As an advanced technique, compressed sensing has been used for rapid magnetic resonance imaging in recent years, Two-step Iterative Shrinkage Thresholding Algorithm (TwIST) is a popular algorithm based on Iterative Thresholding Shrinkage Algorithm (ISTA) for fast MR image reconstruction. However TwIST algorithms cannot dynamically adjust shrinkage factor according to the degree of convergence. So it is difficult to balance speed and efficiency. In this paper, we proposed an algorithm which can dynamically adjust the shrinkage factor to rebalance the fidelity item and regular item during TwIST iterative process. The shrinkage factor adjusting is judged by the previous reconstructed results throughout the iteration cycle. It can greatly accelerate the iterative convergence while ensuring convergence accuracy. We used MR images with 2 body parts and different sampling rates to simulate, the results proved that the proposed algorithm have a faster convergence rate and better reconstruction performance. We also used 60 MR images of different body parts for further simulation, and the results proved the universal superiority of the proposed algorithm.
Citation: Xiuhan Li, Rui Feng, Funan Xiao, Yue Yin, Da Cao, Xiaoling Wu, Songsheng Zhu, Wei Wang. Sparse reconstruction of magnetic resonance image combined with two-step iteration and adaptive shrinkage factor[J]. Mathematical Biosciences and Engineering, 2022, 19(12): 13214-13226. doi: 10.3934/mbe.2022618
[1] | Muhammad Uzair Awan, Nousheen Akhtar, Artion Kashuri, Muhammad Aslam Noor, Yu-Ming Chu . 2D approximately reciprocal ρ-convex functions and associated integral inequalities. AIMS Mathematics, 2020, 5(5): 4662-4680. doi: 10.3934/math.2020299 |
[2] | Yousaf Khurshid, Muhammad Adil Khan, Yu-Ming Chu . Conformable fractional integral inequalities for GG- and GA-convex functions. AIMS Mathematics, 2020, 5(5): 5012-5030. doi: 10.3934/math.2020322 |
[3] | Hengxiao Qi, Muhammad Yussouf, Sajid Mehmood, Yu-Ming Chu, Ghulam Farid . Fractional integral versions of Hermite-Hadamard type inequality for generalized exponentially convexity. AIMS Mathematics, 2020, 5(6): 6030-6042. doi: 10.3934/math.2020386 |
[4] | Mehmet Eyüp Kiriş, Miguel Vivas-Cortez, Gözde Bayrak, Tuğba Çınar, Hüseyin Budak . On Hermite-Hadamard type inequalities for co-ordinated convex function via conformable fractional integrals. AIMS Mathematics, 2024, 9(4): 10267-10288. doi: 10.3934/math.2024502 |
[5] | Ghulam Farid, Saira Bano Akbar, Shafiq Ur Rehman, Josip Pečarić . Boundedness of fractional integral operators containing Mittag-Leffler functions via (s,m)-convexity. AIMS Mathematics, 2020, 5(2): 966-978. doi: 10.3934/math.2020067 |
[6] | Yousaf Khurshid, Muhammad Adil Khan, Yu-Ming Chu . Conformable integral version of Hermite-Hadamard-Fejér inequalities via η-convex functions. AIMS Mathematics, 2020, 5(5): 5106-5120. doi: 10.3934/math.2020328 |
[7] | Yue Wang, Ghulam Farid, Babar Khan Bangash, Weiwei Wang . Generalized inequalities for integral operators via several kinds of convex functions. AIMS Mathematics, 2020, 5(5): 4624-4643. doi: 10.3934/math.2020297 |
[8] | M. Emin Özdemir, Saad I. Butt, Bahtiyar Bayraktar, Jamshed Nasir . Several integral inequalities for (α, s,m)-convex functions. AIMS Mathematics, 2020, 5(4): 3906-3921. doi: 10.3934/math.2020253 |
[9] | Muhammad Zakria Javed, Muhammad Uzair Awan, Loredana Ciurdariu, Omar Mutab Alsalami . Pseudo-ordering and δ1-level mappings: A study in fuzzy interval convex analysis. AIMS Mathematics, 2025, 10(3): 7154-7190. doi: 10.3934/math.2025327 |
[10] | Xiuzhi Yang, G. Farid, Waqas Nazeer, Muhammad Yussouf, Yu-Ming Chu, Chunfa Dong . Fractional generalized Hadamard and Fejér-Hadamard inequalities for m-convex functions. AIMS Mathematics, 2020, 5(6): 6325-6340. doi: 10.3934/math.2020407 |
As an advanced technique, compressed sensing has been used for rapid magnetic resonance imaging in recent years, Two-step Iterative Shrinkage Thresholding Algorithm (TwIST) is a popular algorithm based on Iterative Thresholding Shrinkage Algorithm (ISTA) for fast MR image reconstruction. However TwIST algorithms cannot dynamically adjust shrinkage factor according to the degree of convergence. So it is difficult to balance speed and efficiency. In this paper, we proposed an algorithm which can dynamically adjust the shrinkage factor to rebalance the fidelity item and regular item during TwIST iterative process. The shrinkage factor adjusting is judged by the previous reconstructed results throughout the iteration cycle. It can greatly accelerate the iterative convergence while ensuring convergence accuracy. We used MR images with 2 body parts and different sampling rates to simulate, the results proved that the proposed algorithm have a faster convergence rate and better reconstruction performance. We also used 60 MR images of different body parts for further simulation, and the results proved the universal superiority of the proposed algorithm.
The inequalities discovered by C. Hermite and J. Hadamard for convex functions are considerable significant in the literature (see, e.g., [9], [18], [27,p.137]). These inequalities state that if f:I→R is a convex function on the interval I of real numbers and a,b∈I with a<b, then
f(a+b2)≤1b−ab∫af(x)dx≤f(a)+f(b)2. | (1.1) |
Both inequalities hold in the reversed direction if f is concave.
