Figure 1.
Illustration of the four orientation detection neurons.
In recent years, there has been an increasing emphasis on STEM education and the need for effective professional development (PD) programs to increase teachers' competencies in planning and implementing STEM activities. In this study, the aim was to investigate in depth the effect of a comprehensive STEM PD program on science teachers' competencies in STEM lesson plan preparation and classroom implements. The research design was a holistic single case. The study group consisted of 10 science teachers working in public secondary schools. The STEM PD program included an 80-hour training consisting of four stages in a 6-month period. During the PD program, the teachers prepared at least two STEM lesson plans and carried out STEM classroom implements. The data source of the study consisted of the lesson plans prepared by the teachers and the researchers' observations (observation form and field notes) of the teachers' classroom implements. The results show that the STEM PD program, developed within the scope of this research, contributed to the preparation of lesson plans and in-class implements by teachers. It is recommended that teachers engage in both theoretical learning and practical workshops, distributed across multiple sessions, to enable planning, execution, and experience sharing. The process of identifying deficiencies during implementation and receiving feedback from peers and experts is highlighted as a key strength of the program.
Citation: İrfan Gümüş, Esra Bozkurt Altan. Professional development in STEM for science teachers: Examining improvements in lesson planning and implements†[J]. STEM Education, 2025, 5(3): 425-447. doi: 10.3934/steme.2025021
| [1] | Han-Cheng Dan, Yongcheng Long, Hui Yao, Songlin Li, Yanhao Liu, Quanfeng Zhou . Investigation on the fractal characteristic of asphalt pavement texture roughness incorporating 3D reconstruction technology. Electronic Research Archive, 2023, 31(4): 2337-2357. doi: 10.3934/era.2023119 |
| [2] | Abul Bashar . Employing combined spatial and frequency domain image features for machine learning-based malware detection. Electronic Research Archive, 2024, 32(7): 4255-4290. doi: 10.3934/era.2024192 |
| [3] | Hao Yang, Peihan Wang, Fang Han, Qingyun Wang . An interpretable mechanism for grating-induced cross-inhibition and gamma oscillation based on a visual cortical neuronal network model. Electronic Research Archive, 2024, 32(4): 2936-2954. doi: 10.3934/era.2024134 |
| [4] | Eray Önler . Feature fusion based artificial neural network model for disease detection of bean leaves. Electronic Research Archive, 2023, 31(5): 2409-2427. doi: 10.3934/era.2023122 |
| [5] | Xue Yu . Orientable vertex imprimitive complete maps. Electronic Research Archive, 2024, 32(4): 2466-2477. doi: 10.3934/era.2024113 |
| [6] | Yogesh Kumar Rathore, Rekh Ram Janghel, Chetan Swarup, Saroj Kumar Pandey, Ankit Kumar, Kamred Udham Singh, Teekam Singh . Detection of rice plant disease from RGB and grayscale images using an LW17 deep learning model. Electronic Research Archive, 2023, 31(5): 2813-2833. doi: 10.3934/era.2023142 |
| [7] | Chao Chen, Hua Kong, Bin Wu . Edge detection of remote sensing image based on Grünwald-Letnikov fractional difference and Otsu threshold. Electronic Research Archive, 2023, 31(3): 1287-1302. doi: 10.3934/era.2023066 |
| [8] | Chunyan An, Wei Bai, Donglei Zhang . Meet-in-the-middle differential fault analysis on Midori. Electronic Research Archive, 2023, 31(11): 6820-6832. doi: 10.3934/era.2023344 |
| [9] | Chetan Swarup, Kamred Udham Singh, Ankit Kumar, Saroj Kumar Pandey, Neeraj varshney, Teekam Singh . Brain tumor detection using CNN, AlexNet & GoogLeNet ensembling learning approaches. Electronic Research Archive, 2023, 31(5): 2900-2924. doi: 10.3934/era.2023146 |
| [10] | Yuhang Liu, Jun Chen, Yuchen Wang, Wei Wang . Interpretable machine learning models for detecting fine-grained transport modes by multi-source data. Electronic Research Archive, 2023, 31(11): 6844-6865. doi: 10.3934/era.2023346 |
In recent years, there has been an increasing emphasis on STEM education and the need for effective professional development (PD) programs to increase teachers' competencies in planning and implementing STEM activities. In this study, the aim was to investigate in depth the effect of a comprehensive STEM PD program on science teachers' competencies in STEM lesson plan preparation and classroom implements. The research design was a holistic single case. The study group consisted of 10 science teachers working in public secondary schools. The STEM PD program included an 80-hour training consisting of four stages in a 6-month period. During the PD program, the teachers prepared at least two STEM lesson plans and carried out STEM classroom implements. The data source of the study consisted of the lesson plans prepared by the teachers and the researchers' observations (observation form and field notes) of the teachers' classroom implements. The results show that the STEM PD program, developed within the scope of this research, contributed to the preparation of lesson plans and in-class implements by teachers. It is recommended that teachers engage in both theoretical learning and practical workshops, distributed across multiple sessions, to enable planning, execution, and experience sharing. The process of identifying deficiencies during implementation and receiving feedback from peers and experts is highlighted as a key strength of the program.
The visual system is a component of the nervous system, which, as one of the most basic human sensory systems, gives humans the ability to visualize and perceive [1]. More than one-third of the human cerebral cortex is related to the visual system, and for humans, the vast majority of external information is obtained through vision. Thus, the visual system contributes to human cognition, decision-making, emotional behavior and other behaviors [2]. The complete visual system consists of the eye (especially the retina), the optic nerve, the optic cross, the visual tract, the lateral geniculate body, the visual cortex and the visual association cortex [3]. These structures are divided into the anterior and posterior visual pathways from the lateral geniculate body. The eye, as the first station in the anterior visual pathway, is responsible for the initial processing of visual information [4]. Light is refracted by the eye and then projected onto the retina. Photoreceptor cells in the retina convert light signals into electrical signals, transmit them to bipolar cells and ganglion cells and finally transmit action potential signals to the brain via the optic nerve [5]. Amacrine cells and horizontal cells are involved in lateral information transmission in the retina, thus forming various complex visual receptive fields, which are usually sensitive to certain specific features of visual information, such as color, size, distance and orientation [6]. Hence, visual information is usually initially processed in the retina into channels carrying various specific features [7]. Retinal cells that carry similar visual information features are classified under the same visual channel. To date, more than 30 visual channels have been verified by genetics and anatomy [8]. In the present study, we concentrate on the visual channels associated with orientation features.
Orientation-selective cells were first identified in the pigeon retina by Maturana and Frenk in 1963 [9], and Levick demonstrated similar orientation selectivity in the rabbit retina in 1967 [10]. Subsequent research was reported on the orientation selectivity of retinal cells in cats [11,12,13,14], turtles [15], mice [16,17,18], goldfish [19,20] and zebrafish [21,22]. The congruence between the functional roles of amacrine cells and retinal ganglion cells in orientation selection circuits was established by Paride Antinucci in 2013 [23]. The study of the cell-adhesion molecule Tenm3 further demonstrated that AC is a critical component of orientation selection in retinal ganglion cells [24]. In a study of rabbit retinal cells, researchers found that orientation-selective amacrine cells (OSACs) have radially symmetrical dendrites, and the receptive field of OSACs can be approximated as a circle [25,26]. A common feature evident in these vertebrate visual models is that an orientation-selective amacrine cell becomes an essential element in the visual circuitry for tuning neurons, and it is always present in the initial reception and transmission of visual information. In these models, the orientation tuning function exhibited by the amacrine cells tends to be determined more by their own morphological features than by the inhibition of the superficial cells. That is, the dendritic orientation of the amacrine cell itself determines the implementation of orientation-selective functions in initial visual information processing to a greater extent. OSACs are sensitive to orientation information stimuli that are consistent with the direction of dendrite growth and insensitive to orientation information stimuli that are inconsistent with the direction of dendrite growth; hence, OSACs have two types of expressions for orientation information stimuli: ON response and OFF response [27]. In essence, OSACs are activated only by stimuli from a specific orientation and do not respond to stimuli from other orientations.
There is great potential value in employing the biological properties and mechanisms of OSACs in the field of engineering. Accordingly, in this study, a perceptron-based orientation detection model (PODM) is proposed, and the effectiveness of the model for the orientation detection of objects in images is experimentally verified. As a single neuron is sensitive only to stimuli in a specific orientation, four neurons are inserted into the mechanism to detect information in four orientations. These neurons receive information in the receptive field and are activated by the corresponding information, in the same manner as OSACs receive information from photoreceptor cells and are sensitive to information in a specific orientation. This indicates that the PODM is highly compatible with the biological properties of the cell. The global orientation selection is dependent on the aggregation of neuronal activation in all receptive fields. To ensure the validity of the mechanism's orientation detection, the object features (e.g., color, location and shape) in the experimental dataset are randomly generated. The experimental results confirm that the PODM is always efficient in detecting the orientation of objects, regardless of changes in features such as the color, position and shape of the objects. Based on the experimental results, it is likely that more angles of orientation recognition can be achieved by adding more neurons to the model or by interacting information between a limited number of neurons. This may inspire new ideas to unravel the mysteries of the functioning of the visual system.
