fusion, 2021, 2, 2662-995X, 10.1007/s42979-021-00734-1
    72. Mengxi Liu, Haojun Deng, Wen Dong, Identification of Mangrove Invasive Plant Derris Trifoliate Using UAV Images and Deep Learning Algorithms, 2022, 15, 1939-1404, 10017, 10.1109/JSTARS.2022.3223227
    73. Berk Arican , Erdem Safa Akkul, Levent Çarkacioglu, Behçet U. Töreyin, Nazzareno Pierdicca, Lorenzo Bruzzone, Francesca Bovolo, 2022, Compressed domain classification of remote sensing scene images based on sub-band data fusion, 9781510655379, 8, 10.1117/12.2635862
    74. Ratnabali Pal, Samarjit Kar, Arif Ahmed Sekh, 2023, Chapter 53, 978-981-19-6630-9, 771, 10.1007/978-981-19-6631-6_53
    75. Zhongbin Su, Wei Li, Zheng Ma, Rui Gao, An improved U-Net method for the semantic segmentation of remote sensing images, 2022, 52, 0924-669X, 3276, 10.1007/s10489-021-02542-9
    76. Genji Yuan, Jianbo Li, Yue Wang, Xianglong Meng, 2022, Chapter 37, 978-3-031-19213-5, 441, 10.1007/978-3-031-19214-2_37
    77. Sk Maidul Islam, Subhankar Joardar, Arif Ahmed Sekh, DSSN: dual shallow Siamese network for fashion image retrieval, 2022, 1380-7501, 10.1007/s11042-022-14204-0
    78. Pablo Gomez, Gabriele Meoni, MSMatch: Semisupervised Multispectral Scene Classification With Few Labels, 2021, 14, 1939-1404, 11643, 10.1109/JSTARS.2021.3126082
    79. Biswajit Jena, Gopal Krishna Nayak, Sanjay Saxena, Convolutional neural network and its pretrained models for image classification and object detection: A survey, 2022, 34, 1532-0626, 10.1002/cpe.6767
    80. Bharani Basapathy Rudra, Gururaj Murtugudde, Remote sensing scene classification using visual geometry group 19 model and multi objective grasshopper optimization algorithm, 2022, 13, 0975-6809, 3017, 10.1007/s13198-022-01790-3
    81. Khalid A. AlAfandy, Hicham Omara, Hala S. El-Sayed, Mohammed Baz, Mohamed Lazaar, Osama S. Faragallah, Mohammed Al Achhab, Efficient Classification of Remote Sensing Images Using Two Convolution Channels and SVM, 2022, 72, 1546-2226, 739, 10.32604/cmc.2022.022457
    82. Erdem Safa Akkul, Berk Arıcan, Behçet Uğur Töreyin, 2021, Chapter 54, 978-3-030-88112-2, 669, 10.1007/978-3-030-88113-9_54
    83. Xiaosong Xie, Guangsong Yang, Mengxi Jiang, Qiubo Ye, Chen-Fu Yang, A Kind of Wireless Modulation Recognition Method Based on DenseNet and BLSTM, 2021, 9, 2169-3536, 125706, 10.1109/ACCESS.2021.3111406
    84. You Xuan Thung, Tom Benson, Nikita Klimenko, 2022, Detecting languages in streetscapes using deep convolutional neural networks, 978-1-6654-8045-1, 1711, 10.1109/BigData55660.2022.10020745
    85. Jiaogen Zhou, Yang Wang, Caiyun Zhang, Wenbo Wu, Yanzhu Ji, Yeai Zou, Eyebirds: Enabling the Public to Recognize Water Birds at Hand, 2022, 12, 2076-2615, 3000, 10.3390/ani12213000
    86. Tong Liu, Tao Chen, 2021, A Comparation of CNN and DenseNet for Landslide Detection, 978-1-6654-0369-6, 8440, 10.1109/IGARSS47720.2021.9555000
    87. Jianming Zhang, Kai Wang, Yaoqi He, Lidan Kuang, Visual Object Tracking via Cascaded RPN Fusion and Coordinate Attention, 2022, 132, 1526-1506, 909, 10.32604/cmes.2022.020471
    88. Moisés Cordeiro-Costas, Daniel Villanueva, Pablo Eguía-Oller, Enrique Granada-Álvarez, Machine Learning and Deep Learning Models Applied to Photovoltaic Production Forecasting, 2022, 12, 2076-3417, 8769, 10.3390/app12178769
    89. Wassana Naku, Anand K. Nambisan, Muhammad Roman, Chen Zhu, Rex E. Gerald, Jie Huang, Identification of Volatile Organic Liquids by Combining an Array of Fiber-Optic Sensors and Machine Learning, 2023, 8, 2470-1343, 4597, 10.1021/acsomega.2c05451
    90. Ying Xu, Zefeng Lu, Licai Liu, Daseng Cai, A Classification Algorithm Based on Complex Number Feature, 2020, 8, 2169-3536, 17842, 10.1109/ACCESS.2020.2966017
    91. 2022, Customized EfficientNet for Histopathology Image Representation, 978-1-6654-8768-9, 01, 10.1109/SSCI51031.2022.10022191