The Hermite-Hadamard inequality, which is the first fundamental result for convex mappings with a natural geometrical interpretation and many applications, has drawn attention much interest in elementary mathematics. A number of mathematicians have devoted their efforts.
The most well-known inequalities related to the integral mean of a convex function are the Hermite Hadamard inequalities or its weighted versions, the so-called Hermite-Hadamard-Fejér inequalities. In [17], Fejer gave a weighted generalization of the inequalities (1.1) as the following:
Theorem 1. f:[a,b]→R, be a convex function, then the inequality
f(a+b2)b∫ag(x)dx≤b∫af(x)g(x)dx≤f(a)+f(b)2b∫ag(x)dx | (1.2) |
holds, where g:[a,b]→R is nonnegative, integrable, and symmetric about x=a+b2 (i.e. g(x)=g(a+b−x)).
In this paper we will establish some new Fejér type inequalities for the new concept of co-ordinated hyperbolic ρ-convex functions.
The overall structure of the paper takes the form of four sections including introduction. The paper is organized as follows: we first give the definition of co-ordinated convex functions, the definition of fractional integrals and related Hermite-Hadamard inequality in Section 1. We also recall the concept of hyperbolic ρ-convex functions and co-ordinated hyperbolic ρ-convex functions introduced by Özçelik et. al in [23]. Moreover, we give a lemma and a theorem which will be frequently used in the next section. Some Hermite-Hadamard-Fejer type inequalities for co-ordinated hyperbolic ρ-convex functions are obtained and some special cases of the results are also given in Section 2. Then, we also apply the inequalities obtained in Section 2 to establish some fractional Fejer type inequalities in Section 3. Finally, in Section 4, some conclusions and further directions of research are discussed.
A formal definition for co-ordinated convex function may be stated as follows:
Definition 1. A function f:Δ:=[a,b]×[c,d]→R is called co-ordinated convex on Δ, for all (x,u),(y,v)∈Δ and t,s∈[0,1], if it satisfies the following inequality:
f(tx+(1−t) y,su+(1−s) v)≤ts f(x,u)+t(1−s)f(x,v)+s(1−t)f(y,u)+(1−t)(1−s)f(y,v). | (1.3) |
The mapping f is a co-ordinated concave on Δ if the inequality (1.3) holds in reversed direction for all t,s∈[0,1] and (x,u),(y,v)∈Δ.
In [11], Dragomir proved the following inequalities which is Hermite-Hadamard type inequalities for co-ordinated convex functions on the rectangle from the plane R2.
Theorem 2. Suppose that f:Δ:=[a,b]×[c,d]→R is co-ordinated convex, then we have the following inequalities:
f(a+b2,c+d2)≤12[1b−ab∫af(x,c+d2)dx+1d−cd∫cf(a+b2,y)dy]≤1(b−a)(d−c)b∫ad∫cf(x,y)dydx≤14[1b−ab∫af(x,c)dx+1b−ab∫af(x,d)dx+1d−cd∫cf(a,y)dy+1d−cd∫cf(b,y)dy]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4. | (1.4) |
The above inequalities are sharp. The inequalities in (1.4) hold in reverse direction if the mapping f is a co-ordinated concave mapping.
Over the years, the numerous studies have focused on to establish generalization of the inequality (1.1) and (1.4). For some of them, please see ([1,2,3,4,5,6,7,8], [19,20,21,22,23,24,25,26], [28,29,30,31,32,33,34,35,36]).
Definition 2. [29] Let f∈L1(Δ).The Riemann-Lioville integrals Jα,βa+,c+,Jα,βa+,d−,+Jα,βb−,c+ and Jα,βb−,d−of order α,β>0 with a,c≥0 are defined by
Jα,βa+,c+f(x,y)=1Γ(α)Γ(β)x∫ay∫c(x−t)α−1(y−s)β−1f(t,s)dsdt, x>a, y>c,Jα,βa+,d−f(x,y)=1Γ(α)Γ(β)x∫ad∫y(x−t)α−1(s−y)β−1f(t,s)dsdt, x>a, y>d,Jα,βb−,c+f(x,y)=1Γ(α)Γ(β)b∫xy∫c(t−x)α−1(y−s)β−1f(t,s)dsdt, x<b, y>c,Jα,βb−,d−f(x,y)=1Γ(α)Γ(β)b∫xd∫y(t−x)α−1(s−y)β−1f(t,s)dsdt, x<b, y<d, |
respectively. Here, Γ is the Gamma funtion,
J0,0a+,c+f(x,y)=J0,0a+,d−f(x,y)=J0,0b−,c+f(x,y)=J0,0b−,d−f(x,y) |
and
J1,1a+,c+f(x,y)=x∫ay∫cf(t,s)dsdt. |
First, we give the definition of hyperbolic ρ-convex functions and some related inequalities. Then we define the co-ordinated hyperbolic ρ -convex functions.
Definition 3. [10] A function f:I→R is said to be hyperbolic ρ-convex, if for any arbitrary closed subinterval [a,b] of I such that we have
f(x)≤sinh[ρ(b−x)]sinh[ρ(b−a)]f(a)+sinh[ρ(x−a)]sinh[ρ(b−a)]f(b) | (1.5) |
for all x∈[a,b]. If we take x=(1−t)a+tb, t∈[0,1] in (1.5), then the condition (1.5) becomes
f((1−t)a+tb)≤sinh[ρ(1−t)(b−a)]sinh[ρ(b−a)]f(a)+sinh[ρt(b−a)]sinh[ρ(b−a)]f(b). | (1.6) |
If the inequality (1.5) holds with "≥", then the function will be called hyperbolic ρ-concave on I.
The following Hermite-Hadamard inequality for hyperbolic ρ-convex function is proved by Dragomir in [10].
Theorem 3. Suppose that f:I→R is hyperbolic ρ-convex on I. Then for any a,b∈I, we have
2ρf(a+b2)sinh[ρ(b−a)2]≤b∫af(x)dx≤f(a)+f(b)ρtanh[ρ(b−a)2]. | (1.7) |
Moreover in [12], Dragomir prove the following Hermite Hadamard-Fejer type inequalities for hyperbolic ρ-convex functions.
Theorem 4. Assume that the function f:I→R is hyperbolic ρ-convex on I and a,b∈I. Assume also that p:[a,b]⟶R is a positive, symmetric and integrable function on [a,b], then we have
f(a+b2)b∫acosh[ρ(x−a+b2)]p(x)dx≤b∫af(x)p(x)dx≤f(a)+f(b)2sech[ρ(b−a)2]b∫acosh[ρ(x−a+b2)]p(x)dx. | (1.8) |
For the other inequalities for hyperbolic ρ-convex functions, please refer to ([12,13,14,15]).
Now we give the definition of co-ordinated hyperbolic ρ-convex functions.
Definition 4. [23] A function f:Δ→R is said to co-ordinated hyperbolic ρ-convex on Δ, if the inequality
f(x,y)≤sinh[ρ1(b−x)]sinh[ρ1(b−a)]sinh[ρ2(d−y)]sinh[ρ2(d−c)]f(a,c)+sinh[ρ1(b−x)]sinh[ρ1(b−a)]sinh[ρ2(y−c)]sinh[ρ2(d−c)]f(a,d)+sinh[ρ1(x−a)]sinh[ρ1(b−a)]sinh[ρ2(d−y)]sinh[ρ2(d−c)]f(b,c)+sinh[ρ1(x−a)]sinh[ρ1(b−a)]sinh[ρ2(y−c)]sinh[ρ2(d−c)]f(b,d). | (1.9) |
holds.
If the inequality (1.9) holds with "≥", then the function will be called co-ordinated hyperbolic ρ-concave on Δ.
If we take x=(1−t)a+tb and y=(1−s)c+sd for t,s,∈[0,1], then the inequality (1.9) can be written as
f((1−t)a+tb,(1−s)c+sd)≤sinh[ρ1(1−t)(b−a)]sinh[ρ1(b−a)]sinh[ρ2(1−s)(d−y)]sinh[ρ2(d−c)]f(a,c)+sinh[ρ1(1−t)(b−a)]sinh[ρ1(b−a)]sinh[ρ2s(d−y)]sinh[ρ2(d−c)]f(a,d)+sinh[ρ1t(b−a)]sinh[ρ1(b−a)]sinh[ρ2(1−s)(d−y)]sinh[ρ2(d−c)]f(b,c)+sinh[ρ1(b−a)]sinh[ρ1(b−a)]sinh[ρ2s(d−y)]sinh[ρ2(d−c)]f(b,d). | (1.10) |
Now we give the following useful lemma:
Lemma 1. [23] If f:Δ=[a,b]×[c,d]→R is co-ordinated ρ-convex function on Δ, then we have the following inequality
cosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]f(a+b2,c+d2)≤14[f(x,y)+f(x,c+d−y)+f(a+b−x,y)+f(a+b−x,c+d−y)]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4cosh[ρ1(x−a+b2)]cosh[ρ1(b−a)2]cosh[ρ2(y−c+d2)]cosh[ρ2(d−c)2] | (1.11) |
for all (x,y)∈Δ.
Theorem 5. Let p:Δ→R be a positive, integrable and symmetric about a+b2 and c+d2. Let, f:Δ→R be a co-ordinated hyperbolic ρ-convex functions on Δ. We have the following Hermite-Hadamard-Fejer type inequalities:
f(a+b2,c+d2)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx≤b∫ad∫cf(x,y)p(x,y)dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4cosh[ρ1(b−a)2]cosh[ρ2(d−c)2]×b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx. | (2.1) |
Proof. Multiplying the inequality (1.1) by p(x,y)>0 and then integrating with respect to (x,y) on Δ, we obtain
f(a+b2,c+d2)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx≤14b∫ad∫c[f(x,y)+f(x,c+d−y)+f(a+b−x,y)+f(a+b−x,c+d−y)]p(x,y)dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4cosh[ρ1(b−a)2]cosh[ρ2(d−c)2]×b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx | (2.2) |
Since p is symmetric about a+b2 and c+d2, one can show that
b∫ad∫cf(x,c+d−y)p(x,y)dydx=b∫ad∫cf(a+b−x,y)p(x,y)dydx=b∫ad∫cf(a+b−x,c+d−y)p(x,y)dydx=b∫ad∫cf(x,y)p(x,y)dydx. |
This completes the proof.
Remark 1. If we choose p(x,y)=1 in Theorem 5, then we have the following the inequality
4ρ1ρ2sinh[ρ1(b−a)2]sinh[ρ2(d−c)2]f(a+b2,c+d2)≤b∫ad∫cf(x,y)dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)ρ1ρ2tanh[ρ1(b−a)2]tanh[ρ2(d−c)2] |
which is proved by Özçelik et. al in [23].
Corollary 1. Suppose that all assumptions of Theorem 5 are satisfied. Then we have the following inequality,
f(a+b2,c+d2)b∫ad∫cw(x,y)dydx≤b∫ad∫cf(x,y)w(x,y)sech[ρ1(x−a+b2)]sech[ρ2(y−c+d2)]dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]b∫ad∫cw(x,y)dydx. | (2.3) |
Proof. Let us define the function p(x,y) by
w(x,y)=p(x,y)cosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]. |
Clearly, w(x.y) is a a positive, integrable and symmetric about a+b2 and c+d2. If we apply Theorem 5 for the function w(x,y) then we establish the desired inequality (2.3).
Remark 2. If we choose w(x,y)=1 for all (x,y)ϵΔ in Corollary 1, then we have the following the inequality
f(a+b2,c+d2)≤1(b−a)(d−c)b∫ad∫cf(x,y)sech[ρ1(x−a+b2)]sech[ρ2(y−c+d2)]dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]. | (2.4) |
which is proved by Özçelik et. al in [23].
Theorem 6. Let p:Δ→R be a positive, integrable and symmetric about a+b2 and c+d2. Let f:Δ→R be a co-ordinated hyperbolic ρ-convex on Δ, then we have the following Hermite-Hadamard-Fejer type inequalities
f(a+b2,c+d2)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx≤12[b∫ad∫cf(x,c+d2)cosh[ρ2(y−c+d2)]p(x,y)dydx+b∫ad∫cf(a+b2,y)cosh[ρ1(x−a+b2)]p(x,y)dydx]≤b∫ad∫cf(x,y)p(x,y)dydx≤14[sech[ρ2(d−c)2]b∫ad∫c[f(x,c)+f(x,d)]cosh[ρ2(y−c+d2)]p(x,y)dydx+sech[ρ1(b−a)2]b∫ad∫c[f(a,y)+f(b,y)]cosh[ρ1(x−a+b2)]p(x,y)dydx]≤f(a,c)+f(b,c)+f(a,d)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]×b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx. | (2.5) |
Proof. Since f is co-ordinated hyperbolic ρ-convex on Δ, if we define the mappings fx:[c,d]→R, fx(y)=f(x,y) and px:[c,d]→R, px(y)=p(x,y), then fx(y) is hyperbolic ρ-convex on [c,d] and px(y) is positive, integrable and symmetric about c+d2 for all x∈[a,b]. If we apply the inequality (1.8) for the hyperbolic ρ-convex function fx(y), then we have
fx(c+d2)d∫ccosh[ρ2(y−c+d2)]px(y)dy≤d∫cfx(y)px(y)dy≤fx(c)+fx(d)2sech[ρ2(d−c)2]d∫ccosh[ρ2(y−c+d2)]px(y)dy. | (2.6) |
That is,
f(x,c+d2)d∫ccosh[ρ2(y−c+d2)]p(x,y)dy≤d∫cf(x,y)p(x,y)dy≤f(x,c)+f(x,d)2sech[ρ2(d−c)2]d∫ccosh[ρ2(y−c+d2)]p(x,y)dy. | (2.7) |
Integrating the inequality (2.7) with respect to x from a to b, we obtain
b∫ad∫cf(x,c+d2)cosh[ρ2(y−c+d2)]p(x,y)dydx≤b∫ad∫cf(x,y)p(x,y)dydx≤12b∫ad∫c[f(x,c)+f(x,d)]sech[ρ2(d−c)2]cosh[ρ2(y−c+d2)]p(x,y)dydx. | (2.8) |
Similarly, as f is co-ordinated hyperbolic ρ-convex on Δ, if we define the mappings fy:[a,b]→R, fy(x)=f(x,y) and py:[a,b]→R, py(x)=p(x,y), then fy(x) is hyperbolic ρ-convex on [a,b] and py(x) is positive, integrable and symmetric about a+b2 for all y∈[c,d]. Utilizing the inequality (1.8) for the hyperbolic ρ-convex function fy(x), then we obtain the inequality
fy(a+b2)b∫acosh[ρ1(x−a+b2)]py(x)dx≤b∫afy(x)py(x)dx≤fy(a)+fy(b)2sech[ρ1(b−a)2]b∫acosh[ρ1(x−a+b2)]py(x)dx | (2.9) |
i.e.
f(a+b2,y)b∫acosh[ρ1(x−a+b2)]p(x,y)dx≤b∫af(x,y)p(x,y)dx≤f(a,y)+f(b,y)2sech[ρ1(b−a)2]b∫acosh[ρ1(x−a+b2)]p(x,y)dx. | (2.10) |
Integrating the inequality (2.10) with respect to y on [c,d], we get
b∫ad∫cf(a+b2,y)cosh[ρ1(x−a+b2)]p(x,y)dydx≤b∫ad∫cf(x,y)p(x,y)dydx≤12b∫ad∫c[f(a,y)+f(b,y)]sech[ρ1(b−a)2]cosh[ρ1(x−a+b2)]p(x,y)dydx. | (2.11) |
Summing the inequalities (2.8) and (2.11), we obtain the second and third inequalities in (2.5).
Since f(a+b2,y) is hyperbolic ρ-convex on [c,d] and px(y) is positive, integrable and symmetric about c+d2, using the first inequality in (1.8), we have
f(a+b2,c+d2)d∫ccosh[ρ2(y−c+d2)]p(x,y)dy≤d∫cf(a+b2,y)p(x,y)dy. | (2.12) |
Multiplying the inequality (2.12) by cosh[ρ1(x−a+b2)] and integrating resulting inequality with respect to x on [a,b], we get
f(a+b2,c+d2)b∫ad∫ccosh[ρ2(y−c+d2)]cosh[ρ1(x−a+b2)]p(x,y)dydx≤b∫ad∫cf(a+b2,y)cosh[ρ1(x−a+b2)]p(x,y)dydx. | (2.13) |
Since f(x,c+d2) is hyperbolic ρ-convex on [a,b] and py(x) is positive, integrable and symmetric about a+b2, utilizing the first inequality in (1.8), we have the following inequality
f(a+b2,c+d2)b∫acosh[ρ1(x−a+b2)]p(x,y)dx≤b∫af(x,c+d2)p(x,y)dx. | (2.14) |
Multiplying the inequality (2.14) by cosh[ρ2(y−c+d2)] and integrating resulting inequality with respect to y on [c,d], we get
f(a+b2,c+d2)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx≤b∫ad∫cf(x,c+d2)cosh[ρ2(y−c+d2)]p(x,y)dydx. | (2.15) |
From the inequalities (2.13) and (2.15), we obtain the first inequality in (2.5).
For the proof of last inequality in (2.5), using the second inequality in (1.8) for the hyperbolic ρ-convex functions f(x,c) and f(x,d) on [a,b] and for the symmetric function py(x), we obtain the inequalities
b∫af(x,c)p(x,y)dx≤f(a,c)+f(b,c)2sech[ρ1(b−a)2]b∫acosh[ρ1(x−a+b2)]p(x,y)dx | (2.16) |
and
b∫af(x,d)p(x,y)dx≤f(a,d)+f(b,d)2sech[ρ1(b−a)2]b∫acosh[ρ1(x−a+b2)]p(x,y)dx. | (2.17) |
If we multiply the inequalities (2.16) and (2.17) by sech[ρ2(d−c)2]cosh[ρ2(y−c+d2)] and integrating the resulting inequalities on [c,d], then we have
b∫ad∫cf(x,c)sech[ρ2(d−c)2]cosh[ρ2(y−c+d2)]p(x,y)dydx≤f(a,c)+f(b,c)2sech[ρ1(b−a)2]sech[ρ2(d−c)2]×b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx | (2.18) |
and
b∫ad∫cf(x,d)sech[ρ2(d−c)2]cosh[ρ2(y−c+d2)]p(x,y)dydx≤f(a,d)+f(b,d)2sech[ρ1(b−a)2]sech[ρ2(d−c)2]×b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]p(x,y)dydx. | (2.19) |
Similarly, applying the second inequality in (1.8) for the hyperbolic ρ-convex functions f(a,y) and f(b,y) on [c,d] and for the symmetric function px(y), we have
d∫cf(a,y)p(x,y)dy≤f(a,c)+f(a,d)2sech[ρ2(d−c)2]d∫ccosh[ρ2(y−c+d2)]p(x,y)dy | (2.20) |
and
d∫cf(b,y)p(x,y)dy≤f(b,c)+f(b,d)2sech[ρ2(d−c)2]d∫ccosh[ρ2(y−c+d2)]p(x,y)dy. | (2.21) |
Multiplying the inequalities (2.20) and (2.21) by sech[ρ1(b−a)2]cosh[ρ1(x−a+b2)] and integrating the resulting inequalities on [a,b], then we have
b∫ad∫cf(a,y)sech[ρ1(b−a)2]cosh[ρ1(x−a+b2)]p(x,y)dydx≤f(a,c)+f(a,d)2sech[ρ2(d−c)2]sech[ρ1(b−a)2]×b∫ad∫ccosh[ρ2(y−c+d2)]cosh[ρ1(x−a+b2)]p(x,y)dydx | (2.22) |
and
b∫ad∫cf(b,y)sech[ρ1(b−a)2]cosh[ρ1(x−a+b2)]p(x,y)dydx≤f(b,c)+f(b,d)2sech[ρ2(d−c)2]sech[ρ1(b−a)2]×b∫ad∫ccosh[ρ2(y−c+d2)]cosh[ρ1(x−a+b2)]p(x,y)dydx. | (2.23) |
Summing the inequalities (2.18), (2.19), (2.22) and (2.23), we establish the last inequality in (2.5). This completes the proof.
Remark 3. If we choose p(x,y)=1 in Theorem 6, then we have
4ρ1ρ2sinh[ρ1(b−a)2]sinh[ρ2(d−c)2]f(a+b2,c+d2)≤1ρ1sinh[ρ1(b−a)2]d∫cf(a+b2,y)dy+1ρ2sinh[ρ2(d−c)2]b∫af(x,c+d2)dx≤b∫ad∫cf(x,y)dydx≤12[1ρ2tanh[ρ2(d−c)2]b∫a[f(x,c)+f(x,d)]dx+1ρ1tanh[ρ1(b−a)2]d∫c[f(a,y)+f(b,y)]dy]≤tanh[ρ1(b−a)2]tanh[ρ2(d−c)2]f(a,c)+f(a,d)+f(b,c)+f(b,d)ρ1ρ2 | (2.24) |
which is proved by Özçelik et. al in [23].
Remark 4. Choosing ρ1=ρ2=0 in Theorem 6, we obtain
f(a+b2,c+d2)b∫ad∫cp(x,y)dydx≤12b∫ad∫c[f(x,c+d2)+f(a+b2,y)]p(x,y)dydx≤b∫ad∫cf(x,y)p(x,y)dydx≤14b∫ad∫c[f(x,c)+f(x,d)+f(a,y)+f(b,y)]p(x,y)dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4b∫ad∫cp(x,y)dydx. |
which is proved by Budak and Sarikaya in [5].
Corollary 2. Let g1:[a,b]→R and g1:[c,d]→R be two positive, integrable and symmetric about a+b2 and c+d2, respectively. If we choose p(x,y)=g1(x)g2(y)G1G2 for all (x,y)∈Δ in Theorem 6, then we have
f(a+b2,c+d2)≤12[1G1b∫af(x,c+d2)g1(x)dx+1G2d∫cf(a+b2,y)g2(y)dy]≤1G1G2b∫ad∫cf(x,y)g1(x)g2(y)dydx≤14[sech[ρ2(d−c)2]1G1b∫a[f(x,c)+f(x,d)]g1(x)dx+sech[ρ1(b−a)2]1G2d∫c[f(a,y)+f(b,y)]g2(y)dy]≤f(a,c)+f(b,c)+f(a,d)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2] | (2.25) |
where
G1=b∫acosh[ρ1(x−a+b2)]g1(x)dxandG2=d∫ccosh[ρ2(y−c+d2)]g2(y)dy. |
Remark 5. If we choose ρ1=ρ2=0 in Corollary 2, then we have
f(a+b2,c+d2)≤12[1G1b∫af(x,c+d2)g1(x)dx+1G2d∫cf(a+b2,y)g2(y)dy]≤1G1G2b∫ad∫cf(x,y)g1(x)g2(y)dydx≤14[1G1b∫a[f(x,c)+f(x,d)]g1(x)dx+1G2d∫c[f(a,y)+f(b,y)]g2(y)dy]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4 |
which is proved by Farid et al. in [16].
In this section we obtain some fractional Hermite-Hadamard an Fejer type inequalities for co-ordinated hyperbolic ρ-convex functions.
Theorem 7. If f:Δ→R is a co-ordinated hyperbolic ρ-convex functions on Δ, then we have the following Hermite-Hadamard and Fejer type inequalities,
f(a+b2,c+d2)H(α,β)≤[Jα,βa+,c+f(b,d)+Jα,βa+,d−f(b,c)+Jα,βb−,c+f(a,d)+Jα,βb−,d−f(a,c)]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]H(α,β) |
where
H(α,β)=1Γ(α)Γ(β)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]×[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx. |
Proof. If we apply Theorem 5 for the symmetric function
p(x,y)=1Γ(α)Γ(β)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1], |
then we get the following inequality
f(a+b2,c+d2)H(α,β)≤1Γ(α)Γ(β)b∫ad∫cf(x,y)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]H(α,β). |
From the definition of the double fractional integrals we have
1Γ(α)Γ(β)b∫ad∫cf(x,y)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx=[Jα,βa+,c+f(b,d)+Jα,βa+,d−f(b,c)+Jα,βb−,c+f(a,d)+Jα,βb−,d−f(a,c)] |
which completes the proof.
Remark 6. If we choose ρ1=ρ2=0 in Theorem 7, then we have the following fractional Hermite-Hadamard inequality,
f(a+b2,c+d2)≤Γ(α+1)Γ(β+1)4(b−a)α(d−c)β[Jα,βa+,c+f(b,d)+Jα,βa+,d−f(b,c)+Jα,βb−,c+f(a,d)+Jα,βb−,d−f(a,c)]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4 |
which was proved by Sarikaya in [29,Theorem 4].
Remark 7. If we choose α =β=1 in Theorem 7, then we have
H(1,1)=16ρ1ρ2sinh(ρ1(b−a)2)sinh(ρ2(d−c)2). |
Thus, we get the following Hermite-Hadamard inequality,
4ρ1ρ2f(a+b2,c+d2)sinh(ρ1(b−a)2)sinh(ρ2(d−c)2)≤b∫ad∫cf(x,y)dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)ρ1ρ2tanh[ρ1(b−a)2]tanh[ρ2(d−c)2] |
which is proved by Özçelik et al. in [23].
Theorem 8. Let p:Δ→R be a positive, integrable and symmetric about a+b2 and c+d2. If f:Δ→R is a co-ordinated hyperbolic ρ-convex functions on Δ, then we have the following Hermite-Hadamard-Fejer type inequalities,
f(a+b2,c+d2)Hp(α,β)≤[Jα,βa+,c+(fp)(b,d)+Jα,βa+,d−(fp)(b,c)+Jα,βb−,c+(fp)(a,d)+Jα,βb−,d−(fp)(a,c)]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]Hp(α,β) |
where
Hp(α,β)=1Γ(α)Γ(β)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]×[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]p(x,y)dydx. |
Proof. Let us define the function k(x,y) by
k(x,y)=p(x,y)Γ(α)Γ(β)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1], |
Clearly, k(x.y) is a a positive, integrable and symmetric about a+b2 and c+d2. If we apply Theorem 5 for the function k(x,y) then we obtain,
f(a+b2,c+d2)Hp(α,β)≤1Γ(α)Γ(β)b∫ad∫cf(x,y)p(x,y)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4cosh[ρ1(b−a)2]cosh[ρ2(d−c)2]Hp(α,β). |
From the definition of the double fractional integrals we have
1Γ(α)Γ(β)b∫ad∫cf(x,y)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]p(x,y)dydx=[Jα,βa+,c+(fp)(b,d)+Jα,βa+,d−(fp)(b,c)+Jα,βb−,c+(fp)(a,d)+Jα,βb−,d−(fp)(a,c)]. |
This completes the proof.
Remark 8. If we choose ρ1=ρ2=0 in Theorem 3, then we have the following fractional Hermite-Hadamard inequality,
f(a+b2,c+d2)[Jα,βa+,c+p(b,d)+Jα,βa+,d−p(b,c)+Jα,βb−,c+p(a,d)+Jα,βb−,d−p(a,c)]≤[Jα,βa+,c+(fp)(b,d)+Jα,βa+,d−(fp)(b,c)+Jα,βb−,c+(fp)(a,d)+Jα,βb−,d−(fp)(a,c)]≤f(a,c)+f(a,d)+f(b,c)+f(b,d)4[Jα,βa+,c+p(b,d)+Jα,βa+,d−p(b,c)+Jα,βb−,c+p(a,d)+Jα,βb−,d−p(a,c)] |
which is proved by Yaldız et all in [34].
Remark 9. If we choose α =β=1 in Theorem 3, then we have Theorem 1.3 reduces to Theorem 5.
Theorem 9. If f:Δ→R is a co-ordinated hyperbolic ρ-convex functions on Δ. Then we have the following Hermite-Hadamard type inequalities for fractional integrals,
f(a+b2,c+d2)H1(α,β)≤12[(Jαa+f(b,c+d2)+Jαb−f(a,c+d2))H2(β)+Jβc+f(d,a+b2)+Jβd−f(c,a+b2)H3(α)]≤[Jα,βa+,c+f(b,d)+Jα,βa+,d−f(b,c)+Jα,βb−,c+f(a,d)+Jα,βb−,d−f(a,c)]≤14[sech[ρ2(d−c)2](Jαa+f(b,c)+Jαa+f(b,d)+Jαb−f(a,c)+Jαb−f(a,d))H2(β)+sech[ρ1(b−a)2](Jβc+f(a,d)+Jβc+f(b,d)+Jβd−f(a,c)+Jβd−f(b,c))H3(α)]≤f(a,c)+f(b,c)+f(a,d)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]H1(α,β) | (3.1) |
where
H1(α,β)=1Γ(α)Γ(β)b∫ad∫ccosh[ρ1(x−a+b2)]cosh[ρ2(y−c+d2)]×[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx, |
H2(β)=1Γ(β)d∫ccosh[ρ2(y−c+d2)][(d−y)β−1+(y−c)β−1]dy |
and
H3(α,β)=1Γ(α)b∫acosh[ρ1(x−a+b2)][(b−x)α−1+(x−a)α−1]dx. |
Proof. If we apply Theorem 6 for the symmetric function
p(x,y)=1Γ(α)Γ(β)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1], |
then we get the following inequality
f(a+b2,c+d2)H1(α,β)≤12[(1Γ(α)b∫af(x,c+d2)[(b−x)α−1+(x−a)α−1]dx)H2(β)+(1Γ(β)d∫cf(a+b2,y)[(d−y)β−1+(y−c)β−1]dy)H3(α)]≤1Γ(α)Γ(β)b∫ad∫cf(x,y)[(b−x)α−1(d−y)β−1+(b−x)α−1(y−c)β−1+(x−a)α−1(d−y)β−1+(x−a)α−1(y−c)β−1]dydx≤14[sech[ρ2(d−c)2](1Γ(α)b∫a[f(x,c)+f(x,d)][(b−x)α−1+(x−a)α−1]dx)H2(β)+sech[ρ1(b−a)2](1Γ(β)b∫a[f(a,y)+f(b,y)][(d−y)β−1+(y−c)β−1]dx)H3(α)]≤f(a,c)+f(b,c)+f(a,d)+f(b,d)4sech[ρ1(b−a)2]sech[ρ2(d−c)2]H1(α,β). |
This completes the proof.
Remark 10. Under assumptions of Theorem 9 with α=β=1, the inequalities (3.1) reduce to inequalities (2.5) proved by Özçelik et. al in [23].
Remark 11. Under assumptions of Theorem 9 with ρ1=ρ2=0, the inequalities (3.1) reduce to inequalities proved by Sarikaya in [29,Theorem 4]
In this paper, we establish some Fejer type inequalities for co-ordinated hyperbolic ρ-convex functions. By using these inequalities we present some inequalities for Riemann-Liouville fractional integrals. In the future works, authors can prove similar inequalities for other fractional integrals.
All authors declare no conflicts of interest.
[1] |
D. Donoho, Compressed sensing, IEEE Trans. Inf. Theory, 52 (2006), 1289–1306. https://doi.org/10.1109/TIT.2006.871582 doi: 10.1109/TIT.2006.871582
![]() |
[2] |
M. Lustig, D. Donoho, J. M. Pauly, Sparse MRI: The application of compressed sensing for rapid MR imaging, Magn. Reson. Med., 58 (2007), 1182–1195. https://doi.org/10.1002/mrm.21391 doi: 10.1002/mrm.21391
![]() |
[3] |
I. Daubechies, M. Defrise, C. D. Mol, An iterative thresholding algorithm for linear inverse problems with a sparsity constraint, Commun. Pure Appl. Math., 57 (2004), 1413–1457. https://doi.org/10.1002/cpa.20042 doi: 10.1002/cpa.20042
![]() |
[4] |
T. Goldstein, S. Osher, The Split Bregman method for L1-egularized problems, SIAM J. Imaging Sci., 2 (2009), 1–21. https://doi.org/10.1137/080725891 doi: 10.1137/080725891
![]() |
[5] |
W. W. Hager, H. Zhang, A survey of nonlinear conjugate gradient methods, Pac. J. Optim., 2 (2006), 35–58. https://doi.org/10.1006/jsco.1995.1040 doi: 10.1006/jsco.1995.1040
![]() |
[6] |
H. Nien, J. A. Fessler, A convergence proof of the split Bregman method for regularized least-squares problems, Mathematics, 2014 (2014). https://doi.org/10.48550/arXiv.1402.4371 doi: 10.48550/arXiv.1402.4371
![]() |
[7] |
J. D. Benamou, G. Carlier, M. Cuturi, L. Nenna, G. Peyré, Iterative Bregman projections for regularized transportation problems, SIAM J. Sci. Comput., 37 (2015). https://doi.org/10.1137/141000439 doi: 10.1137/141000439
![]() |
[8] |
E. G. Birgin, J. M. Martínez, A spectral conjugate gradient method for unconstrained optimization, Appl. Math. Optim., 43 (2001), 117–128. https://doi.org/10.1007/s00245-001-0003-0 doi: 10.1007/s00245-001-0003-0
![]() |
[9] |
M. M. Dehnavi, D. M. Fernandez, D. Giannacopoulos, Enhancing the performance of conjugate gradient solvers on graphic processing units, IEEE Trans. Magn., 47 (2011), 1162–1165. https://doi.org/10.1109/TMAG.2010.2081662 doi: 10.1109/TMAG.2010.2081662
![]() |
[10] | S. Wang, Z. Su, L. Ying, X. Peng, S. Zhu, F. Liang, et al., Accelerating magnetic resonance imaging via deep learning, in 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI), (2016), 514–517. https://doi.org/10.1109/ISBI.2016.7493320 |
[11] |
D. Liang, J. Cheng, Z. Ke, L. Ying, Deep magnetic resonance image reconstruction: Inverse problems meet neural networks, IEEE Signal Process. Mag., 37 (2020), 141–151. https://doi.org/10.1109/MSP.2019.2950557 doi: 10.1109/MSP.2019.2950557
![]() |
[12] |
J. M. Bioucas-Dias, M. A. T. Figueiredo, A new twIst: Two-step iterative shrinkage/thresholding algorithms for image restoration, IEEE Trans. Image Process., 16 (2007), 2992–3004. https://doi.org/10.1109/tip.2007.909319 doi: 10.1109/tip.2007.909319
![]() |
[13] | A. Beck, M. Teboulle, A fast Iterative Shrinkage-Thresholding Algorithm with application to wavelet-based image deblurring, in 2009 IEEE International Conference on Acoustics, Speech and Signal Processing, (2009), 693–696. https://doi.org/10.1109/ICASSP.2009.4959678 |
[14] |
Y. Zhang, Z. Dong, P. Phillips, S. Wang, G. Ji, J. Yang, Exponential Wavelet Iterative Shrinkage Thresholding Algorithm for compressed sensing magnetic resonance imaging, Inf. Sci., 322 (2015), 115–132. https://doi.org/10.1016/j.ins.2015.06.017 doi: 10.1016/j.ins.2015.06.017
![]() |
[15] | X. Li, J. Wang, S. Tan, Hessian Schatten-norm regularization for CBCT image reconstruction using fast iterative shrinkage-thresholding algorithm, in Medical Imaging 2015: Physics of Medical Imaging, 2015. https://doi.org/10.1117/12.2082424 |
[16] |
G. Wu, S. Luo, Adaptive fixed-point iterative shrinkage/thresholding algorithm for MR imaging reconstruction using compressed sensing, Magn. Reson. Imaging, 32 (2014), 372–378. https://doi.org/10.1016/j.mri.2013.12.009 doi: 10.1016/j.mri.2013.12.009
![]() |
[17] |
K. Shang, Y. Li, Z. Huang, Iterative p-shrinkage thresholding algorithm for low Tucker rank tensor recovery, Inf. Sci., 482 (2019), 374–391. https://doi.org/10.1016/j.ins.2019.01.031 doi: 10.1016/j.ins.2019.01.031
![]() |
[18] | L. Zhang, H. Wang, Y. Xu, A shrinkage-thresholding method for the inverse problem of Electrical Resistance Tomography, in 2012 IEEE International Instrumentation and Measurement Technology Conference Proceedings, (2012), 2425–2429. https://doi.org/10.1109/I2MTC.2012.6229564 |
[19] |
A. Beck, M. Teboulle, A fast iterative shrinkage-thresholding algorithm for linear inverse problems, SIAM J. Imaging Sci., 2 (2009), 183–202. https://doi.org/10.1137/080716542 doi: 10.1137/080716542
![]() |
[20] |
A. Chambolle, C. Dossal, On the Convergence of the Iterates of the "Fast Iterative Shrinkage/Thresholding Algorithm", J. Optim. Theory Appl., 166 (2015), 1–15. https://doi.org/10.1007/s10957-015-0746-4 doi: 10.1007/s10957-015-0746-4
![]() |
[21] |
İ. Bayram, On the convergence of the iterative shrinkage/thresholding algorithm with a weakly convex penalty, IEEE Trans. Signal Process., 64 (2016), 1597–1608. https://doi.org/10.1109/TSP.2015.2502551 doi: 10.1109/TSP.2015.2502551
![]() |
[22] |
W. Hao, J. Li, X. Qu, Z. Dong, Fast iterative contourlet thresholding for compressed sensing MRI, Electron. Lett., 49 (2013), 1206. https://doi.org/10.1049/el.2013.1483 doi: 10.1049/el.2013.1483
![]() |
[23] |
S. Dirksen, G. Lecue, H. Rauhut, On the gap between restricted isometry properties and sparse recovery conditions, IEEE Trans. Inf. Theory, 64 (2018), 5478–5487. https://doi.org/10.1109/TIT.2016.2570244 doi: 10.1109/TIT.2016.2570244
![]() |
[24] |
Y. Yang, C. M. Kramer, P. W. Shaw, C. H. Meyer, M. Salerno, First-pass myocardial perfusion imaging with whole-heart coverage using L1-SPIRiT accelerated variable density spiral trajectories, Magn. Reson. Med., 76 (2016), 1375–1387. https://doi.org/10.1002/mrm.26014 doi: 10.1002/mrm.26014
![]() |
[25] |
V. P. Gopi, P. Palanisamy, K. A. Wahid, P. Babyn, D. Cooper, Multiple regularization based MRI reconstruction, Signal Process., 103 (2014), 103–113. https://doi.org/10.1016/j.sigpro.2013.11.001 doi: 10.1016/j.sigpro.2013.11.001
![]() |
[26] |
C. S. Xydeas, V. S. Petrovic, Objective image fusion performance measure, Electron. Lett., 36 (2000), 308–309. https://doi.org/10.1117/12.381668 doi: 10.1117/12.381668
![]() |
1. | Dumitru Baleanu, Artion Kashuri, Pshtiwan Othman Mohammed, Badreddine Meftah, General Raina fractional integral inequalities on coordinates of convex functions, 2021, 2021, 1687-1847, 10.1186/s13662-021-03241-y | |
2. | Han Li, Muhammad Shoaib Saleem, Imran Ahmed, Kiran Naseem Aslam, Hermite–Hadamard and Fejér-type inequalities for strongly reciprocally (p, h)-convex functions of higher order, 2023, 2023, 1029-242X, 10.1186/s13660-023-02960-y | |
3. | Silvestru Sever Dragomir, Mohamed Jleli, Bessem Samet, Hermite-Hadamard-type inequalities for generalized trigonometrically and hyperbolic ρ-convex functions in two dimension, 2024, 22, 2391-5455, 10.1515/math-2024-0028 |