The perceptron was invented by Frank Rosenblatt in 1957 [28]. As a type of artificial neural network (ANN), it was designed from its inception to mimic the working mechanism of nerve cells. The state of a nerve cell depends on the strength of the information received from other nerve cells. When the strength of the information exceeds a certain threshold, the nerve cell is activated and generates action potentials, which are then transmitted to other neurons via synapses [29]. Corresponding underlying concepts in the perceptron include weight ω corresponding to synapses, bias b corresponding to thresholds and activation functions corresponding to cell bodies. The equation of the perceptron is shown below:
| f(x)={1ifω⋅x+b>00else, | (2.1) |
where x is the input received by the current neuron. OSACs usually receive input from multiple photoreceptors to determine the orientation of the object comprehensively, rather than relying on a single photoreceptor [30]. In this work, the neuron is set to receive the grayscale values of two adjacent points to determine whether the neuron corresponding to the orientation where these two points are located is activated or not. When the grayscale values of the two adjacent points are equal, the object is considered to be in the orientation corresponding to the two points, and the corresponding neuron is activated. Considering that the human eye has a limited grayscale recognition rate, and the colors of real objects are not exactly the same, a threshold needs to be added to the mechanism to determine whether the grayscale values are the same (approximately) [31]. Since human individuals differ, the ability of the human eye to recognize the minimum grayscale difference varies. For our model, the threshold value, as a user-set parameter, is both a switch for the model to work properly and a fault tolerance for the model detection. The smaller that the threshold is set, the better the model will recognize the smallest grayscale difference, and the more it will be affected by the background color change; the larger that the threshold is set, the weaker the model will be at recognizing the smallest grayscale, and it is likely to lose the detection ability. So, we need to give the threshold a suitable value to give the model a certain error tolerance. During our experiment, we found that the model can maintain a more ideal working state when the threshold is set to 3, and thus we define the threshold as 3. When the difference between two points is less than or equal to the threshold value, the two points are considered to have the same (approximate) grayscale value, and the neurons of the corresponding orientation are activated; otherwise, the two points are regarded as having different grayscale values, and the neurons of the corresponding orientation are not activated. In the receptive field of neuronal cells, the central point is selected as the reference point, the grayscale values of eight adjacent points are compared with the reference point, and the neuronal cells of the corresponding orientation are activated if the difference in grayscale values is less than the threshold. Based on the aforementioned basic principles, we propose an equation to adapt our mechanism as shown below:
| Response={ONif|x∗−xi|<thresholdOFFelse, | (2.2) |
where x∗ represents the grayscale value of the central reference point of the perceptual field, and the xi represents the grayscale value of the points adjacent to the reference point.
In this study, four neurons are set up in the mechanism to detect the four orientations of 0°, 45°, 90° and 135°, as shown in Figure 1. We define the photoreceptor cell in the center of the receptive field as the reference point with coordinates (i, j), and Xi,j represents the signal received from the photoreceptor cell by the amacrine cell. In such a receptive field, the horizontally oriented neuron (0 degrees) is activated when the signal received from the photoreceptor cell located at (i, j+1) or (i, j-1) is close to the signal at the reference point. A vertically oriented neuron (90 degrees) is activated when Xi+1,j or Xi−1,j is close to Xi,j. The neuron corresponding to 135 degrees is activated when the Xi+1,j+1 or Xi−1,j−1 is similar to Xi,j. The neuron corresponding to 45 degrees is activated when Xi−1,j+1 or Xi+1,j−1 is approaching Xi,j.
Neuronal cells based on this mechanism would be able to be activated by an object as small as 1*2 pixels. A simple demonstration is given in Figure 2. In the retina, photoreceptor cells receive light signals that are converted into electrical signals and then transmit information to cells in the posterior layer in turn. Two of the photoreceptor cells receive light signals, which are reflected in the current receptive field as the central point with the same signal as its horizontal neighbor, and activate the neuron in the amacrine cell layer in the corresponding horizontal orientation (0°).
When processing a complete image, it is necessary for the neurons to globally detect the image to detect the orientation of objects in the image. Therefore, the sliding window scanning mechanism from the convolutional neural network (CNN) [32] is utilized in this study. That is, the perceptual field slides sequentially from the beginning of the image to the next position in a fixed step, scanning over the whole image line by line to read the information of the whole image as an input to the model. The neurons corresponding to the four orientations are activated during the scanning process, the frequencies of activation are recorded and summarized at the same time, and the orientation detection results of the model are output after substitution into the activation function calculation. The equation of the activation function is shown below:
| f(x)=eXi∑ni=1Xi | (2.3) |
where Xi represents the activation frequencies, and n equals 4, meaning that a total of four orientation-selective neurons are employed here. It is worth noting that the output here is the probability of selecting each orientation, and the sum of the probabilities of the four orientations being selected is 1. The final detection result of the model is taken from the one with the highest probability among the four orientations, i.e., the orientation with the highest activation frequency is considered to be the final detection result of the model. A simplified diagram of the model is shown in Figure 3. An example of the specific sliding window scanning mechanism is shown in Figure 4. Figure 4(a) shows an object which is sized at 1*3 and in the orientation of 135°, placed on an image with a gradient grayscale background. Figure 4(b) shows the whole process of receiving this image by the model. A receptive field of size 3*3 slides across the entire image from left to right line by line. In each receptive field, the neurons in the corresponding orientation are activated and recorded separately (activation is shown in highlighted blue). Figure 4(c) summarizes the result of the scan, and the direction with the highest number of neuron activations is considered to be the orientation of the object. It can be seen that the global detection result obtained by the four neurons matches the actual orientation of the object. Unlike the sliding window scanning in the CNN, the proposed neuron needs minimal information interaction to read the features of the whole picture completely, and thus using a 3*3 receptive field to scan the whole picture will cause unnecessary waste of computational resources. Therefore, the size of the receptive field is collapsed from 3*3 to 2*3, and the activation level of neurons in the corresponding orientation is weighted accordingly, as shown in Figure 5. The experimental results verify that this improvement saves approximately one-third of the computational resources while ensuring detection accuracy.
A series of experiments were conducted to evaluate the effectiveness of the suggested model. In the first subsection, the composition of the experimental dataset and the method it generates are described. In the second subsection, the mechanism proposed above is utilized to detect the orientation of each object in the dataset to validate the feasibility of the mechanism. In the third subsection, the PODM is compared with CNN to verify the robustness of the mechanism by detecting simulated realistic images and the detection accuracy when subjected to the same level of interference. The fourth subsection then compares the performances of the mechanism when different sizes of perceptual fields are applied.
All datasets were randomly generated and according to the following guidelines: Each dataset contains 2500 images, each image has a resolution of 100*100 pixels, the background color of the image is a randomly generated grayscale color, an object is placed on each image, and the orientation and color of the object are randomly generated. Figure 6 displays a partial sample of the dataset. The experiments are classified only according to the size of the object pixel values to check the accuracy and reliability of the model when dealing with objects of different sizes. Further experiments are applied to examine the ability of the proposed mechanism to cope with different backgrounds by replacing the filled form of the image background.
It is well known that a complete image is a combination of grayscale maps of multiple color channels. In this set of experiments, the validity of the proposed mechanism is verified in the monochromatic grayscale images. In this dataset, the background of the image and the color of the object in the image are each randomly generated monochromatic grayscale colors. The sizes of the objects in the images are divided into four categories: 50,100,500 and 1000 pixel values. The percentage of the object occupying the image ranges from 0.5 to 10%, which allows the tolerance of the proposed mechanism to object size to be fully investigated. Some samples from the monochrome grayscale image dataset are shown in Figure 6. It can be seen that the background of the image is a randomly generated monochrome grayscale, and the position of the object in the image, as well as its color, is random, while the aspect ratio (shape) of the object also has different variations. Figure 7 shows the detection result of the mechanism for one of the images in which the object has a size of approximately 500 pixels and an orientation of 135°. The activation intensity of neurons in the 135° orientation reaches 924, which is significantly higher than the activation intensity in the other three directions. The mechanism detection results point to 135°, which is consistent with the actual orientation of the object. It indicates that the proposed mechanism successfully detects the orientation of the object in that image. Table 1 presents the results obtained by the mechanism in this dataset. The mechanism had an accurate detection of the orientation of each object with a success rate of 100%, which demonstrates that the mechanism is always able to efficiently detect the orientation of the object in the images, regardless of the object's size, orientation, shape or color. Thus, the results indicate the mechanism has a stable recognition rate and good robustness in detecting the orientation of objects in monochrome grayscale images. Also, since monochrome grayscale images of multiple channels can be superimposed to form color images, the mechanism is likely also capable of detecting the orientation of objects in color images.
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 646 | 608 | 627 | 619 | 2500 |
| Predicted number | 646 | 608 | 627 | 619 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 641 | 589 | 644 | 626 | 2500 |
| Predicted number | 641 | 589 | 644 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 634 | 646 | 585 | 635 | 2500 |
| Predicted number | 634 | 646 | 585 | 635 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 600 | 623 | 643 | 634 | 2500 |
| Predicted number | 600 | 623 | 643 | 634 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
The effectiveness of the mechanism in monochromatic grayscale images was verified in the previous set of experiments. However, ambient light in the real world often appears as gradient colors because the intensity of light is usually affected by many factors, such as temperature, humidity, light source proximity and irradiation angle. Therefore, in this set of experiments, the background of the images is replaced with a gradient grayscale color. It should be noted that to restore the background under various lighting conditions to the extent possible, the gradient direction of the grayscale background is random, and the grayscale value of the background is also random. An example is given in Figure 8. The background of this sample image is a grayscale color that fades to white (grayscale values become larger) from the upper left to the lower right, and there is an object of approximately 100 pixels in size and 45° orientation in the image. The detection result on the right side of the image shows that the activation intensity of neurons in the 45° direction reaches 336, which is stronger than the activation in other directions. Thus, the detection result of the mechanism is 45°, which matches with the actual orientation of the object. The results of this set of experiments are shown in Table 2, and it is clearly observed that the detection accuracy remains at 100% in all orientations. This indicates that even if the background is changed to a random gradient grayscale color, the proposed mechanism still recognizes the object independent of the background grayscale, object size, location, etc. It further supports the robustness and feasibility of the mechanism.
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 580 | 647 | 653 | 620 | 2500 |
| Predicted number | 580 | 647 | 653 | 620 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 611 | 635 | 648 | 606 | 2500 |
| Predicted number | 611 | 635 | 648 | 606 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 602 | 602 | 670 | 626 | 2500 |
| Predicted numbers | 602 | 602 | 670 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 642 | 616 | 591 | 651 | 2500 |
| Predicted number | 642 | 616 | 591 | 651 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
Considering the excellent performance of CNNs in image processing, it is informative to compare the performance of the PODM with the CNN. The images for this set of experiments are still set to have gradient grayscale backgrounds, while, to increase the difficulty, different levels of salt-and-pepper noise are added to all of the images to test noise resistance. Salt-and-pepper noise, also known as impulse noise, is often observed in images. It is a random appearance of white or black dots, either as black pixels in bright areas or white pixels in dark areas (or both). The cause of such noise in an image is usually a sudden and strong disturbance of the transmitted signal. A partial failure of the sensor produces black dots (pepper) on the image, whereas an oversaturation of the sensor produces white dots (salt)[33]. Since the size of the largest object in the dataset in this study is about 1000 pixels, which occupies only 10% of the image size, a noise value is added in the range of 1–10% of the image size so that the maximum noise value is controlled to be no more than the size of the object. Two example sets of images with added noise are presented in Figure 10. The original image without noise is placed on the left, while the examples on the right present the actual image when the added noise is incremented from 1 to 10%. The added pepper noise is evenly and randomly distributed over the image, while the ratio of pepper noise to salt noise is 1:1. Figure 10(a) shows a horizontally placed object of size 50 pixels. When excessive noise is added, the shape of the object becomes greatly disturbed, offering a severe test for the detection mechanism. Figure 10(b) shows an object with an orientation of 135° and a size of 1000 pixels. Both sets of images are added with equal noise values, and it is obvious to the naked eye that the shapes of large objects are easier to distinguish compared with small objects. The detection results made by PODM for the orientation of the objects on these images are also similar to the perception of human eye observation, which to some extent indicates that PODM is built in accordance with the logic of human eye work.
The CNN is a feedforward neural network that includes convolutional computation and has a deep structure [34]. The "neocognitron" neural network proposed by Fukushima in 1980 is considered to be the inspiration for the CNN [35]. Alex formally proposed the first CNN called "time delay network" in 1987, which is mainly applied to the field of sound recognition [36]. In 1988, Wei Zhang proposed the first two-dimensional CNN [37]. Recently, the application area of CNNs has been extended to include portrait recognition [38] and gesture recognition [39]. An input layer, hidden layers and an output layer are the three components of a CNN. The hidden layer usually contains n convolutional and pooling layers and a fully connected layer. CNN usually consumes a large number of computational resources in the training process, and the more convolutional layers there are, the more computational resources are required. Therefore, to ensure the fairness of the experimental comparison, we allocate the computational resources of both methods using the time required to implement both methods on the same platform as a uniform metric (with an error of no more than 10%). In this experiment, only two convolutional layers are set, the size of the convolutional kernel is 3*3, the step size is 1, and the ratio of training set to test set is 7:3. A simplified diagram of the CNN for the present experiment is displayed in Figure 9. With such acceptable computational resource consumption, each set of experiments runs 30 times, and the results derived from the experiments will eventually be tested by hypothesis testing to verify whether the results are significantly different, as shown in Tables 3–6. These tables give the mean and standard deviation (SD) differences of the detection accuracies obtained from the two methods running 30 times each in the four sets of control experiments when adding different levels of noise to the images. Higher accuracy rates are bolded in the table. It is clear that PODM outperforms CNN in most instances at accurately detecting the orientation of an object. The detection accuracies of PODM and CNN are only comparable in images with large objects and high noise values. We employed a P-test to examine whether the accuracies of the two methods are fundamentally different. The P-test is a statistical method that is applied to examine the validity of a generally accepted hypothesis about the aggregate. The smaller the p-value is, the more evidence there is that the null hypothesis should be rejected, and the alternative hypothesis is more plausible [40]. In this set of experiments, the p-value is less than 0.05, which implies that there is a significant difference between the detection results of PODM and CNN. In other words, the detection results of PODM are significantly better than those of CNN. As can be seen from the results in the tables, all of the p-values are less than 0.05, which means there is a significant difference between the results obtained by the two methods. This indicates the PODM is always capable of achieving better detection results than those obtained by the CNN regardless of object size.
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0149 | 67.83 | ± | 0.1596 |
| 2% | 99.40 | ± | 0.1228 | 52.71 | ± | 0.1674 |
| 3% | 97.16 | ± | 0.3379 | 45.52 | ± | 0.1447 |
| 4% | 92.95 | ± | 0.4359 | 41.59 | ± | 0.1298 |
| 5% | 87.69 | ± | 0.7140 | 41.61 | ± | 0.1813 |
| 6% | 81.84 | ± | 0.5256 | 35.97 | ± | 0.1021 |
| 7% | 76.59 | ± | 0.7738 | 33.96 | ± | 0.1017 |
| 8% | 70.99 | ± | 0.7243 | 31.58 | ± | 0.1052 |
| 9% | 66.31 | ± | 0.8904 | 30.90 | ± | 0.0748 |
| 10% | 62.28 | ± | 0.7972 | 32.30 | ± | 0.0889 |
| P-value | - | 8.64E-07 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.98 | ± | 0.0267 | 93.03 | ± | 0.0236 |
| 2% | 99.54 | ± | 0.1267 | 90.87 | ± | 0.0399 |
| 3% | 98.60 | ± | 0.2016 | 85.30 | ± | 0.1019 |
| 4% | 96.99 | ± | 0.2789 | 84.34 | ± | 0.1290 |
| 5% | 95.34 | ± | 0.3437 | 79.01 | ± | 0.1528 |
| 6% | 93.11 | ± | 0.4067 | 77.87 | ± | 0.1796 |
| 7% | 90.43 | ± | 0.4366 | 78.39 | ± | 0.1530 |
| 8% | 87.91 | ± | 0.6577 | 76.80 | ± | 0.1522 |
| 9% | 84.91 | ± | 0.5372 | 72.51 | ± | 0.1787 |
| 10% | 82.16 | ± | 0.7730 | 75.30 | ± | 0.1598 |
| P-value | - | 9.53E-04 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 100.00 | ± | 0.0100 | 97.58 | ± | 0.0067 |
| 2% | 99.94 | ± | 0.0503 | 97.17 | ± | 0.0068 |
| 3% | 99.69 | ± | 0.1204 | 96.93 | ± | 0.0072 |
| 4% | 99.34 | ± | 0.1709 | 95.96 | ± | 0.0121 |
| 5% | 98.91 | ± | 0.1891 | 96.08 | ± | 0.0092 |
| 6% | 98.42 | ± | 0.2377 | 95.80 | ± | 0.0078 |
| 7% | 97.81 | ± | 0.3068 | 95.25 | ± | 0.0084 |
| 8% | 97.19 | ± | 0.3345 | 95.26 | ± | 0.0063 |
| 9% | 96.52 | ± | 0.2720 | 94.83 | ± | 0.0100 |
| 10% | 95.74 | ± | 0.3695 | 94.67 | ± | 0.0077 |
| P-value | - | 5.13E-04 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0072 | 96.71 | ± | 0.0065 |
| 2% | 99.81 | ± | 2.8279 | 96.49 | ± | 0.0055 |
| 3% | 99.43 | ± | 0.7075 | 96.29 | ± | 0.0070 |
| 4% | 98.85 | ± | 0.1964 | 96.13 | ± | 0.0064 |
| 5% | 98.16 | ± | 0.1602 | 95.85 | ± | 0.0101 |
| 6% | 97.41 | ± | 0.2816 | 95.66 | ± | 0.0058 |
| 7% | 96.71 | ± | 0.3358 | 95.90 | ± | 0.0075 |
| 8% | 95.70 | ± | 0.3497 | 95.43 | ± | 0.0115 |
| 9% | 94.91 | ± | 0.3863 | 95.93 | ± | 0.0087 |
| 10% | 93.77 | ± | 0.3876 | 95.58 | ± | 0.0096 |
| P-value | - | 4.76E-02 | ||||
DownLoad:
CSV
The results are also shown in the form of a line graph in Figure 11 for a more in-depth visual analysis of the differences between the two methods. It is obvious that the CNN responds more obviously to the effect of noise, and the obtained results are usually accompanied by certain volatility. This also indicates that the PODM has better noise tolerance than the CNN. Furthermore, in the recognition of small objects, especially when the noise value is larger than that of small objects, the recognition accuracy of the CNN decreases sharply, whereas the proposed PODM maintains acceptable accuracy. The results of this set of experiments reveal that the PODM has many advantages over the CNN, such as higher accuracy in detecting the orientation of objects (especially for small objects), better noise tolerance, less sensitivity to the influence of external factors and better robustness.
In this subsection, a comparison between the original PODM with a 3*3 receptive field and the evolved PODM with a 2*3 receptive field is performed. Gradient grayscale images make up this dataset for the comparative experiment. The size of the detected objects in the images is still divided into four categories, and the images are also added with different levels of noise. The results are shown in Tables 7–10. As can be seen from these tables, the detection accuracies achieved by both models with the same noise impact are very close, regardless of how the size of the object varies. The p-values obtained from the statistical hypothesis tests are all greater than 0.05, which indicates there is no difference in the final results between the two mechanisms with different receptive fields. Unlike the differences between the objects compared in the previous section, neither mechanism in this group imposes a high computational load on the computer, so the length of time taken is the only criterion for comparing the two mechanisms. The time recorded at the bottom of the table clearly shows that the mechanism with a perceptual field of 3*3 takes a significantly longer time. Furthermore, the reduced perceptual field of the mechanism does not reduce the accuracy of recognition. In other words, the improved PODM reduces the reuse rate of information while still maintaining the efficiency and robustness of orientation detection. This indicates that it is necessary and beneficial for the model to change the shape of the receptive field from 3*3 to 2*3.
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.98 | ± | 0.0246 | 100.00 | ± | 0.0147 |
| 2% | 99.17 | ± | 0.1496 | 99.36 | ± | 0.1210 |
| 3% | 96.42 | ± | 0.3577 | 96.96 | ± | 0.3363 |
| 4% | 92.11 | ± | 0.6329 | 92.80 | ± | 0.4348 |
| 5% | 86.64 | ± | 0.5642 | 88.00 | ± | 0.7038 |
| 6% | 80.74 | ± | 0.9089 | 81.68 | ± | 0.5230 |
| 7% | 75.04 | ± | 0.8508 | 77.48 | ± | 0.7652 |
| 8% | 70.24 | ± | 0.8117 | 71.56 | ± | 0.7256 |
| 9% | 65.31 | ± | 0.8048 | 66.24 | ± | 0.8806 |
| 10% | 61.68 | ± | 0.9211 | 62.28 | ± | 0.7914 |
| P-value | - | 8.84E-01 | ||||
| Time cost | 2193.02 s | 3378.61 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.95 | ± | 0.0377 | 99.96 | ± | 0.0266 |
| 2% | 99.39 | ± | 0.1232 | 99.72 | ± | 0.1251 |
| 3% | 98.23 | ± | 0.2248 | 98.68 | ± | 0.1983 |
| 4% | 96.71 | ± | 0.3103 | 97.00 | ± | 0.2752 |
| 5% | 94.97 | ± | 0.3515 | 95.08 | ± | 0.3446 |
| 6% | 92.70 | ± | 0.4397 | 92.88 | ± | 0.4001 |
| 7% | 90.19 | ± | 0.4157 | 90.44 | ± | 0.4311 |
| 8% | 87.31 | ± | 0.6059 | 87.44 | ± | 0.6692 |
| 9% | 84.57 | ± | 0.5317 | 84.36 | ± | 0.5287 |
| 10% | 81.49 | ± | 0.7801 | 82.12 | ± | 0.7805 |
| P-value | - | 9.40E-01 | ||||
| Time cost | 2195.05 s | 3312.37 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0149 | 100.00 | ± | 0.0098 |
| 2% | 99.91 | ± | 0.0572 | 99.84 | ± | 0.0507 |
| 3% | 99.65 | ± | 0.1274 | 99.72 | ± | 0.1222 |
| 4% | 99.27 | ± | 0.1609 | 99.40 | ± | 0.1699 |
| 5% | 98.78 | ± | 0.2088 | 98.84 | ± | 0.1891 |
| 6% | 98.32 | ± | 0.2011 | 98.20 | ± | 0.2344 |
| 7% | 97.63 | ± | 0.1843 | 97.64 | ± | 0.3094 |
| 8% | 97.04 | ± | 0.2397 | 97.56 | ± | 0.3291 |
| 9% | 96.41 | ± | 0.3563 | 96.56 | ± | 0.2679 |
| 10% | 95.69 | ± | 0.3067 | 95.72 | ± | 0.3724 |
| P-value | - | 9.06E-01 | ||||
| Time cost | 2214.19 s | 3379.97 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0181 | 100.00 | ± | 0.0071 |
| 2% | 99.81 | ± | 0.0696 | 99.84 | ± | 2.7832 |
| 3% | 99.43 | ± | 0.1429 | 99.56 | ± | 0.6968 |
| 4% | 98.85 | ± | 0.1799 | 99.00 | ± | 0.1933 |
| 5% | 98.16 | ± | 0.2300 | 98.24 | ± | 0.1582 |
| 6% | 97.41 | ± | 0.3432 | 97.32 | ± | 0.2833 |
| 7% | 96.71 | ± | 0.3401 | 96.36 | ± | 0.3413 |
| 8% | 95.70 | ± | 0.3142 | 95.60 | ± | 0.3461 |
| 9% | 94.91 | ± | 0.3150 | 95.08 | ± | 0.3803 |
| 10% | 93.77 | ± | 0.3777 | 94.00 | ± | 0.3815 |
| P-value | - | 9.84E-01 | ||||
| Time cost | 2335.16 s | 3583.72 s | ||||
DownLoad:
CSV
A set of experiments on the orientation detection of natural objects is added in this section to further validate the confidence level of the PODM. This dataset contains 50 images of size 100*100 with objects such as pens of various colors, contrails of airplanes, elongated stars, elongated drops of water, etc. The orientation of these objects in the images also varies, and there are various light disturbances in the background of the images because of the shooting angle. This imposes some demands on the ability of the PODM to detect the orientation. Meanwhile, to further evaluate the performance of the model, we conducted a downsampling operation on all the images in this dataset, and the size of the images after the operation became 50*50. The image downsampling operation usually reduces the image quality while decreasing the image size, so it can perfectly reproduce the actual quality reduction process of the image delivery. Some of the examples are shown in Figure 12, and the detection results on both the original images and downsampled images are displayed in Table 11. It can be seen that the PODM still achieves 100% correctness in dealing with these practical problems, which proves that the proposed model is also effective in dealing with practical problems and that the stability of getting correct results can be trusted.
| Angle | 0° | 45° | 90° | 135° | Total |
| Original images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
| Downsampled images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
The motivation of this study was to propose a perceptron-based orientation detection mechanism (PODM) inspired by the working mechanism of amacrine cells and to verify the effectiveness of the mechanism through a series of experiments. As color images are usually superimposed by multiple channels of grayscale images, it is reasonable to believe that the successful orientation detection of the classical monochrome grayscale image means the mechanism will also be effective for object orientation detection in color images. Images with gradient grayscale backgrounds can restore the state of actual objects to some extent, and the successful detection of object orientation in such images verifies that the mechanism is also competent for actual object orientation recognition. To further corroborate the effectiveness of the mechanism, it is compared with CNN, the state-of-the-art method in image recognition processing. Different levels of noise are added to the dataset to compare the accuracy of both methods for object orientation recognition under the same interference conditions. The results also confirm that the PODM is superior to the CNN in terms of accuracy, noise tolerance and robustness. Because each feature of the object in the dataset is randomly generated, it can be concluded that the orientation detection of the object by the PODM is always efficient and consistent, regardless of the color, size or shape of the object. The detection results of PODM for noise-affected images closely resemble the perception of the human eye, and we are confident that this model can account for the functioning of the human visual system. In other words, it can give biologists a fresh perspective when conducting research on the visual system.
The authors declare there is no conflict of interest.
| [1] |
Affouneh, S., Salha, S., Burgos, D., Khlaif, Z.N., Saifi, A.G., Mater, N., et al., Factors that foster and deter STEM professional development among teachers. Science Education, 2020,104(5): 857‒872. https://doi.org/10.1002/sce.21591 doi: 10.1002/sce.21591
|
| [2] |
Kelley, T.R. and Knowles, J.G., A conceptual framework for integrated STEM education. International Journal of STEM Education, 2016, 3: 1‒11. https://doi.org/10.1186/s40594-016-0046-z doi: 10.1186/s40594-016-0046-z
|
| [3] | Moore, T.J., Johnston, A.C. and Glancy, A.W., STEM integration: A synthesis of conceptual frameworks and definitions. In Handbook of Research on STEM Education, C. C. Johnson, M. J. Mohr-Schroeder, T.J. Moore, and L.D. English, Eds., 2020, 3‒16. New York: Routledge Taylor and Francis Group. |
| [4] | Bybee, R.W., The Case for STEM Education: Challenges and Opportunities, 2013. USA: NSTA Press. |
| [5] | National Research Council[NRC] (1996). National Science Education Standards. Washington, DC: National Academy Press - Joseph Henry Press. |
| [6] |
Roehrig, G.H., Moore, T.J., Wang, H.H. and Park, M.S., Is adding the E enough? Investigating the impact of K‐12 engineering standards on the implementation of STEM integration. School Science and Mathematics, 2012,112(1): 31‒44. https://doi.org/10.1111/j.1949-8594.2011.00112.x doi: 10.1111/j.1949-8594.2011.00112.x
|
| [7] |
Han, J., Kelley, T. and Knowles, J.G., Factors influencing student STEM learning: Self-efficacy and outcome expectancy, 21st century skills, and career awareness. Journal for STEM Education Research, 2021, 4(2): 117‒137. https://doi.org/10.1007/s41979-021-00053-3 doi: 10.1007/s41979-021-00053-3
|
| [8] |
Stehle, S.M. and Peters-Burton, E.E., Developing student 21st Century skills in selected exemplary inclusive STEM high schools. International Journal of STEM Education, 2019, 6(1): 1‒15. https://doi.org/10.1186/s40594-019-0192-1 doi: 10.1186/s40594-019-0192-1
|
| [9] |
Chen, Y., Chow, S.C.F. and So, W.W.M.., School-STEM professional collaboration to diversify stereotypes and increase interest in STEM careers among primary school students. Asia Pacific Journal of Education, 2022, 42(3): 556‒573. https://doi.org/10.1080/02188791.2020.1841604 doi: 10.1080/02188791.2020.1841604
|
| [10] |
Purzer, Ş. and Quintana-Cifuentes, J.P., Integrating engineering in K-12 science education: spelling out the pedagogical, epistemological, and methodological arguments. Disciplinary and Interdisciplinary Science Education Research, 2019, 1(1): 1‒13. https://doi.org/10.1186/s43031-019-0010-0 doi: 10.1186/s43031-019-0010-0
|
| [11] |
Rehmat, A.P. and Hartley, K., Building engineering awareness: Problem based learning approach for STEM integration. Interdisciplinary Journal of Problem-Based Learning, 2020, 14(1). https://doi.org/10.14434/ijpbl.v14i1.28636 doi: 10.14434/ijpbl.v14i1.28636
|
| [12] |
Wendell, K.B. and Rogers, C., Engineering design‐based science, science content performance, and science attitudes in elementary school. Journal of Engineering Education, 2013,102(4): 513‒540. https://doi.org/10.1002/jee.20026 doi: 10.1002/jee.20026
|
| [13] |
Huang, B., Jong, M.S.Y., Tu, Y.F., Hwang, G.J., Chai, C.S. and Jiang, M.Y.C., Trends and exemplary implements of STEM teacher professional development programs in K-12 contexts: A systematic review of empirical studies. Computers & Education, 2022,189: 104577. https://doi.org/10.1016/j.compedu.2022.104577 doi: 10.1016/j.compedu.2022.104577
|
| [14] | Lau, M. and Multani, S., Engineering STEM teacher learning: using a museum-based field experience to foster stem teachers' pedagogical content knowledge for engineering. In Pedagogical Content Knowledge in STEM: Research to Implement, S.M. Uzzo, S.B. Graves, E. Shay, M. Harford, andR. Thompson, Eds., 2018,195‒213. Springer. |
| [15] |
Nesmith, S.M. and Cooper, S., Engineering process as a focus: STEM professional development with elementary STEM‐focused professional development schools. School Science and Mathematics, 2019,119(8): 487‒498. https://doi.org/10.1111/ssm.12376 doi: 10.1111/ssm.12376
|
| [16] |
Ring, E.A., Dare, E.A., Crotty, E.A. and Roehrig, G.H., The evolution of teacher conceptions of STEM education throughout an intensive professional development experience. Journal of Science Teacher Education, 2017, 28(5): 444‒467. https://doi.org/10.1080/1046560X.2017.1356671 doi: 10.1080/1046560X.2017.1356671
|
| [17] |
Rinke, C.R., Gladstone‐Brown, W., Kinlaw, C.R. and Cappiello, J., Characterizing STEM teacher education: Affordances and constraints of explicit STEM preparation for elementary teachers. School Science and Mathematics, 2016,116(6): 300‒309. https://doi.org/10.1111/ssm.12185 doi: 10.1111/ssm.12185
|
| [18] |
Sahin, E., Sarı, U. and Şen, Ö.F., STEM professional development program for gifted education teachers: STEM lesson plan design competence, self-efficacy, computational thinking and entrepreneurial skills. Thinking Skills and Creativity, 2024, 51: 101439. https://doi.org/10.1016/j.tsc.2023.101439 doi: 10.1016/j.tsc.2023.101439
|
| [19] |
Du, W., Liu, D., Johnson, C.C., Sondergeld, T.A., Bolshakova, V.L. and Moore, T.J., The impact of integrated STEM professional development on teacher quality. School Science and Mathematics, 2019,119(2): 105‒114. https://doi.org/10.1111/ssm.12318 doi: 10.1111/ssm.12318
|
| [20] |
Kelley, T.R., Knowles, J.G., Holland, J.D. and Han, J., Increasing high school teachers self-efficacy for integrated STEM instruction through a collaborative community of implement. International Journal of STEM Education, 2020, 7(1): 1‒13. https://doi.org/10.1186/s40594-020-00211-w doi: 10.1186/s40594-020-00211-w
|
| [21] |
Penuel, W.R., Fishman, B.J., Yamaguchi, R. and Gallagher, L.P., What makes professional development effective? Strategies that foster curriculum implementation. American Educational Research Journal, 2007, 44(4): 921‒958. https://doi.org/10.3102/0002831207308221 doi: 10.3102/0002831207308221
|
| [22] |
Johnson, C.C. and Fargo, J.D., Urban school reform enabled by transformative professional development: Impact on teacher change and student learning of science. Urban Education, 2010, 45(1): 4‒29. https://doi.org/10.1177/0042085909352073 doi: 10.1177/0042085909352073
|
| [23] |
Supovitz, J.A. and Turner, H.M., The effects of professional development on science teaching implements and classroom culture. Journal of Research in Science Teaching, 2000, 37(9): 963‒980. https://doi.org/10.1002/1098-2736(200011)37:9<963::AID-TEA6>3.0.CO;2-0 doi: 10.1002/1098-2736(200011)37:9<963::AID-TEA6>3.0.CO;2-0
|
| [24] |
Richmond, G., Dershimer, R.C., Ferreira, M., Maylone, N., Kubitskey, B. and Meriweather, A., Developing and sustaining an educative mentoring model of STEM teacher professional development through collaborative partnership. Mentoring & Tutoring: Partnership in Learning, 2017, 25(1): 5‒26. https://doi.org/10.1080/13611267.2017.1308097 doi: 10.1080/13611267.2017.1308097
|
| [25] |
Heller, J.I., Daehler, K.R., Wong, N., Shinohara, M. and Miratrix, L.W., Differential effects of three professional development models on teacher knowledge and student achievement in elementary science. Journal of Research in Science Teaching, 2012, 49(3): 333‒362. https://doi.org/10.1002/tea.21004 doi: 10.1002/tea.21004
|
| [26] | Krajcik, J., The importance of viable models in the construction of professional development. In Models and approaches to STEM professional development, B.S. Wojnowski, and C. Pea, Eds., 2014, 33‒44. NSTA Press |
| [27] | Moore, T.J., Stohlmann, M.S., Wang, H.H., Tank, K.M. and Roehrig, G.H., Implementation and integration of engineering in K-12 STEM education. In Engineering in Precollege Settings: Research into Implement, J. Strobel, S. Purzer, and M. Cardella, Eds., 2014, 35‒60. Sense Publishers. |
| [28] | Sanders, M., STEM, STEM education, STEMmania. The Technology Teacher, 2009, 68(4): 20‒26. |
| [29] | Felix, A. and Harris, J., A project-based STEM integrated: Alternative energy team challenge for teachers. The Technology Teacher, 2010, 69(5): 29‒34. |
| [30] |
Lou, S.J., Shih, R.C., Ray Diez, C. and Tseng, K.H., The impact of problem-based learning strategies on STEM knowledge integration and attitudes: an exploratory study among female Taiwanese senior high school students. International Journal of Technology and Design Education, 2011, 21: 195‒215. https://doi.org/10.1007/s10798-010-9114-8 doi: 10.1007/s10798-010-9114-8
|
| [31] |
Han, S., Capraro, R. and Capraro, M.M., How science, technology, engineering, and mathematics (STEM) project-based learning (PBL) affects high, middle, and low achievers differently: The impact of student factors on achievement. International Journal of Science and Mathematics Education, 2015, 13: 1089‒1113. https://doi.org/10.1007/s10763-014-9526-0 doi: 10.1007/s10763-014-9526-0
|
| [32] |
Bozkurt Altan, E. and Tan, S., Concepts of creativity in design based learning in STEM education. International Journal of Technology and Design Education, 2021, 31(3): 503‒529. https://doi.org/10.1007/s10798-020-09569-y doi: 10.1007/s10798-020-09569-y
|
| [33] | Duch, B.J., Groh, S.E. and Allen, D.E., The Power of Problem-Based Learning: A Practical" How to" for Teaching Undergraduate Courses in Any Discipline, 1st ed. 2001, Stylus Publishing. |
| [34] | Sahin, A., STEM project-based learning. In STEM Project-Based Learning an Integrated Science, Technology, Engineering, and Mathematics (STEM) Approach: Capraro, R.M., Capraro, M.M., Morgan, J.R., Eds., 2013, 59‒64. SensePublishers. https://doi.org/10.1007/978-94-6209-143-6_7 |
| [35] | Ozel, S., W3 of project-based learning. In STEM-Project Based Learning. an Integrated Science, Technology, Engineering, and Mathematics (STEM) Approach, R.M. Capraro, M.M. Capraro, and J.R. Morgan, Eds., 2013, 43‒49. SensePublishers. https://doi.org/10.1007/978-94-6209-143-6_5 |
| [36] |
Mehalik, M.M., Doppelt, Y. and Schuun, C.D., Middle‐school science through design‐based learning versus scripted inquiry: Better overall science concept learning and equity gap reduction. Journal of Engineering Education, 2008, 97(1): 71‒85. https://doi.org/10.1002/j.2168-9830.2008.tb00955.x doi: 10.1002/j.2168-9830.2008.tb00955.x
|
| [37] | Geng, J., Jong, M.S.Y., and Chai, C.S., Hong Kong teachers' self-efficacy and concerns about STEM education. The Asia-Pacific Education Researcher, 2019, 28: 35‒45. |
| [38] | Margot, K.C. and Kettler, T., Teachers' perception of STEM integration and education: a systematic literature review. International Journal of STEM Education, 2019, 6(1): 1‒16. |
| [39] | Samara, V. and Kotsis, K.T., Primary school teachers' perceptions of using STEM in the classroom attitudes, obstacles, and suggestions: A literature review. Contemporary Mathematics and Science Education, 2023, 4(2): ep23018. |
| [40] |
Bozkurt Altan, E. and Hacıoğlu, Y., Investigation of problem statement developed by science teachers to perform STEM focused activities in their courses. Necatibey Faculty of Education Electronic Journal of Science and Mathematics Education, 2018, 12(2): 487‒507. https://doi.org/10.17522/balikesirnef.506462 doi: 10.17522/balikesirnef.506462
|
| [41] |
Johnson, C.C., Sondergeld, T. and Walton, J.B., A statewide implementation of the critical features of professional development: Impact on teacher outcomes. School Science and Mathematics, 2017,117(7-8): 341‒349. https://doi.org/10.1111/ssm.12251 doi: 10.1111/ssm.12251
|
| [42] | Zhou, X., Shu, L., Xu, Z. and Padrón, Y., The effect of professional development on in-service STEM teachers' self-efficacy: A meta-analysis of experimental studies. International Journal of STEM Education, 2023, 10(1): 37. |
| [43] |
Kennedy, M.M., How does professional development improve teaching? Review of Educational Research, 2016, 86(4): 945–980. https://doi.org/10.3102/0034654315626800 doi: 10.3102/0034654315626800
|
| [44] |
Desimone, L.M., Improving impact studies of teachers' professional development: toward better conceptualizations and measures. Educational Researcher, 2009, 38: 181–199. https://doi.org/10.3102/0013189X08331140 doi: 10.3102/0013189X08331140
|
| [45] |
Darling-Hammond, L., Teacher education around the world: what can we learn from international implement?. European Journal of Teacher Education, 2017, 40(3): 291‒309. https://doi.org/10.1080/02619768.2017.1315399 doi: 10.1080/02619768.2017.1315399
|
| [46] | Yin, R.K., Case Study Methods: Design and Methods (5th Edition), 2009. Sage Publications. |
| [47] | O'reilly, M. and Parker, N., 'Unsatisfactory saturation': a critical exploration of the notion of saturated sample sizes in qualitative research. Qualitative Research, 2013, 13(2): 190‒197. |
| [48] | Onwuegbuzie, A.J. and Leech, N.L., Sampling designs in qualitative research: making the sampling process more public. Qualitative Report, 2007, 12(2): 238‒254. |
| [49] |
Corbin, J.M. and Strauss, A., Grounded theory research: Procedures, canons, and evaluative criteria. Qualitative Sociology, 1990, 13: 3–21. https://doi.org/10.1007/BF00988593 doi: 10.1007/BF00988593
|
| [50] | Darling-Hammond, L., Changing conceptions of teaching and teacher development. Teacher Education Quarterly, 1995, 22(4): 9‒26. |
| [51] | Sherman, R.R., and Webb, R.B., Qualitative research in education: Focus and methods, 2005. New York: Routledge/Falmer. |
| [52] | Creswell, J.W., Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, 2014. Sage Publications. |
| [53] | Eisenhart, M.A. and Howe, K.R., Validity in educational research. In The Handbook of Qualitative Research in Education, M. LeCompte, W. Millroy, J. Preissle, Eds., 1992,642‒680. Academic Press. |
| [54] | Merriam, S.B., Qualitative Research and Case Study Applications in Education: Revised and Expanded from Case Study Research in Education, 1998. Sansome St, San Francisco: Jossey-Bass Publishers. |
| [55] |
Chai, C.S., Teacher professional development for science, technology, engineering and mathematics (STEM) education: A review from the perspectives of technological pedagogical content (TPACK). The Asia-Pacific Education Researcher, 2019, 28(1): 5‒13. https://doi.org/10.1007/s40299-018-0400-7 doi: 10.1007/s40299-018-0400-7
|
| [56] |
English, L.D., STEM education K-12: perspectives on integration. International Journal of STEM Education, 2016, 3: 1‒8. https://doi.org/10.1186/s40594-016-0036-1 doi: 10.1186/s40594-016-0036-1
|
| [57] | Honey, M., Pearson, G. and Schweingruber, H., STEM integration in k-12 education: status, prospects, and an agenda for research, 2014. National Academy of Engineering; National Research Council, National Academies Press, USA. |
| [58] |
Lin, P., Brummelen, J.V., Lukin, G., Williams, R. and Breazeal, C., Zhorai: Designing a conversational agent for children to explore machine learning concepts. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(09): 13381–13388. https://doi.org/10.1609/aaai.v34i09.7061 doi: 10.1609/aaai.v34i09.7061
|
| [59] |
Parker C.E., Stylinski, C.D., Bonney C.R., Schillaci, R. and McAuliffe, C., Examining the quality of technology implementation in STEM classrooms: demonstration of an evaluative framework. Journal of Research on Technology in Education, 2015, 47(2): 105‒121. https://doi.org/10.1080/15391523.2015.999640 doi: 10.1080/15391523.2015.999640
|
| [60] |
Sun, D., Zhan, Y., Wan, Z.H., Yang, Y. and Looi, C., Identifying the roles of technology: A systematic review of STEM education in primary and secondary schools from 2015 to 2023. Research in Science & Technological Education, 2023, 43(1): 145‒169. https://doi.org/10.1080/02635143.2023.2251902 doi: 10.1080/02635143.2023.2251902
|
| [61] | Kiazai, A.N., Siddiqua, N. and Waheed, Z., Challenges in implementing STEM education and role of teacher education programs in mitigating these challenges. International Journal of Distance Education and E-Learning, 2020, 5(2): 123‒137. |
| 1. | Shangce Gao, Rong-Long Wang, Dongbao Jia, Ting Jin, Special Issue: Artificial intelligence and computational intelligence, 2023, 31, 2688-1594, 7556, 10.3934/era.2023381 |
İrfan Gümüş, Esra Bozkurt Altan. Professional development in STEM for science teachers: Examining improvements in lesson planning and implements†[J]. STEM Education, 2025, 5(3): 425-447. doi: 10.3934/steme.2025021
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 646 | 608 | 627 | 619 | 2500 |
| Predicted number | 646 | 608 | 627 | 619 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 641 | 589 | 644 | 626 | 2500 |
| Predicted number | 641 | 589 | 644 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 634 | 646 | 585 | 635 | 2500 |
| Predicted number | 634 | 646 | 585 | 635 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 600 | 623 | 643 | 634 | 2500 |
| Predicted number | 600 | 623 | 643 | 634 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 580 | 647 | 653 | 620 | 2500 |
| Predicted number | 580 | 647 | 653 | 620 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 611 | 635 | 648 | 606 | 2500 |
| Predicted number | 611 | 635 | 648 | 606 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 602 | 602 | 670 | 626 | 2500 |
| Predicted numbers | 602 | 602 | 670 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 642 | 616 | 591 | 651 | 2500 |
| Predicted number | 642 | 616 | 591 | 651 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0149 | 67.83 | ± | 0.1596 |
| 2% | 99.40 | ± | 0.1228 | 52.71 | ± | 0.1674 |
| 3% | 97.16 | ± | 0.3379 | 45.52 | ± | 0.1447 |
| 4% | 92.95 | ± | 0.4359 | 41.59 | ± | 0.1298 |
| 5% | 87.69 | ± | 0.7140 | 41.61 | ± | 0.1813 |
| 6% | 81.84 | ± | 0.5256 | 35.97 | ± | 0.1021 |
| 7% | 76.59 | ± | 0.7738 | 33.96 | ± | 0.1017 |
| 8% | 70.99 | ± | 0.7243 | 31.58 | ± | 0.1052 |
| 9% | 66.31 | ± | 0.8904 | 30.90 | ± | 0.0748 |
| 10% | 62.28 | ± | 0.7972 | 32.30 | ± | 0.0889 |
| P-value | - | 8.64E-07 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.98 | ± | 0.0267 | 93.03 | ± | 0.0236 |
| 2% | 99.54 | ± | 0.1267 | 90.87 | ± | 0.0399 |
| 3% | 98.60 | ± | 0.2016 | 85.30 | ± | 0.1019 |
| 4% | 96.99 | ± | 0.2789 | 84.34 | ± | 0.1290 |
| 5% | 95.34 | ± | 0.3437 | 79.01 | ± | 0.1528 |
| 6% | 93.11 | ± | 0.4067 | 77.87 | ± | 0.1796 |
| 7% | 90.43 | ± | 0.4366 | 78.39 | ± | 0.1530 |
| 8% | 87.91 | ± | 0.6577 | 76.80 | ± | 0.1522 |
| 9% | 84.91 | ± | 0.5372 | 72.51 | ± | 0.1787 |
| 10% | 82.16 | ± | 0.7730 | 75.30 | ± | 0.1598 |
| P-value | - | 9.53E-04 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 100.00 | ± | 0.0100 | 97.58 | ± | 0.0067 |
| 2% | 99.94 | ± | 0.0503 | 97.17 | ± | 0.0068 |
| 3% | 99.69 | ± | 0.1204 | 96.93 | ± | 0.0072 |
| 4% | 99.34 | ± | 0.1709 | 95.96 | ± | 0.0121 |
| 5% | 98.91 | ± | 0.1891 | 96.08 | ± | 0.0092 |
| 6% | 98.42 | ± | 0.2377 | 95.80 | ± | 0.0078 |
| 7% | 97.81 | ± | 0.3068 | 95.25 | ± | 0.0084 |
| 8% | 97.19 | ± | 0.3345 | 95.26 | ± | 0.0063 |
| 9% | 96.52 | ± | 0.2720 | 94.83 | ± | 0.0100 |
| 10% | 95.74 | ± | 0.3695 | 94.67 | ± | 0.0077 |
| P-value | - | 5.13E-04 | ||||
DownLoad:
CSV
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0072 | 96.71 | ± | 0.0065 |
| 2% | 99.81 | ± | 2.8279 | 96.49 | ± | 0.0055 |
| 3% | 99.43 | ± | 0.7075 | 96.29 | ± | 0.0070 |
| 4% | 98.85 | ± | 0.1964 | 96.13 | ± | 0.0064 |
| 5% | 98.16 | ± | 0.1602 | 95.85 | ± | 0.0101 |
| 6% | 97.41 | ± | 0.2816 | 95.66 | ± | 0.0058 |
| 7% | 96.71 | ± | 0.3358 | 95.90 | ± | 0.0075 |
| 8% | 95.70 | ± | 0.3497 | 95.43 | ± | 0.0115 |
| 9% | 94.91 | ± | 0.3863 | 95.93 | ± | 0.0087 |
| 10% | 93.77 | ± | 0.3876 | 95.58 | ± | 0.0096 |
| P-value | - | 4.76E-02 | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.98 | ± | 0.0246 | 100.00 | ± | 0.0147 |
| 2% | 99.17 | ± | 0.1496 | 99.36 | ± | 0.1210 |
| 3% | 96.42 | ± | 0.3577 | 96.96 | ± | 0.3363 |
| 4% | 92.11 | ± | 0.6329 | 92.80 | ± | 0.4348 |
| 5% | 86.64 | ± | 0.5642 | 88.00 | ± | 0.7038 |
| 6% | 80.74 | ± | 0.9089 | 81.68 | ± | 0.5230 |
| 7% | 75.04 | ± | 0.8508 | 77.48 | ± | 0.7652 |
| 8% | 70.24 | ± | 0.8117 | 71.56 | ± | 0.7256 |
| 9% | 65.31 | ± | 0.8048 | 66.24 | ± | 0.8806 |
| 10% | 61.68 | ± | 0.9211 | 62.28 | ± | 0.7914 |
| P-value | - | 8.84E-01 | ||||
| Time cost | 2193.02 s | 3378.61 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.95 | ± | 0.0377 | 99.96 | ± | 0.0266 |
| 2% | 99.39 | ± | 0.1232 | 99.72 | ± | 0.1251 |
| 3% | 98.23 | ± | 0.2248 | 98.68 | ± | 0.1983 |
| 4% | 96.71 | ± | 0.3103 | 97.00 | ± | 0.2752 |
| 5% | 94.97 | ± | 0.3515 | 95.08 | ± | 0.3446 |
| 6% | 92.70 | ± | 0.4397 | 92.88 | ± | 0.4001 |
| 7% | 90.19 | ± | 0.4157 | 90.44 | ± | 0.4311 |
| 8% | 87.31 | ± | 0.6059 | 87.44 | ± | 0.6692 |
| 9% | 84.57 | ± | 0.5317 | 84.36 | ± | 0.5287 |
| 10% | 81.49 | ± | 0.7801 | 82.12 | ± | 0.7805 |
| P-value | - | 9.40E-01 | ||||
| Time cost | 2195.05 s | 3312.37 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0149 | 100.00 | ± | 0.0098 |
| 2% | 99.91 | ± | 0.0572 | 99.84 | ± | 0.0507 |
| 3% | 99.65 | ± | 0.1274 | 99.72 | ± | 0.1222 |
| 4% | 99.27 | ± | 0.1609 | 99.40 | ± | 0.1699 |
| 5% | 98.78 | ± | 0.2088 | 98.84 | ± | 0.1891 |
| 6% | 98.32 | ± | 0.2011 | 98.20 | ± | 0.2344 |
| 7% | 97.63 | ± | 0.1843 | 97.64 | ± | 0.3094 |
| 8% | 97.04 | ± | 0.2397 | 97.56 | ± | 0.3291 |
| 9% | 96.41 | ± | 0.3563 | 96.56 | ± | 0.2679 |
| 10% | 95.69 | ± | 0.3067 | 95.72 | ± | 0.3724 |
| P-value | - | 9.06E-01 | ||||
| Time cost | 2214.19 s | 3379.97 s | ||||
DownLoad:
CSV
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0181 | 100.00 | ± | 0.0071 |
| 2% | 99.81 | ± | 0.0696 | 99.84 | ± | 2.7832 |
| 3% | 99.43 | ± | 0.1429 | 99.56 | ± | 0.6968 |
| 4% | 98.85 | ± | 0.1799 | 99.00 | ± | 0.1933 |
| 5% | 98.16 | ± | 0.2300 | 98.24 | ± | 0.1582 |
| 6% | 97.41 | ± | 0.3432 | 97.32 | ± | 0.2833 |
| 7% | 96.71 | ± | 0.3401 | 96.36 | ± | 0.3413 |
| 8% | 95.70 | ± | 0.3142 | 95.60 | ± | 0.3461 |
| 9% | 94.91 | ± | 0.3150 | 95.08 | ± | 0.3803 |
| 10% | 93.77 | ± | 0.3777 | 94.00 | ± | 0.3815 |
| P-value | - | 9.84E-01 | ||||
| Time cost | 2335.16 s | 3583.72 s | ||||
DownLoad:
CSV
| Angle | 0° | 45° | 90° | 135° | Total |
| Original images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
| Downsampled images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
DownLoad:
CSV
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 646 | 608 | 627 | 619 | 2500 |
| Predicted number | 646 | 608 | 627 | 619 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 641 | 589 | 644 | 626 | 2500 |
| Predicted number | 641 | 589 | 644 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 634 | 646 | 585 | 635 | 2500 |
| Predicted number | 634 | 646 | 585 | 635 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 600 | 623 | 643 | 634 | 2500 |
| Predicted number | 600 | 623 | 643 | 634 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
| Object Size | Angle | 0° | 45° | 90° | 135° | Total |
| 50 | Number of images | 580 | 647 | 653 | 620 | 2500 |
| Predicted number | 580 | 647 | 653 | 620 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 100 | Number of images | 611 | 635 | 648 | 606 | 2500 |
| Predicted number | 611 | 635 | 648 | 606 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 500 | Number of images | 602 | 602 | 670 | 626 | 2500 |
| Predicted numbers | 602 | 602 | 670 | 626 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% | |
| 1000 | Number of images | 642 | 616 | 591 | 651 | 2500 |
| Predicted number | 642 | 616 | 591 | 651 | 2500 | |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0149 | 67.83 | ± | 0.1596 |
| 2% | 99.40 | ± | 0.1228 | 52.71 | ± | 0.1674 |
| 3% | 97.16 | ± | 0.3379 | 45.52 | ± | 0.1447 |
| 4% | 92.95 | ± | 0.4359 | 41.59 | ± | 0.1298 |
| 5% | 87.69 | ± | 0.7140 | 41.61 | ± | 0.1813 |
| 6% | 81.84 | ± | 0.5256 | 35.97 | ± | 0.1021 |
| 7% | 76.59 | ± | 0.7738 | 33.96 | ± | 0.1017 |
| 8% | 70.99 | ± | 0.7243 | 31.58 | ± | 0.1052 |
| 9% | 66.31 | ± | 0.8904 | 30.90 | ± | 0.0748 |
| 10% | 62.28 | ± | 0.7972 | 32.30 | ± | 0.0889 |
| P-value | - | 8.64E-07 | ||||
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.98 | ± | 0.0267 | 93.03 | ± | 0.0236 |
| 2% | 99.54 | ± | 0.1267 | 90.87 | ± | 0.0399 |
| 3% | 98.60 | ± | 0.2016 | 85.30 | ± | 0.1019 |
| 4% | 96.99 | ± | 0.2789 | 84.34 | ± | 0.1290 |
| 5% | 95.34 | ± | 0.3437 | 79.01 | ± | 0.1528 |
| 6% | 93.11 | ± | 0.4067 | 77.87 | ± | 0.1796 |
| 7% | 90.43 | ± | 0.4366 | 78.39 | ± | 0.1530 |
| 8% | 87.91 | ± | 0.6577 | 76.80 | ± | 0.1522 |
| 9% | 84.91 | ± | 0.5372 | 72.51 | ± | 0.1787 |
| 10% | 82.16 | ± | 0.7730 | 75.30 | ± | 0.1598 |
| P-value | - | 9.53E-04 | ||||
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 100.00 | ± | 0.0100 | 97.58 | ± | 0.0067 |
| 2% | 99.94 | ± | 0.0503 | 97.17 | ± | 0.0068 |
| 3% | 99.69 | ± | 0.1204 | 96.93 | ± | 0.0072 |
| 4% | 99.34 | ± | 0.1709 | 95.96 | ± | 0.0121 |
| 5% | 98.91 | ± | 0.1891 | 96.08 | ± | 0.0092 |
| 6% | 98.42 | ± | 0.2377 | 95.80 | ± | 0.0078 |
| 7% | 97.81 | ± | 0.3068 | 95.25 | ± | 0.0084 |
| 8% | 97.19 | ± | 0.3345 | 95.26 | ± | 0.0063 |
| 9% | 96.52 | ± | 0.2720 | 94.83 | ± | 0.0100 |
| 10% | 95.74 | ± | 0.3695 | 94.67 | ± | 0.0077 |
| P-value | - | 5.13E-04 | ||||
| Noise | PODM | CNN | ||||
| Mean(%) | ± | SD | Mean(%) | ± | SD | |
| 1% | 99.99 | ± | 0.0072 | 96.71 | ± | 0.0065 |
| 2% | 99.81 | ± | 2.8279 | 96.49 | ± | 0.0055 |
| 3% | 99.43 | ± | 0.7075 | 96.29 | ± | 0.0070 |
| 4% | 98.85 | ± | 0.1964 | 96.13 | ± | 0.0064 |
| 5% | 98.16 | ± | 0.1602 | 95.85 | ± | 0.0101 |
| 6% | 97.41 | ± | 0.2816 | 95.66 | ± | 0.0058 |
| 7% | 96.71 | ± | 0.3358 | 95.90 | ± | 0.0075 |
| 8% | 95.70 | ± | 0.3497 | 95.43 | ± | 0.0115 |
| 9% | 94.91 | ± | 0.3863 | 95.93 | ± | 0.0087 |
| 10% | 93.77 | ± | 0.3876 | 95.58 | ± | 0.0096 |
| P-value | - | 4.76E-02 | ||||
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.98 | ± | 0.0246 | 100.00 | ± | 0.0147 |
| 2% | 99.17 | ± | 0.1496 | 99.36 | ± | 0.1210 |
| 3% | 96.42 | ± | 0.3577 | 96.96 | ± | 0.3363 |
| 4% | 92.11 | ± | 0.6329 | 92.80 | ± | 0.4348 |
| 5% | 86.64 | ± | 0.5642 | 88.00 | ± | 0.7038 |
| 6% | 80.74 | ± | 0.9089 | 81.68 | ± | 0.5230 |
| 7% | 75.04 | ± | 0.8508 | 77.48 | ± | 0.7652 |
| 8% | 70.24 | ± | 0.8117 | 71.56 | ± | 0.7256 |
| 9% | 65.31 | ± | 0.8048 | 66.24 | ± | 0.8806 |
| 10% | 61.68 | ± | 0.9211 | 62.28 | ± | 0.7914 |
| P-value | - | 8.84E-01 | ||||
| Time cost | 2193.02 s | 3378.61 s | ||||
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.95 | ± | 0.0377 | 99.96 | ± | 0.0266 |
| 2% | 99.39 | ± | 0.1232 | 99.72 | ± | 0.1251 |
| 3% | 98.23 | ± | 0.2248 | 98.68 | ± | 0.1983 |
| 4% | 96.71 | ± | 0.3103 | 97.00 | ± | 0.2752 |
| 5% | 94.97 | ± | 0.3515 | 95.08 | ± | 0.3446 |
| 6% | 92.70 | ± | 0.4397 | 92.88 | ± | 0.4001 |
| 7% | 90.19 | ± | 0.4157 | 90.44 | ± | 0.4311 |
| 8% | 87.31 | ± | 0.6059 | 87.44 | ± | 0.6692 |
| 9% | 84.57 | ± | 0.5317 | 84.36 | ± | 0.5287 |
| 10% | 81.49 | ± | 0.7801 | 82.12 | ± | 0.7805 |
| P-value | - | 9.40E-01 | ||||
| Time cost | 2195.05 s | 3312.37 s | ||||
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0149 | 100.00 | ± | 0.0098 |
| 2% | 99.91 | ± | 0.0572 | 99.84 | ± | 0.0507 |
| 3% | 99.65 | ± | 0.1274 | 99.72 | ± | 0.1222 |
| 4% | 99.27 | ± | 0.1609 | 99.40 | ± | 0.1699 |
| 5% | 98.78 | ± | 0.2088 | 98.84 | ± | 0.1891 |
| 6% | 98.32 | ± | 0.2011 | 98.20 | ± | 0.2344 |
| 7% | 97.63 | ± | 0.1843 | 97.64 | ± | 0.3094 |
| 8% | 97.04 | ± | 0.2397 | 97.56 | ± | 0.3291 |
| 9% | 96.41 | ± | 0.3563 | 96.56 | ± | 0.2679 |
| 10% | 95.69 | ± | 0.3067 | 95.72 | ± | 0.3724 |
| P-value | - | 9.06E-01 | ||||
| Time cost | 2214.19 s | 3379.97 s | ||||
| Receptive field | 2*3 | 3*3 | ||||
| Noise | Mean(%) | ± | SD | Mean(%) | ± | SD |
| 1% | 99.99 | ± | 0.0181 | 100.00 | ± | 0.0071 |
| 2% | 99.81 | ± | 0.0696 | 99.84 | ± | 2.7832 |
| 3% | 99.43 | ± | 0.1429 | 99.56 | ± | 0.6968 |
| 4% | 98.85 | ± | 0.1799 | 99.00 | ± | 0.1933 |
| 5% | 98.16 | ± | 0.2300 | 98.24 | ± | 0.1582 |
| 6% | 97.41 | ± | 0.3432 | 97.32 | ± | 0.2833 |
| 7% | 96.71 | ± | 0.3401 | 96.36 | ± | 0.3413 |
| 8% | 95.70 | ± | 0.3142 | 95.60 | ± | 0.3461 |
| 9% | 94.91 | ± | 0.3150 | 95.08 | ± | 0.3803 |
| 10% | 93.77 | ± | 0.3777 | 94.00 | ± | 0.3815 |
| P-value | - | 9.84E-01 | ||||
| Time cost | 2335.16 s | 3583.72 s | ||||
| Angle | 0° | 45° | 90° | 135° | Total |
| Original images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
| Downsampled images | 16 | 10 | 13 | 11 | 50 |
| Predicted number | 16 | 10 | 13 | 11 | 50 |
| Accuracy rate | 100% | 100% | 100% | 100% | 100% |