    92. Sara S. Satake, Rodrigo Calvo, Alceu S. Britto, Yandre M. G. Costa, 2022, Chapter 10, 978-3-030-96877-9, 108, 10.1007/978-3-030-96878-6_10
    93. Yan Jiang, Guisheng Yin, Ye Yuan, Qingan Da, Mohamed Amine Ferrag, Project Gradient Descent Adversarial Attack against Multisource Remote Sensing Image Scene Classification, 2021, 2021, 1939-0122, 1, 10.1155/2021/6663028
    94. Naftaly Wambugu, Yiping Chen, Zhenlong Xiao, Kun Tan, Mingqiang Wei, Xiaoxue Liu, Jonathan Li, Hyperspectral image classification on insufficient-sample and feature learning using deep neural networks: A review, 2021, 105, 15698432, 102603, 10.1016/j.jag.2021.102603
    95. Shamik Tiwari, Anurag Jain, A lightweight capsule network architecture for detection of COVID ‐19 from lung CT scans , 2022, 32, 0899-9457, 419, 10.1002/ima.22706
    96. Heesang Eom, Jongryun Roh, Yuli Sun Hariyani, Suwhan Baek, Sukho Lee, Sayup Kim, Cheolsoo Park, Deep Learning-Based Optimal Smart Shoes Sensor Selection for Energy Expenditure and Heart Rate Estimation, 2021, 21, 1424-8220, 7058, 10.3390/s21217058
    97. Ivica Dimitrovski, Ivan Kitanovski, Dragi Kocev, Nikola Simidjievski, Current trends in deep learning for Earth Observation: An open-source benchmark arena for image classification, 2023, 197, 09242716, 18, 10.1016/j.isprsjprs.2023.01.014
    98. Yang Zhang, Ruohan Zong, Lanyu Shang, Md Tahmid Rashid, Dong Wang, 2021, SuperClass: A Deep Duo-Task Learning Approach to Improving QoS in Image-driven Smart Urban Sensing Applications, 978-1-6654-1494-4, 1, 10.1109/IWQOS52092.2021.9521332
    99. Wassana Naku, Chen Zhu, Anand K. Nambisan, Rex E. Gerald, Jie Huang, Machine learning identifies liquids employing a simple fiber-optic tip sensor, 2021, 29, 1094-4087, 40000, 10.1364/OE.441144
    100. Law Kumar Singh, Hitendra Garg, Munish Khanna, Performance evaluation of various deep learning based models for effective glaucoma evaluation using optical coherence tomography images, 2022, 81, 1380-7501, 27737, 10.1007/s11042-022-12826-y
    101. Junli Hu, Qiangyi Li, Study on the Lightweighting Strategy of Target Detection Model with Deep Learning, 2022, 2022, 1687-5699, 1, 10.1155/2022/7234888
    102. Waid Nimri, Yong Wang, Ziang Zhang, Chengbin Deng, Kristofor Sellstrom, Data-driven approaches and model-based methods for detecting and locating leaks in water distribution systems: a literature review, 2023, 0941-0643, 10.1007/s00521-023-08497-x
    103. Junfu Fan, Zongwen Shi, Zhoupeng Ren, Yuke Zhou, Min Ji, DDPM-SegFormer: Highly refined feature land use and land cover segmentation with a fused denoising diffusion probabilistic model and transformer, 2024, 133, 15698432, 104093, 10.1016/j.jag.2024.104093
    104. Aakash Thapa, Teerayut Horanont, Bipul Neupane, Jagannath Aryal, Deep Learning for Remote Sensing Image Scene Classification: A Review and Meta-Analysis, 2023, 15, 2072-4292, 4804, 10.3390/rs15194804
    105. Shuhao Liu, Xinwei Yao, Xiangjie Kong, 2024, Obstruction detection and recovery in weather radar images using deep learning, 9781510680258, 74, 10.1117/12.3029064
    106. Zhang Qingmiao, Zhang Dinghua, Novel hawk swarm‐optimized deep learning classification with K‐nearest neighbor based decision making for autonomous vehicle movement controller, 2024, 36, 1532-0626, 10.1002/cpe.8241
    107. Rajashree Gadhave, R. R. Sedamkar, Sujata Alegavi, 2023, 2984, 0094-243X, 020005, 10.1063/5.0175906
    108. Adekanmi Adeyinka Adegun, Serestina Viriri, Jules-Raymond Tapamo, Review of deep learning methods for remote sensing satellite images classification: experimental survey and comparative analysis, 2023, 10, 2196-1115, 10.1186/s40537-023-00772-x
    109. Hazret Tekin, Yılmaz Kaya, A new approach for heart disease detection using Motif transform-based CWT’s time-frequency images with DenseNet deep transfer learning methods, 2024, 69, 0013-5585, 407, 10.1515/bmt-2023-0580
    110. Bo Wang, Hong Zhao, Xiaoyi Wang, Guanting Lyu, Kuangmin Chen, Jinfeng Xu, Guishan Cui, Liheng Zhong, Le Yu, Huabing Huang, Qinghong Sheng, Bamboo classification based on GEDI, time-series Sentinel-2 images and whale-optimized, dual-channel DenseNet: A case study in Zhejiang province, China, 2024, 209, 09242716, 312, 10.1016/j.isprsjprs.2024.02.002
    111. Sk Maidul Islam, Subhankar Joardar, Arif Ahmed Sekh, BangleFIR: bridging the gap in fashion image retrieval with a novel dataset of bangles, 2024, 1573-7721, 10.1007/s11042-024-19698-4
    112. Rakesh Meena, Sunil Joshi, Sandeep Raghuwanshi, 2023, Chapter 30, 978-981-99-2099-0, 373, 10.1007/978-981-99-2100-3_30
    113. Salma N, Madhuri G R, Basavaraj Jagadale, Advancing remote sensing: a unified deep learning approach with pretrained and custom architectures for high-precision classification, 2024, 99, 0031-8949, 116012, 10.1088/1402-4896/ad8491
    114. Mohammed Diykh, Mumtaz Ali, Mehdi Jamei, Shahab Abdulla, Md Palash Uddin, Aitazaz Ahsan Farooque, Abdulhaleem H. Labban, Hussein Alabdally, Empirical curvelet transform based deep DenseNet model to predict NDVI using RGB drone imagery data, 2024, 221, 01681699, 108964, 10.1016/j.compag.2024.108964
    115. Yifan Liu, Qiudong Yu, Shuze Geng, Real-time and lightweight detection of grape diseases based on Fusion Transformer YOLO, 2024, 15, 1664-462X, 10.3389/fpls.2024.1269423
    116. Moisés Cordeiro-Costas, Daniel Villanueva, Pablo Eguía-Oller, Miguel Martínez-Comesaña, Sérgio Ramos, Load Forecasting with Machine Learning and Deep Learning Methods, 2023, 13, 2076-3417, 7933, 10.3390/app13137933
    117. Chen Zhao, Renjun Shuai, Li Ma, Wenjia Liu, Die Hu, Menglin Wu, Dermoscopy Image Classification Based on StyleGAN and DenseNet201, 2021, 9, 2169-3536, 8659, 10.1109/ACCESS.2021.3049600
    118. Matheus H. C. Leme, Vinicius S. Simm, Douglas Rorie Tanno, Yandre M. G. Costa, Marcos Aurélio Domingues, 2024, Chapter 52, 978-3-031-49017-0, 730, 10.1007/978-3-031-49018-7_52
    119. Chao Liu, Xueqing Yu, Dingyu Wang, Tingting Jiang, 2024, Chapter 6, 978-3-031-72085-7, 57, 10.1007/978-3-031-72086-4_6
    120. Lirong Yin, Lei Wang, Jianqiang Li, Siyu Lu, Jiawei Tian, Zhengtong Yin, Shan Liu, Wenfeng Zheng, YOLOV4_CSPBi: Enhanced Land Target Detection Model, 2023, 12, 2073-445X, 1813, 10.3390/land12091813
    121. Farhan Ullah, Irfan Ullah, Khalil Khan, Salabat Khan, Farhan Amin, Advances in deep neural network-based hyperspectral image classification and feature learning with limited samples: a survey, 2025, 55, 0924-669X, 10.1007/s10489-024-06139-w
    122. Mingqi Wang, Shengpeng Wan, Junsong Yu, EFPI Solution Type and Concentration Sensing System Based on ConvNeXt Network Model, 2024, 73, 0018-9456, 1, 10.1109/TIM.2024.3472888
    123. Tianzhen Yin, Yankun Peng, Kuanglin Chao, Yongyu Li, Emerging trends in SERS-based veterinary drug detection: multifunctional substrates and intelligent data approaches, 2025, 9, 2396-8370, 10.1038/s41538-025-00393-z
    124. Seyyed Ali Jahani, Keyvan Mohebbi, Farsad Zamani Boroujeni, Improving Scene Recognition in Remote Sensing Using Deep Learning and Feature Selector, 2024, 21, 0365-3196, 69, 10.61186/jsdp.21.3.69
    125. Ritu Aggarwal, Gulbir Singh, S. B. Goyal, Anurag Jain, Tanupriya Choudhury, 2025, Chapter 4, 978-3-031-86071-3, 40, 10.1007/978-3-031-86072-0_4
    126. Karim Malik, Colin Robertson, Structural Similarity-Guided Siamese U-Net Model for Detecting Changes in Snow Water Equivalent, 2025, 17, 2072-4292, 1631, 10.3390/rs17091631
  • Reader Comments
  • © 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(11634) PDF downloads(4039) Cited by(125)

Figures and Tables

Figures(11)  /  Tables(7)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog