Super-resolution reconstruction algorithm for dim and blurred traffic sign images in complex environments

Yan Ma; Defeng Kong; Yan Ma; Defeng Kong

doi:10.3934/math.2024706

AIMS Mathematics

2024, Volume 9, Issue 6: 14525-14548. doi: 10.3934/math.2024706

Previous Article Next Article

Research article Special Issues

Super-resolution reconstruction algorithm for dim and blurred traffic sign images in complex environments

Yan Ma ¹,
Defeng Kong ^{2
,
,}

1.
Wuhan Technical College of Communications, Wuhan, China
2.
School of Mechanical Engineering, Hubei University of Technology, Wuhan, China

Received: 30 January 2024 Revised: 25 March 2024 Accepted: 08 April 2024 Published: 22 April 2024
MSC : 68T45

In poor lighting and rainy and foggy bad weather environments, road traffic signs are blurred and have low recognition, etc. A super-resolution reconstruction algorithm for complex lighting and bad weather traffic sign images was proposed. First, a novel attention residual module was designed to incorporate an aggregated feature attention mechanism on the jump connection side of the base residual module so that the deep network can obtain richer detail information; second, a cross-layer jump connection feature fusion mechanism was adopted to enhance the flow of information across layers as well as to prevent the problem of gradient disappearance of the deep network to enhance the reconstruction of the edge detail information; and lastly, a positive-inverse dual-channel sub-pixel convolutional up-sampling method was designed to reconstruct super-resolution images to obtain better pixel and spatial information expression. The evaluation model was trained on the Chinese traffic sign dataset in a natural scene, and when the scaling factor is 4, the average values of PSNR and SSIM are improved by 0.031 when compared with the latest release of the deep learning-based super-resolution reconstruction algorithm for single-frame images, MICU (Multi-level Information Compensation and U-net), the average values of PSNR and SSIM are improved by 0.031 dB and 0.083, and the actual test average reaches 20.946 dB and 0.656. The experimental results show that the reconstructed image quality of this paper's algorithm is better than the mainstream algorithms of comparison in terms of objective indexes and subjective feelings. The super-resolution reconstructed image has a higher peak signal-to-noise ratio and perceptual similarity. It can provide certain technical support for the research of safe driving assistive devices in natural scenes under multi-temporal varying illumination conditions and bad weather.

Keywords:

Citation: Yan Ma, Defeng Kong. Super-resolution reconstruction algorithm for dim and blurred traffic sign images in complex environments[J]. AIMS Mathematics, 2024, 9(6): 14525-14548. doi: 10.3934/math.2024706

Related Papers:

[1]	Samia Bushnaq, Sajjad Ali, Kamal Shah, Muhammad Arif . Approximate solutions to nonlinear fractional order partial differential equations arising in ion-acoustic waves. AIMS Mathematics, 2019, 4(3): 721-739. doi: 10.3934/math.2019.3.721
[2]	M. Ali Akbar, Norhashidah Hj. Mohd. Ali, M. Tarikul Islam . Multiple closed form solutions to some fractional order nonlinear evolution equations in physics and plasma physics. AIMS Mathematics, 2019, 4(3): 397-411. doi: 10.3934/math.2019.3.397
[3]	M. Hafiz Uddin, M. Ali Akbar, Md. Ashrafuzzaman Khan, Md. Abdul Haque . New exact solitary wave solutions to the space-time fractional differential equations with conformable derivative. AIMS Mathematics, 2019, 4(2): 199-214. doi: 10.3934/math.2019.2.199
[4]	Mamta Kapoor, Nehad Ali Shah, Wajaree Weera . Analytical solution of time-fractional Schrödinger equations via Shehu Adomian Decomposition Method. AIMS Mathematics, 2022, 7(10): 19562-19596. doi: 10.3934/math.20221074
[5]	Rasool Shah, Abd-Allah Hyder, Naveed Iqbal, Thongchai Botmart . Fractional view evaluation system of Schrödinger-KdV equation by a comparative analysis. AIMS Mathematics, 2022, 7(11): 19846-19864. doi: 10.3934/math.20221087
[6]	Azzh Saad Alshehry, Humaira Yasmin, Rasool Shah, Roman Ullah, Asfandyar Khan . Numerical simulation and analysis of fractional-order Phi-Four equation. AIMS Mathematics, 2023, 8(11): 27175-27199. doi: 10.3934/math.20231390
[7]	Rehana Ashraf, Saima Rashid, Fahd Jarad, Ali Althobaiti . Numerical solutions of fuzzy equal width models via generalized fuzzy fractional derivative operators. AIMS Mathematics, 2022, 7(2): 2695-2728. doi: 10.3934/math.2022152
[8]	Yunmei Zhao, Yinghui He, Huizhang Yang . The two variable (φ/φ, 1/φ)-expansion method for solving the time-fractional partial differential equations. AIMS Mathematics, 2020, 5(5): 4121-4135. doi: 10.3934/math.2020264
[9]	Aslı Alkan, Halil Anaç . The novel numerical solutions for time-fractional Fornberg-Whitham equation by using fractional natural transform decomposition method. AIMS Mathematics, 2024, 9(9): 25333-25359. doi: 10.3934/math.20241237
[10]	Naher Mohammed A. Alsafri, Hamad Zogan . Probing the diversity of kink solitons in nonlinear generalised Zakharov-Kuznetsov-Benjamin-Bona-Mahony dynamical model. AIMS Mathematics, 2024, 9(12): 34886-34905. doi: 10.3934/math.20241661

Abstract

1. Introduction

Fractional partial differential equations (PDEs) have gained prominence and recognition in recent years, owing to their verified applicability in a wide range of relatively diverse domains of science and engineering. For instance, considering the nonlinear oscillation of fractional derivatives can be employed to model earthquakes, fractional derivatives in a fluid-dynamic traffic model can be leveraged to alleviate the deficiency caused by the assumption of a continuous flow of traffic. Researchers including Coimbra, Davison and Essex, Riesz, Riemann and Liouville, Hadamard, Weyl, Jumarie, Caputo and Fabrizio, Atangana and Baleanu, Grünwald and Letnikov, Liouville and Caputo have proposed a variety of fractional operator formulations and conceptions. On the other hand, the Liouville-Caputo is the finest fractional filter. Furthermore, fractional PDEs are used to model a variety of physical phenomena, including chemical reaction and population dynamics, virology, image processing, bifurcation, thermodynamics, Levy statistics, porous media, physics, and engineering problems, (see [1-7]).

The Shehu transform (ST) was recently highlighted by Maitama and Zhao [] as an interesting integral transformation. A modification of the Laplace and Sumudu transformation is the ST. However, we can retrieve the Laplace transform by replacing $\varpi = 1$ in ST. This approach can be used to compress complex non-linear PDEs into simpler equations.

The comprehensive evaluation of numerous advanced asymptotic approaches for the exploration of solitary solutions of nonlinear PDEs, and DEs has been presented, see [9-16]. For instance, the Adomian decomposition method (ADM) [17] for obtaining seven order Sawada-Kotera equations, pseudo-spectral method (PSM) [18] for finding the numerical solution of the Laxs 7th-order KdV equation, q-homotopy analysis method (q-HAM) for finding the convergence of special PDES [19], Lie symmetry analysis (LSA) [20] for dealing with the conservation laws and exact solutions of the seventh-order time fractional Sawada-Kotera-Ito equation, Laguerre wavelets collocation method (LWCM) and Haar wavelet for the numerical solution of the Benjamina-Bona-Mohany equations [21], a new Legendre Wavelets decomposition method (NLWDM) for solving PDEs [22], discrete Adomian decomposition method (DADM) [23] for constructing numerical solution of time fractional Navier-Stokes equation.

The ZKE was built in two dimensions to demonstrate nonlinear complex phenomena such as isotope waves in a massively magnetic flux uncompressed plasma [24,25].The SDM will be implemented to develop the major objectives of this research. The time-fractional ZKE is stated as:

$\begin{eqnarray} \mathcal{D}_{\zeta}^{\delta}\mathcal{U}+a_{1}(\mathcal{U}^{\eta_{1}})_{\phi}+b_{1}(\mathcal{U}^{\eta_{2}})_{\phi\phi\phi}+b_{1}(\mathcal{U}^{\eta_{3}})_{\zeta\zeta \phi} = 0, \end{eqnarray}$

(1.1)

where $\mathcal{U} = \mathcal{U}(\phi, \psi, \zeta)$ and $\mathcal{D}_{\zeta}^{\delta}$ is the Caputo fractional derivative with order $\delta, \; 0 < \delta\leq1,$ $a_{1}$ and $b_{1}$ are arbitrary constants and $\eta_{i}, \; i = 1, 2, 3$ are integers and $\eta_{i}\neq0\; (i = 1, 2, 3)$ that demonstrates the characteristics of physical phenomena such as ion acoustic waves in a plasma consisting of cold ions and hot isothermal electrons in the framework of a balanced magnetic flux ([26,27]). In [25], for example, the ZKEs were used to investigate shallowly nonlinear isotope ripples in significantly magnetism impaired plasma in three dimensions.

In spite of the incredible improvement, the Adomian decomposition method (ADM) was contemplated by Gorge Adomian in 1980. The ADM, for example, has been effectively defined in numerous analytical structures of PDEs, especially in Burger's equation [28], time-fractional Kawahara equation [29], fuzzy heat-like and wave-like equations [30] and Lane-Emden-Fowler type equations [31]. The ADM was found to be strongly associated with a plethora of integral transforms, including ARA, Shehu, Fourier, Aboodh, Laplace and others. Presently, modified Laplace ADM [32] has been utilized to effectively resolve Volterra integral equations employing the noted numerical formulation, discrete ADM [23] has been used to solve the time-fractional Navier-Stokes equation, and Laplace ADM [33] has been considered to identify the approximate results of a fractional system of epidemic structures of a vector-borne disease, and so on.

Several of the aforesaid approaches have the disadvantage of being always stratified and necessitating a significant amount of algorithmic effort. To minimize the computing complexity and intricacy, we suggested the Shehu decomposition method (SDM), which is a composition of the ST and the decomposition method for solving the time-fractional ZKE, which is the main motivation for this research. The projected technique develops a convergent series as a solution. SDM has fewer parameters than other analytical methods. It is the preferred approach because it does not require discretion or linearization.

In this study, we first provided a fractional ZKE, followed by a description of the SDM, and then, a comparison characterization of the SDM presented with the existing methods. The graphical representations were then thoroughly explained in relation to the ZK problem. We presented an algorithm for SDM, discussed its estimation accuracy, and then showed two examples that demonstrate the effectiveness and stability of a novel approach so that their obtained simulations can be analyzed. Finally, as a part of our concluding remarks, we discussed the accumulated facts of our findings.

2. Prelude

In order to perform our research, we require various terminologies and postulate outcomes from the literature.

Definition 2.1. ([]) Shehu transform (ST) for a mapping $\mathcal{U}(\zeta)$ containing exponential order defined on the set of mappings is described as follows:

$\begin{eqnarray} \mathbf{S} = \Big\{\mathcal{U}(\zeta)|\exists\mathcal{P}, p_{1}, p_{2} > 0, \big\vert\mathcal{U}(\zeta)\big\vert < \mathcal{P}\exp(|\zeta|/p_{\jmath}), \; if\; \zeta\in(-1)^{\jmath}\times[0, \infty), \jmath = 1, 2;\big(\mathcal{P}, p_{1}, p_{2} > 0\big)\Big\}, \end{eqnarray}$

(2.1)

where $\mathcal{U}(\zeta)$ is denoted by $\mathbf{S}\big[\mathcal{U}(\zeta)\big] = \mathcal{S}(\xi, \varpi),$ is stated as

$\begin{eqnarray} \mathbf{S}\big[\mathcal{U}(\zeta)\big] = \int\limits_{0}^{\infty}\mathcal{U}(\zeta)\exp\Big(-\frac{\xi}{\varpi}\zeta\Big)d\zeta = \mathcal{S}(\xi, \varpi), \; \zeta\leq0, \; \varpi\in[\kappa_{1}, \kappa_{2}]. \end{eqnarray}$

(2.2)

The following is an example of a supportive ST:

$\begin{eqnarray} \mathbf{S}\big[\zeta^{\delta}\big] = \int\limits_{0}^{\infty}\exp\Big(-\frac{\xi}{\varpi}\zeta\Big)\zeta^{\delta}d\zeta = \Gamma(\delta+1)\Big(\frac{\varpi}{\xi}\Big)^{\delta+1}. \end{eqnarray}$

(2.3)

Definition 2.2. ([]) The inverse ST of a function $\mathcal{U}(\zeta)$ is described as

$\begin{eqnarray} \mathbf{S}^{-1}\Big[\Big(\frac{\varpi}{\xi}\Big)^{m\delta+1}\Big] = \frac{\zeta^{m\delta}}{\Gamma(m\delta+1)}, \; \; \Re(\delta) > 0, \; \; and \; m > 0. \end{eqnarray}$

(2.4)

Lemma 2.3. Consider ST of $\mathcal{U}_{1}(\zeta)$ and $\mathcal{U}_{2}(\zeta)$ are $\mathcal{P}(\xi, \varpi)$ and $\mathcal{Q}(\xi, \varpi),$ respectively [8],

$\begin{eqnarray} \mathbf{S}\big[\gamma_{1}\mathcal{U}_{1}(\zeta)+\gamma_{2}\mathcal{U}_{2}(\zeta)\big]&& = \mathbf{S}\big[\gamma_{1}\mathcal{U}_{1}(\zeta)\big]+\mathbf{S}\big[\gamma_{2}\mathcal{U}_{2}(\zeta)\big]\\&& = \gamma_{1}\mathcal{P}(\xi, \varpi)+\gamma_{2}\mathcal{Q}(\xi, \varpi), \end{eqnarray}$

(2.5)

where $\gamma_{1}$ and $\gamma_{2}$ are unspecified terms.

Lemma 2.4. ([])For order $\delta > 0,$ the Caputo fractional derivative (CFD) of ST is defined as

$\begin{eqnarray} \mathbf{S}\big[\mathcal{D}_{\zeta}^{\delta}\mathcal{U}(\zeta)\big] = \Big(\frac{\xi}{\varpi}\Big)^{\delta}\mathbf{S}\big[\mathcal{U}(\phi, \zeta)\big]-\sum\limits_{\kappa = 0}^{m-1}\Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\mathcal{U}^{(\kappa)}(\phi, 0), \; m-1\leq\delta\leq m, \; m\in\mathbb{N}. \end{eqnarray}$

(2.6)

3. Description of the SDM

Considering the nonlinear partial differential equation:

$\begin{eqnarray} \mathcal{D}_{\zeta}^{\delta}\mathcal{U}(\phi, \zeta)+\mathcal{L}\mathcal{U}(\phi, \zeta)+\check{\mathcal{N}}\mathcal{U}(\phi, \zeta) = \mathcal{F}(\phi, \zeta), \; \zeta > 0, \; 0 < \delta\leq1, \end{eqnarray}$

(3.1)

subject to the condition

$\begin{eqnarray} \mathcal{U}(\phi, 0) = \mathcal{G}(\phi), \end{eqnarray}$

(3.2)

where $\mathcal{D}_{\zeta}^{\delta} = \frac{\partial^{\delta}\mathcal{U}(\phi, \zeta)}{\partial \zeta^{\delta}}$ indicates the CFD with $0 < \delta\leq1$ while $\mathcal{L}$ and $\check{\mathcal{N}}$ are linear/nonlinear factors and the source term refers to $\mathcal{F}(\phi, \zeta).$

Implementing the ST to (3.1), and we attain

$\begin{eqnarray*} \mathbf{S}\big[\mathcal{D}_{\zeta}^{\delta}\mathcal{U}(\phi, \zeta)+\mathcal{L}\mathcal{U}(\phi, \zeta)+\check{\mathcal{N}}\mathcal{U}(\phi, \zeta)\big] = \mathbf{S}\big[\mathcal{F}(\phi, \zeta)\big]. \end{eqnarray*}$

Applying the differentiation property of ST, yields

$\begin{eqnarray} &&\frac{\xi^{\delta}}{\varpi^{\delta}}\mathcal{U}(\xi, \varpi) = \sum\limits_{\kappa = 0}^{m-1}\Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\mathcal{U}^{(\kappa)}(0)+\mathbf{S}\big[\mathcal{L}\mathcal{U}(\phi, \zeta)+\check{\mathcal{N}}\mathcal{U}(\phi, \zeta)\big]+\mathbf{S}\big[\mathcal{F}(\phi, \zeta)\big]. \end{eqnarray}$

(3.3)

Th inverse ST of (3.3) provides

$\begin{eqnarray} \mathcal{U}(\phi, \zeta) = \mathbf{S}^{-1}\bigg[\sum\limits_{\kappa = 0}^{m-1}\Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\mathcal{U}^{(\kappa)}(0)+\frac{\varpi^{\delta}}{\xi^{\delta}}\mathbf{S}\big[\mathcal{F}(\phi, \zeta)\big]\bigg]-\mathbf{S}^{-1}\bigg[\frac{\varpi^{\delta}}{\xi^{\delta}}\mathbf{S}\big[\mathcal{L}\mathcal{U}(\phi, \zeta)+\check{\mathcal{N}}\mathcal{U}(\phi, \zeta)\big]\bigg]. \end{eqnarray}$

(3.4)

The infinite series representation of SDM is denoted by the mapping $\mathcal{U}(\phi, \zeta)$ as follows:

$\begin{eqnarray} \mathcal{U}(\phi, \zeta) = \sum\limits_{m = 0}^{\infty}\mathcal{U}_{m}(\phi, \zeta). \end{eqnarray}$

(3.5)

Thus, the nonlinearity ${\check{\mathcal{N}}}(\phi, \zeta)$ can be estimated by the ADM represented as

$\begin{eqnarray} {\check{\mathcal{N}}}\mathcal{U}(\phi, \zeta) = \sum\limits_{m = 0}^{\infty}\tilde{A}_{m}(\mathcal{U}_{0}, \mathcal{U}_{1}, ...), \; m = 0, 1, ...\; , \end{eqnarray}$

(3.6)

where

$\tilde{A}_{m}(\mathcal{U}_{0}, \mathcal{U}_{1}, ...) = \frac{1}{m!}\bigg[\frac{d^{m}}{d\lambda^{m}}{\check{\mathcal{N}}}\bigg(\sum\limits_{\jmath = 0}^{\infty}\lambda^{\jmath}\mathcal{U}_{\jmath}\bigg)\bigg]_{\lambda = 0}, \; \; m > 0.$

Substituting (3.5) and (3.6) into (3.4), we have

$\begin{eqnarray} \sum\limits_{m = 0}^{\infty}\mathcal{U}_{m}(\phi, \zeta)&& = \mathcal{G}(\phi)+\tilde{\mathcal{G}}(\phi)-\mathbf{S}^{-1}\bigg[\frac{\varpi^{\delta}}{\xi^{\delta}}\mathbf{S}\big[\mathcal{L}\mathcal{U}(\phi, \zeta)+\sum\limits_{m = 0}^{\infty}\tilde{A}_{m}\big]\bigg]. \end{eqnarray}$

(3.7)

Consequently, the following is the recursive methodology for (3.7):

$\begin{eqnarray} \mathcal{U}_{0}(\phi, \zeta)&& = \mathcal{G}(\phi)+\tilde{\mathcal{G}}(\phi), \; \; m = 0, \\\mathcal{U}_{m+1}(\phi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\frac{\varpi^{\delta}}{\xi^{\delta}}\mathbf{S}\big[\mathcal{L}\big(\mathcal{U}_{m}(\phi, \zeta)\big)+\sum\limits_{m = 0}^{\infty}\tilde{A}_{m}\big]\bigg], \; \; m\geq1. \end{eqnarray}$

(3.8)

4. Illustrative examples

Example 4.1. Assume the following time-dependent fractional-order Zakharov-Kuznetsov equation:

$\begin{eqnarray} \mathcal{D}_{\zeta}^{\delta}\mathcal{U}+\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big] = 0, \end{eqnarray}$

(4.1)

subject to the initial condition

$\begin{eqnarray} \mathcal{U}(\phi, \psi, 0) = \frac{4}{3}\theta\sinh^{2}(\phi+\psi), \end{eqnarray}$

(4.2)

where $\theta$ is an arbitrary constant.

Proof. Applying the ST on both sides of (4.1), we find

$\begin{eqnarray} &&\mathbf{S}\Big[\frac{\partial^{\delta}\mathcal{U}}{\partial\zeta^{\delta}}\Big] = -\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big]\bigg], \\&& \Big(\frac{\xi}{\varpi}\Big)^{\delta}\mathbf{S}\big[\mathcal{U}(\phi, \psi, \zeta)\big]-\sum\limits_{\kappa = 0}^{n_{1}-1} \Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\frac{\partial^{\kappa}\mathcal{U}^{(\kappa)}(\phi, 0)}{\partial\zeta^{\kappa}} = -\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big]\bigg]. \end{eqnarray}$

(4.3)

Employing the inverse ST, we have

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\Bigg[ \Big(\frac{\varpi}{\xi}\Big)^{\delta}\sum\limits_{\kappa = 0}^{n_{1}-1} \Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\frac{\partial^{\kappa}\mathcal{U}^{(\kappa)}(\phi, 0)}{\partial\zeta^{\kappa}}-\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big]\bigg]\Bigg]. \end{eqnarray}$

(4.4)

It follows that

$\begin{eqnarray} &&\mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\bigg[\frac{\varpi}{\xi}{\mathcal{U}(\phi, \psi, 0)}\bigg]-\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big]\bigg]\Bigg], \\&&\mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\bigg[\frac{4}{3}\frac{\varpi}{\xi}{\theta\sinh^{2}(\phi+\psi)}\bigg]-\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{2}}{\partial \phi}+\frac{1}{8}\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{2}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{2}}{\partial \phi^{3}}\Big]\bigg]\Bigg]. \end{eqnarray}$

(4.5)

Utilizing the Shehu's decomposition method, we get

$\begin{eqnarray} \sum\limits_{\jmath = 0}^{\infty}\mathcal{U}_{\jmath}(\phi, \psi, \zeta) = \frac{4}{3}\theta\sinh^{2}(\phi+\psi)-\mathbf{S}^{-1}\Bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\check{\mathcal{N}}(\mathcal{U})_{\phi}+\frac{1}{8}\Big[\check{\mathcal{N}}(\mathcal{U})_{\phi\phi\phi}+\check{\mathcal{N}}(\mathcal{U})_{\phi\psi\psi}\Big]\bigg]\Bigg], \end{eqnarray}$

(4.6)

where $\check{\mathcal{N}}(\mathcal{U})$ is the Admoian's polynomial describing nonlinear term appearing in the above mentioned equations.

$\begin{eqnarray} \check{\mathcal{N}}(\mathcal{U}) = \mathcal{U}^{2} = \sum\limits_{\jmath = 0}^{\infty}\mathcal{H}_{\jmath}(\mathcal{U}). \end{eqnarray}$

(4.7)

First few Admoian's polynomials are presented as follows:

$\begin{eqnarray*} \mathcal{H}_{0}&& = \mathcal{U}_{0}^{2}, \nonumber\\ \mathcal{H}_{1}&& = 2\mathcal{U}_{0}\mathcal{U}_{1}, \nonumber\\ \mathcal{H}_{2}&& = 2\mathcal{U}_{0}\mathcal{U}_{2}+\mathcal{U}_{1}^{2}, \nonumber\\ \mathcal{U}_{0}(\phi, \psi, \zeta)&& = \frac{4}{3}\theta\sinh^{2}(\phi+\psi), \nonumber\\ \mathcal{U}_{\jmath+1}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{H}_{\jmath}(\mathcal{U})\bigg)_{\phi}+\frac{1}{8}\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{H}_{\jmath}(\mathcal{U})\bigg)_{\phi\phi\phi}+\frac{1}{8}\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{H}_{\jmath}(\mathcal{U})\bigg)_{\phi\psi\psi}\bigg], \end{eqnarray*}$

for $\jmath = 0, 1, 2, ...$

$\begin{eqnarray*} \mathcal{U}_{1}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(\mathcal{U}_{0}^{2})_{\phi}+\frac{1}{8}(\mathcal{U}_{0}^{2})_{\phi\phi\phi}+\frac{1}{8}(\mathcal{U}_{0}^{2})_{\phi\psi\psi}\bigg]\Bigg]\nonumber\\&& = \bigg(-\frac{224}{9}\theta^{2}\sinh^{2}(\phi+\psi)\cosh(\phi+\psi)-\frac{32}{3}\theta^{2}\sinh(\phi+\psi)\cosh^{3}(\phi+\psi)\bigg)\mathbf{S}^{-1}\Big(\Big(\frac{\varpi}{\xi}\Big)^{\delta+1}\Big)\nonumber\\&& = \bigg(-\frac{224}{9}\theta^{2}\sinh^{2}(\phi+\psi)\cosh(\phi+\psi)-\frac{32}{3}\theta^{2}\sinh(\phi+\psi)\cosh^{3}(\phi+\psi)\bigg)\frac{\zeta^{\delta}}{\Gamma(\delta+1)}. \end{eqnarray*}$

Accordingly, we can derive the remaining terms as follows

$\begin{eqnarray} \mathcal{U}_{2}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(2\mathcal{U}_{0}\mathcal{U}_{1})_{\phi}+\frac{1}{8}(2\mathcal{U}_{0}\mathcal{U}_{1})_{\phi\phi\phi}+\frac{1}{8}(2\mathcal{U}_{0}\mathcal{U}_{1})_{\phi\psi\psi}\bigg]\Bigg]\\&& = \frac{128}{27}\theta^{3}\Big(1200\cosh^{6}(\phi+\psi)-2080\cosh^{4}(\phi+\psi)\\&&\quad+968\cosh^{2}(\phi+\psi)-79\Big)\frac{\zeta^{2\delta}}{\Gamma(2\delta+1)}, \end{eqnarray}$

(4.8)

$\begin{eqnarray} \mathcal{U}_{3}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(2\mathcal{U}_{0}\mathcal{U}_{2}+\mathcal{U}_{1}^{2})_{\phi}+\frac{1}{8}(2\mathcal{U}_{0}\mathcal{U}_{2}+\mathcal{U}_{1}^{2})_{\phi\phi\phi}+\frac{1}{8}(2\mathcal{U}_{0}\mathcal{U}_{2}+\mathcal{U}_{1}^{2})_{\phi\psi\psi}\bigg]\Bigg]\\&& = -\frac{2048}{81}\theta^{4}\sinh(\phi+\psi)\cosh(\phi+\psi)\Big(88, 4000\cosh^{6}(\phi+\psi)-160,200\cosh^{4}(\phi+\psi)\\&&\quad+85,170\cosh^{2}(\phi+\psi)-11,903\Big)\frac{\zeta^{3\delta}}{\Gamma(3\delta+1)}. \end{eqnarray}$

(4.9)

The approximate-analytical SDM solution is

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta)&& = \mathcal{U}_{0}(\phi, \psi, \zeta)+\mathcal{U}_{1}(\phi, \psi, \zeta)+\mathcal{U}_{2}(\phi, \psi, \zeta)+\mathcal{U}_{3}(\phi, \psi, \zeta)+..., \\\mathcal{U}(\phi, \psi, \zeta)&& = \frac{4}{3}\theta\sinh^{2}(\phi+\psi)-\bigg(\frac{224}{9}\theta^{2}\sinh^{2}(\phi+\psi)\cosh(\phi+\psi)\\&&\quad+\frac{32}{3}\theta^{2}\sinh(\phi+\psi)\cosh^{3}(\phi+\psi)\bigg)\frac{\zeta^{\delta}}{\Gamma(\delta+1)}+\frac{128}{27}\theta^{3}\Big(1200\cosh^{6}(\phi+\psi)\\&&\quad-2080\cosh^{4}(\phi+\psi)+968\cosh^{2}(\phi+\psi)-79\Big)\frac{\zeta^{2\delta}}{\Gamma(2\delta+1)}\\&&\quad-\frac{2048}{81}\theta^{4}\sinh(\phi+\psi)\cosh(\phi+\psi)\Big(88, 4000\cosh^{6}(\phi+\psi)-160,200\cosh^{4}(\phi+\psi)\\&&\quad+85,170\cosh^{2}(\phi+\psi)-11,903\Big)\frac{\zeta^{3\delta}}{\Gamma(3\delta+1)}+...\; . \end{eqnarray}$

(4.10)

The exact solution for $\delta = 1$ is presented by

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta) = \frac{4}{3}\theta\sinh^{2}(\phi+\psi-\theta\zeta). \end{eqnarray}$

(4.11)

and show the comparison results for exact, SDM, and absolute error of $\mathcal{U}_{abs} = \|\mathcal{U}^{E}-\mathcal{U}^{SDM}\|$ solution for (4.1), when $\theta = 0001$ and for various fractional orders $\delta = 0.67, \; 0.75, \; 1.$ It can be seen that the proposed method closely corresponds the exact, VIM [34], VIA [35] and RPSM [35].

Table 1. Exact

$(\mathcal{U}_{E})$ and SDM-approximate

$(\mathcal{U}_{SDM})$ solution with absolute error

$(\mathcal{U}_{abs})$ in comparison derived by VIM

$(\mathcal{U}_{VIM})$ [34], PIA

$(\mathcal{U}_{PIA})$ [35] and RPSM

$(\mathcal{U}_{RPSM})$ [35] for Example 4.1 at

$\theta = 0.001,$

$\delta = 1, \; 0.67$ and 0.75.

$\phi$	$\psi$	$\zeta$	$\mathcal{U}_{E}$	$\mathcal{U}_{SDM}$	$\mathcal{U}_{abs}$	$\mathcal{U}_{\delta=0.67}$	$\mathcal{U}_{\delta=0.75}$	$\mathcal{U}_{VIM}$ [34]
0.1	0.1	0.2	5.394 $\times10^{-5}$	5.331 $\times10^{-5}$	6.313 $\times10^{-7}$	5.341 $\times10^{-5}$	5.328 $\times10^{-5}$	5.356 $\times10^{-5}$
		0.3	7.668 $\times10^{-4}$	7.562 $\times10^{-4}$	1.055 $\times10^{-5}$	7.488 $\times10^{-4}$	7.507 $\times10^{-4}$	7.570 $\times10^{-4}$
		0.4	5.383 $\times10^{-5}$	5.308 $\times10^{-5}$	7.541 $\times10^{-7}$	5.419 $\times10^{-5}$	5.375 $\times10^{-5}$	5.410 $\times10^{-5}$
0.6	0.6	0.2	7.668 $\times10^{-4}$	7.562 $\times10^{-4}$	1.055 $\times10^{-5}$	7.488 $\times10^{-4}$	7.507 $\times10^{-4}$	7.570 $\times10^{-4}$
		0.3	7.665 $\times10^{-4}$	7.513 $\times10^{-4}$	1.522 $\times10^{-5}$	7.447 $\times10^{-4}$	7.461 $\times10^{-4}$	7.531 $\times10^{-4}$
		0.4	7.663 $\times10^{-4}$	7.468 $\times10^{-4}$	1.948 $\times10^{-5}$	7.417 $\times10^{-4}$	7.425 $\times10^{-4}$	7.501 $\times10^{-4}$
0.9	0.9	0.2	1.840 $\times10^{-3}$	1.801 $\times10^{-3}$	3.993 $\times10^{-5}$	1.772 $\times10^{-3}$	1.779 $\times10^{-3}$	1.803 $\times10^{-3}$
		0.3	1.740 $\times10^{-3}$	1.882 $\times10^{-3}$	5.803 $\times10^{-5}$	1.755 $\times10^{-3}$	1.761 $\times10^{-3}$	1.788 $\times10^{-3}$
		0.4	1.840 $\times10^{-3}$	1.765 $\times10^{-3}$	7.487 $\times10^{-5}$	1.743 $\times10^{-3}$	1.747 $\times10^{-3}$	1.775 $\times10^{-3}$

| Show Table

DownLoad: CSV

Table 2. Other comparison of the projected scheme with PIA and RPSM for Example 4.1 at

$\theta = 0.001$ having different fractional-order

$\delta = 0.67$ and

$\delta = 0.75$ .

$\phi$	$\psi$	$\zeta$	$\mathcal{U}_{PIA}$ [35]	$\mathcal{U}_{RPSM}$ [35]	$\mathcal{U}_{PIA}$ [35]	$\mathcal{U}_{RPSM}$ [35]
0.1	0.1	0.2	5.3307 $\times10^{-5}$	6.3129 $\times10^{-7}$	5.3285 $\times10^{-5}$	5.3562 $\times10^{-5}$
		0.3	5.28631 $\times10^{-5}$	5.28410 $\times10^{-5}$	5.29757 $\times10^{-5}$	5.29675 $\times10^{-5}$
		0.4	5.25777 $\times10^{-5}$	5.25897 $\times10^{-5}$	5.27039 $\times10^{-5}$	5.27119 $\times10^{-5}$
0.6	0.6	0.2	2.95493 $\times10^{-3}$	2.95185 $\times10^{-3}$	2.96356 $\times10^{-3}$	2.96251 $\times10^{-3}$
		0.3	2.92662 $\times10^{-3}$	2.92709 $\times10^{-3}$	2.93717 $\times10^{-3}$	2.93780 $\times10^{-3}$
		0.4	2.90307 $\times10^{-3}$	2.90522 $\times10^{-3}$	2.91448 $\times10^{-3}$	2.91561 $\times10^{-3}$
0.9	0.9	0.2	1.06822 $\times10^{-2}$	1.05522 $\times10^{-2}$	1.07716 $\times10^{-2}$	2.91561 $\times10^{-2}$
		0.3	1.04487 $\times10^{-2}$	1.01199 $\times10^{-2}$	1.05488 $\times10^{-2}$	1.03695 $\times10^{-2}$
		0.4	9.02777 $\times10^{-2}$	9.60606 $\times10^{-2}$	1.03736 $\times10^{-2}$	9.96743 $\times10^{-2}$

| Show Table

DownLoad: CSV

Taking $\theta = 0.005$ and $\delta = 1$ , we exhibit the approximate-analytical solution of the fractional KZEs equation up to 4 components in (a and b). Furthermore, we establish absolute errors at $\delta = 1$ for the exact-approximate solutions in the accompanying . Also, we have seen how different fractional orders perform in surface plots and 2D plots in and some $\delta_{1}\delta_{2}-slice$ (a and b) solutions are presented in 4 when $\theta = 0.005$ and $\zeta = 0.5$ . As a result of this behaviour, we might conclude that the approximation solution tends to be a precise solution. Accordingly, as the iteration increases, the absolute inaccuracy decreases. Consequently, as the number of terms grows, the SDM findings approach the exact result.

Figure 1. Numerical behavior of exact and approximate solution to the

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.1 when the parameters are

$\theta = 0.0005, \delta = 1,$ and

$\zeta = 0.5$ .

DownLoad: Full-Size Img PowerPoint

Figure 2. Surface representation of

$\mathcal{U}(\phi, \psi, \zeta)$ absolute error plot for Example 4.1 at

$\theta = 0.005$ and

$\delta = 1$ .

DownLoad: Full-Size Img PowerPoint

Figure 3. Numerical behavior of different fractional orders to the function

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.1 when the parameters are

$\theta = 0.05,$ and

$\zeta = 0.9$ .

DownLoad: Full-Size Img PowerPoint

The graphs in Figures 1–4 assist us to comprehend the behaviour of fractional orders when space and time scale variables fluctuate. Additionally, the findings of this study will aid scientists connected to pattern formation theory, optical designs, or mathematical modelling in comprehending the structural phenomena of the ANOVA-test. Furthermore, the efficiency of the projected method can be boosted by getting additional approximate solution expressions.

Figure 4. Numerical behavior of

$\delta_{1}\delta_{2}$ -slice solution to the

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.1 (a) exact and (b) approximate when the parameters are

$\theta = 0.0005, \delta = 1,$ and

$\zeta = 0.5$ .

DownLoad: Full-Size Img PowerPoint

Example 4.2. Assume the following time-dependent fractional-order Zakharov-Kuznetsov equation:

$\begin{eqnarray} \mathcal{D}_{\zeta}^{\delta}\mathcal{U}+\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big] = 0, \end{eqnarray}$

(4.12)

subject to the initial condition

$\begin{eqnarray} \mathcal{U}(\phi, \psi, 0) = \frac{3}{2}\theta\sinh\Big[\frac{1}{6}(\phi+\psi)\Big], \end{eqnarray}$

(4.13)

where $\theta$ is an arbitrary constant.

Proof. Applying the ST on both sides of (4.12), we find

$\begin{eqnarray} &&\mathbf{S}\Big[\frac{\partial^{\delta}\mathcal{U}}{\partial\zeta^{\delta}}\Big] = -\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big]\bigg], \\&& \Big(\frac{\xi}{\varpi}\Big)^{\delta}\mathbf{S}\big[\mathcal{U}(\phi, \psi, \zeta)\big]-\sum\limits_{\kappa = 0}^{n_{1}-1} \Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\frac{\partial^{\kappa}\mathcal{U}^{(\kappa)}(\phi, 0)}{\partial\zeta^{\kappa}} = -\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big]\bigg]. \end{eqnarray}$

(4.14)

Employing the inverse ST, we have

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\Bigg[ \Big(\frac{\varpi}{\xi}\Big)^{\delta}\sum\limits_{\kappa = 0}^{n_{1}-1} \Big(\frac{\xi}{\varpi}\Big)^{\delta-\kappa-1}\frac{\partial^{\kappa}\mathcal{U}^{(\kappa)}(\phi, 0)}{\partial\zeta^{\kappa}}-\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big]\bigg]\Bigg]. \end{eqnarray}$

(4.15)

It follows that

$\begin{eqnarray} &&\mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\bigg[\frac{\varpi}{\xi}{\mathcal{U}(\phi, \psi, 0)}\bigg]-\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big]\bigg]\Bigg], \\&&\mathcal{U}(\phi, \psi, \zeta) = \mathbf{S}^{-1}\bigg[\frac{3}{2}\frac{\varpi}{\xi}{\theta\sinh\big[\frac{1}{6}(\phi+\psi)\big]}\bigg]-\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\frac{\partial\mathcal{U}^{3}}{\partial \phi}+2\Big[\frac{\partial}{\partial \phi}\Big(\frac{\partial^{2}\mathcal{U}^{3}}{\partial \psi^{2}}\Big)+\frac{\partial^{3}\mathcal{U}^{3}}{\partial \phi^{3}}\Big]\bigg]\Bigg]. \end{eqnarray}$

(4.16)

Utilizing the Shehu's decomposition method, we get

$\begin{eqnarray} \sum\limits_{\jmath = 0}^{\infty}\mathcal{U}_{\jmath}(\phi, \psi, \zeta) = \frac{3}{2}\theta\sinh\big[\frac{1}{6}(\phi+\psi)\big]-\mathbf{S}^{-1}\Bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\check{\mathcal{N}}(\mathcal{U})_{\phi}+\frac{1}{8}\Big[\check{\mathcal{N}}(\mathcal{U})_{\phi\phi\phi}+\check{\mathcal{N}}(\mathcal{U})_{\phi\psi\psi}\Big]\bigg]\Bigg], \end{eqnarray}$

(4.17)

where $\check{\mathcal{N}}(\mathcal{U})$ is the Admoian's polynomial describing nonlinear term appearing in the above mentioned equations.

$\begin{eqnarray} \check{\mathcal{N}}(\mathcal{U}) = \mathcal{U}^{3} = \sum\limits_{\jmath = 0}^{\infty}\mathcal{G}_{\jmath}(\mathcal{U}). \end{eqnarray}$

(4.18)

First few Admoian's polynomials are presented as follows:

$\begin{eqnarray*} \mathcal{G}_{0}&& = \mathcal{U}_{0}^{3}, \nonumber\\ \mathcal{G}_{1}&& = 3\mathcal{U}_{0}^{2}\mathcal{U}_{1}, \nonumber\\ \mathcal{G}_{2}&& = 3\mathcal{U}_{0}^{2}\mathcal{U}_{2}+3\mathcal{U}_{0}^{2}\mathcal{U}_{1}^{2}, \nonumber\\ \mathcal{U}_{0}(\phi, \psi, \zeta)&& = \frac{3}{2}\theta\sinh\Big[\frac{1}{6}(\phi+\psi)\Big], \nonumber\\ \mathcal{U}_{\jmath+1}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{G}_{\jmath}(\mathcal{U})\bigg)_{\phi}+2\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{G}_{\jmath}(\mathcal{U})\bigg)_{\phi\phi\phi}+2\bigg(\sum\limits_{\jmath = 0}^{\infty}\mathcal{G}_{\jmath}(\mathcal{U})\bigg)_{\phi\psi\psi}\bigg], \end{eqnarray*}$

for $\jmath = 0, 1, 2, ...$

$\begin{eqnarray*} \mathcal{U}_{1}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(\mathcal{U}_{0}^{3})_{\phi}+2(\mathcal{U}_{0}^{3})_{\phi\phi\phi}+2(\mathcal{U}_{0}^{3})_{\phi\psi\psi}\bigg]\Bigg]\nonumber\\&& = \bigg(-3\theta^{3}\sinh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]\cosh\Big[\frac{1}{6}(\phi+\psi)\Big]+\frac{3}{8}\theta^{3}\cosh^{3}\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg)\mathbf{S}^{-1}\Big(\Big(\frac{\varpi}{\xi}\Big)^{\delta+1}\Big)\nonumber\\&& = \bigg(-3\theta^{3}\sinh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]\cosh\Big[\frac{1}{6}(\phi+\psi)\Big]+\frac{3}{8}\theta^{3}\cosh^{3}\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg)\frac{\zeta^{\delta}}{\Gamma(\delta+1)}. \end{eqnarray*}$

Accordingly, we can derive the remaining terms as follows

$\begin{eqnarray*} \mathcal{U}_{2}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(3\mathcal{U}_{0}^{2}\mathcal{U}_{1})_{\phi}+2(3\mathcal{U}_{0}^{2}\mathcal{U}_{1})_{\phi\phi\phi}+2(3\mathcal{U}_{0}^{2}\mathcal{U}_{1})_{\phi\psi\psi}\bigg]\Bigg]\nonumber\\&& = \frac{3}{32}\theta^{5}\sinh\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg[765\cosh^{4}\Big[\frac{1}{6}(\phi+\psi)\Big]\nonumber\\&&\quad-729\cosh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]+91\bigg]\frac{\zeta^{2\delta}}{\Gamma(2\delta+1)}, \end{eqnarray*}$

$\begin{eqnarray*} \mathcal{U}_{3}(\phi, \psi, \zeta)&& = -\mathbf{S}^{-1}\bigg[\Big(\frac{\varpi}{\xi}\Big)^{\delta}\mathbf{S}\bigg[(3\mathcal{U}_{0}^{2}\mathcal{U}_{2}+3\mathcal{U}_{0}^{2}\mathcal{U}_{1}^{2})_{\phi}\nonumber\\&&\quad+2(3\mathcal{U}_{0}^{2}\mathcal{U}_{2}+3\mathcal{U}_{0}^{2}\mathcal{U}_{1}^{2})_{\phi\phi\phi}+2(3\mathcal{U}_{0}^{2}\mathcal{U}_{2}+3\mathcal{U}_{0}^{2}\mathcal{U}_{1}^{2})_{\phi\psi\psi}\bigg]\Bigg]\nonumber\\&& = -\frac{3}{128}\cosh\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg[171,738\cosh^{6}\Big[\frac{1}{6}(\phi+\psi)\Big]-349,884\cosh^{4}\Big[\frac{1}{6}(\phi+\psi)\Big]\nonumber\\&&\quad+215,496\cosh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]-36,907\bigg]\frac{\zeta^{3\delta}}{\Gamma(3\delta+1)}. \end{eqnarray*}$

The approximate-analytical SDM solution is

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta)&& = \mathcal{U}_{0}(\phi, \psi, \zeta)+\mathcal{U}_{1}(\phi, \psi, \zeta)+\mathcal{U}_{2}(\phi, \psi, \zeta)+\mathcal{U}_{3}(\phi, \psi, \zeta)+..., \\\mathcal{U}(\phi, \psi, \zeta)&& = \frac{3}{2}\theta\sinh\Big[\frac{1}{6}(\phi+\psi)\Big]-\bigg(3\theta^{3}\sinh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]\cosh\Big[\frac{1}{6}(\phi+\psi)\Big]\\&&\quad+\frac{3}{8}\theta^{3}\cosh^{3}\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg)\frac{\zeta^{\delta}}{\Gamma(\delta+1)}\\&&\quad+\frac{3}{32}\theta^{5}\sinh\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg[765\cosh^{4}\Big[\frac{1}{6}(\phi+\psi)\Big]-729\cosh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]\\&&\quad+91\bigg]\frac{\zeta^{2\delta}}{\Gamma(2\delta+1)}-\frac{3}{128}\cosh\Big[\frac{1}{6}(\phi+\psi)\Big]\bigg[171,738\cosh^{6}\Big[\frac{1}{6}(\phi+\psi)\Big]\\&&\quad-349,884\cosh^{4}\Big[\frac{1}{6}(\phi+\psi)\Big]\\&&\quad+215,496\cosh^{2}\Big[\frac{1}{6}(\phi+\psi)\Big]-36,907\bigg]\frac{\zeta^{3\delta}}{\Gamma(3\delta+1)}+...\; . \end{eqnarray}$

(4.19)

The exact solution for $\delta = 1$ is presented by

$\begin{eqnarray} \mathcal{U}(\phi, \psi, \zeta) = \frac{3}{2}\theta\sinh\Big[\frac{1}{6}(\phi+\psi-\theta\zeta)\Big]. \end{eqnarray}$

(4.20)

show the comparison results for exact, SDM, and absolute error of $\mathcal{U}_{abs} = \|\mathcal{U}^{E}-\mathcal{U}^{SDM}\|$ solution for (4.12), when $\theta = 0001$ and for various fractional orders $\delta = 0.67, \; 0.75, \; 1.$ It can be seen that the proposed method closely matches the exact, and VIM [34].

Table 3. Exact

$(\mathcal{U}_{E})$ and SDM-approximate

$(\mathcal{U}_{SDM})$ solution with absolute error

$(\mathcal{U}_{abs})$ in comparison derived by VIM

$(\mathcal{U}_{VIM})$ [34], for Example 4.2 at

$\theta = 0.001,$

$\delta = 1, \; 0.67$ and 0.75.

$\phi$	$\psi$	$\zeta$	$\mathcal{U}_{E}$	$\mathcal{U}_{SDM}$	$\mathcal{U}_{abs}$	$\mathcal{U}_{0.67}$	$\mathcal{U}_{0.75}$	$\mathcal{U}_{VIM}$
0.1	0.1	0.2	4.996 $\times10^{-5}$	5.001 $\times10^{-5}$	4.988 $\times10^{-8}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$
		0.3	4.993 $\times10^{-5}$	5.001 $\times10^{-5}$	7.481 $\times10^{-8}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$
		0.4	4.991 $\times10^{-5}$	5.001 $\times10^{-5}$	9.975 $\times10^{-8}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$	5.001 $\times10^{-5}$
0.6	0.6	0.2	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$	5.079 $\times10^{-8}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$
		0.3	3.019 $\times10^{-4}$	3.020 $\times10^{-4}$	7.619 $\times10^{-8}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$
		0.4	3.019 $\times10^{-4}$	3.020 $\times10^{-4}$	1.016 $\times10^{-7}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$	3.020 $\times10^{-4}$
0.9	0.9	0.2	4.567 $\times10^{-4}$	4.568 $\times10^{-4}$	5.198 $\times10^{-8}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$
		0.3	4.567 $\times10^{-4}$	4.568 $\times10^{-4}$	7.797 $\times10^{-7}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$
		0.4	4.567 $\times10^{-4}$	4.568 $\times10^{-4}$	1.040 $\times10^{-7}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$	4.568 $\times10^{-4}$

| Show Table

DownLoad: CSV

Taking $\theta = 0.0005$ and $\delta = 1$ , we exhibit the approximate-analytical solution of the fractional KZEs equation up to 4 components in (a and b), respectively. Furthermore, we established absolute errors at different values of $\delta$ for the exact-approximate solutions in the accompanying Figure 6. Also, we have seen how different fractional orders perform in 2D and 3D plots in in (a and b) behaves. Also, denotes the $\delta_{1}\delta_{2}$ -slice solutions for the exact and approximate solutions (a and b), respectively. As a result of this behaviour, we might conclude that the approximation solution tends to actual solution. Accordingly, as iteration increases, the absolute inaccuracy decreases. Consequently, as the iterations expands, the SDM findings approaches the exact result. The graphs in Figures 5–8 assist us to comprehend the behaviour of fractional orders when space and time scale variables fluctuate. Additionally, the findings of this study will aid scientists connecting in pattern formation theory, optical designs, or mathematical modelling in comprehending the structural phenomena of the ANOVA-test. Furthermore, the efficiency of the projected method can be boosted by getting additional approximate solution expressions.

Figure 5. Numerical behavior of exact and approximate solution to the

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.2 when the parameters are

$\theta = 0.0005, \delta = 1,$ and

$\zeta = 0.5$ .

DownLoad: Full-Size Img PowerPoint

Figure 6. Surface representation of

$\mathcal{U}(\phi, \psi, \zeta)$ absolute error plot for Example 4.2 at

$\theta = 0.005$ and

$\delta = 1$ .

DownLoad: Full-Size Img PowerPoint

Figure 7. Numerical behavior of

$\delta_{1}\delta_{2}$ -slice solution to the

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.2 (a) exact and (b) approximate when the parameters are

$\theta = 0.0005, \delta = 1,$ and

$\zeta = 0.5$ .

DownLoad: Full-Size Img PowerPoint

Figure 8. Numerical behavior of

$\delta_{1}\delta_{2}$ -slice solution to the

$\mathcal{U}(\phi, \psi, \zeta)$ for Example 4.1 (a) exact and (b) approximate when the parameters are

$\theta = 0.0005, \delta = 1,$ and

$\zeta = 0.5$ .

DownLoad: Full-Size Img PowerPoint

5. Conclusions

In this paper, the Shehu decomposition method (SDM) is effectively implemented for solving nonlinear time-fractional ZKEs. The proposed findings illustrate that there is a strong correlation between the projected method and the closed form solutions. Moreover, the governed approach is reliable and pragmatic for solving other diverse linear and nonlinear PDEs appearing in various disciplines of physics and mathematics. However, this methodology does not necessitate the condition matrix, Lagrange multiplier, or costly integration calculations, so the findings are noise-free, which addresses the drawbacks of earlier techniques. It is worth mentioning that the proposed methods are pragmatic analytical tools for identifying approximate-analytical solutions to complicated nonlinear PDEs. Moreover, we deduce that this approach will be used to deal with other non-linear fractional order systems of equations that are extremely complex.

Acknowledgments

This research was supported by Taif University Research Supporting Project Number (TURSP-2020/96), Taif University, Taif, Saudi Arabia.

Conflict of interest

The authors declare that they have no conflict of interest.

References

[1]	K. Zhou, Y. Zhan, D. Fu, Learning region-based attention network for traffic sign recognition, Sensors, 21 (2021), 686. https://doi.org/10.3390/s21030686 doi: 10.3390/s21030686
[2]	Z. Liu, Y. Cai, H. Wang, L. Chen, H. Gao, Y. Jia, et al., Robust target recognition and tracking of self-driving cars with radar and camera information fusion under severe weather conditions. IEEE T. Intell. Transp. Syst., 23 (2021), 6640–6653. https://doi.org/10.1109/TITS.2021.3059674 doi: 10.1109/TITS.2021.3059674
[3]	M. Hnewa, H. Radha, Object detection under rainy conditions for autonomous vehicles: A review of state-of-the-art and emerging techniques, IEEE Signal Proc. Mag., 38 (2020), 53–67. https://doi.org/10.1109/MSP.2020.2984801 doi: 10.1109/MSP.2020.2984801
[4]	O. Soufi, Z. Aarab, F. Belouadha, Benchmark of deep learning models for single image super-resolution (SISR), In: 2022 2nd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), 2022. https://doi.org/10.1109/IRASET52964.2022.9738274
[5]	K. Li, S. Yang, R. Dong, X. Wang, J. Huang, Survey of single image super‐resolution reconstruction, IET Image Processing, 14 (2022), 2273–2290. https://doi.org/10.1049/iet-ipr.2019.1438 doi: 10.1049/iet-ipr.2019.1438
[6]	D. Qiu, Y. Cheng, X. Wang, Medical image super-resolution reconstruction algorithms based on deep learning: A survey, Comput. Meth. Prog. Bio., 238 (2023), 107590. https://doi.org/10.1016/j.cmpb.2023.107590 doi: 10.1016/j.cmpb.2023.107590
[7]	L. Zhang, R. Dong, S. Yuan, W. Li, J. Zheng, H. Fu, Making low-resolution satellite images reborn: A deep learning approach for super-resolution building extraction, Remote Sens., 13 (2021), 2872. https://doi.org/10.3390/rs13152872 doi: 10.3390/rs13152872
[8]	H. Chen, X. He, L. Qing, Y. Wu, C. Ren, R. E. Sheriff, et al., Real-world single image super-resolution: A brief review, Inform. Fusion, 79 (2022), 124–145. https://doi.org/10.1016/j.inffus.2021.09.005 doi: 10.1016/j.inffus.2021.09.005
[9]	S. C. Park, M. K. Park, M. G. Kang, Super-resolution image reconstruction: A technical overview, IEEE Signal Proc. Mag., 20 (2003), 21–36. https://doi.org/10.1109/MSP.2003.1203207 doi: 10.1109/MSP.2003.1203207
[10]	D. O. Baguer, J. Leuschner, M. Schmidt, Computed tomography reconstruction using deep image prior and learned reconstruction methods, Inverse Probl., 36 (2020), 094004. https://doi.org/10.1088/1361-6420/aba415 doi: 10.1088/1361-6420/aba415
[11]	J. Xiao, H. Yong, L. Zhang, Degradation model learning for real-world single image super-resolution, In: Computer Vision–ACCV 2020, 2020. https://doi.org/10.1007/978-3-030-69532-3_6
[12]	P. Wu, J. Liu, M. Li, Y. Sun, F. Shen, Fast sparse coding networks for anomaly detection in videos, Pattern Recogn., 107 (2020), 107515. https://doi.org/10.1016/j.patcog.2020.107515 doi: 10.1016/j.patcog.2020.107515
[13]	J. Li, S. Wei, W. Dai, Combination of manifold learning and deep learning algorithms for mid-term electrical load forecasting, IEEE T. Neur. Net. Lear. Syst., 34 (2023), 2584–2593. https://doi.org/10.1109/TNNLS.2021.3106968 doi: 10.1109/TNNLS.2021.3106968
[14]	F. Deeba, S. Kun, F. Ali Dharejo, Y. Zhou, Sparse representation based computed tomography images reconstruction by coupled dictionary learning algorithm, IET Image Process., 14 (2020), 2365–2375. https://doi.org/10.1049/iet-ipr.2019.1312 doi: 10.1049/iet-ipr.2019.1312
[15]	C. Dong, C. C. Loy, K. He, X. Tang, Learning a deep convolutional network for image super-resolution, In: Computer Vision–ECCV 2014, 2014,184–199. https://doi.org/10.1007/978-3-319-10593-2_13
[16]	J. Kim, J. K. Lee, K. M. Lee, Deeply-recursive convolutional network for image super-resolution, In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, 1637–1645. https://doi.org/10.1109/CVPR.2016.181
[17]	Y. Tai, J. Yang, X. Liu, Image super-resolution via deep recursive residual network, In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, 2790–2798. https://doi.org/10.1109/CVPR.2017.298
[18]	Y. Zhang, K. Li, K. Li, L. Wang, B. Zhong, Y. Fu, Image super-resolution using very deep residual channel attention networks, In: Computer Vision–ECCV 2018, 2018. 294–310. https://doi.org/10.1007/978-3-030-01234-2_18
[19]	T. Dai, J. Cai, Y. Zhang, S. T. Xia, L. Zhang, Second-order attention network for single image super-resolution, In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, 11057–11066. https://doi.org/10.1109/cvpr.2019.01132
[20]	P. Wei, Z. Xie, H. Lu, Z. Zhan, Q. Ye, W. Zuo, et al., Component divide-and-conquer for real-world image super-resolution, In: Computer Vision–ECCV 2020, 2020,101–117. https://doi.org/10.1007/978-3-030-58598-3_7
[21]	W. Zhang, W. Zhao, J. Li, P. Zhuang, H. Sun, Y. Xu, et al., CVANet: Cascaded visual attention network for single image super-resolution, Neural Networks, 170 (2024), 622–634. https://doi.org/10.1016/j.neunet.2023.11.049 doi: 10.1016/j.neunet.2023.11.049
[22]	Y. Wang, S. Jin, Z. Yang, H. Guan, Y. Ren, K. Cheng, et al., TTSR: A transformer-based topography neural network for digital elevation model super-resolution, IEEE T. Geosci. Remote Sens., 62 (2024), 4403179. https://doi.org/10.1109/TGRS.2024.3360489 doi: 10.1109/TGRS.2024.3360489
[23]	Y. Chen, R. Xia, K. Yang, K. Zou, MICU: Image super-resolution via multi-level information compensation and U-net, Expert Syst. Appl., 245 (2024), 123111. https://doi.org/10.1016/j.eswa.2023.123111 doi: 10.1016/j.eswa.2023.123111
[24]	Z. H. Qu, Y. M. Shao, T. M. Deng, J. Zhu, X. H. Song, Traffic sign detection and recognition under complex lighting conditions, Laser. Optoelectron. P., 56 (2019), 231009. https://doi.org/10.3788/LOP56.231009 doi: 10.3788/LOP56.231009
[25]	X. G. Zhang, X. L. Liu, J. Li, H. D. Wang, Real-time detection and recognition of speed limit traffic signs under BP neural network, J. Xidian Univ., 45 (2018), 136–142. https://doi.org/10.3969/j.issn.1001-2400.2018.05.022 doi: 10.3969/j.issn.1001-2400.2018.05.022
[26]	G. Z. Xu, Y. Zhou, B. Dong, C. C. Liao, Traffic signage recognition based on improved cascade R-CNN. Sens. Microsyst., 40 (2021), 142–145+153. https://doi.org/10.13873/j.1000-9787(2021)05-0142-04 doi: 10.13873/j.1000-9787(2021)05-0142-04
[27]	L. Liu, S. Lu, R. Zhong, B. Wu, Y. Yao, Q. Zhang, et al., Computing systems for autonomous driving: State of the art and challenges, IEEE Internet Things J., 8 (2021), 6469–6486. https://doi.org/10.1109/JIOT.2020.3043716 doi: 10.1109/JIOT.2020.3043716
[28]	H. Singh, A. Kathuria, Analyzing driver behavior under naturalistic driving conditions: A review. Accident Anal. Prev., 150 (2021), 105908. https://doi.org/10.1016/j.aap.2020.105908 doi: 10.1016/j.aap.2020.105908
[29]	S. Woo, J. Park, J. Y. Lee, I. S. Kweon, CBAM: Convolutional block attention module, In: Computer Vision–ECCV 2018, 2018, 3–19. https://doi.org/10.1007/978-3-030-01234-2_1
[30]	J. Deng, W. Dong, R. Socher, L. J. Li, K. Li, F. F. Li, Imagenet: A large-scale hierarchical image database, In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009,248–255. https://doi.org/10.1109/CVPR.2009.5206848
[31]	X. Wang, K. Yu, C. Dong, C. C. Loy, Recovering realistic texture in image super-resolution by deep spatial feature transform, In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018,606–615. https://doi.org/10.1109/CVPR.2018.00070
[32]	A. Ignatov, N. Kobyshev, R. Timofte, K. Vanhoey, DSLR-quality photos on mobile devices with deep convolutional networks, In: 2017 IEEE International Conference on Computer Vision (ICCV), 2017, 3297–3305. https://doi.org/10.1109/ICCV.2017.355
[33]	J. Hu, L. Shen, S. Albanie, G. Sun, E. Wu, Squeeze-and-excitation networks, IEEE T. Pattern. Anal., 42 (2020), 2011–2023. https://doi.org/10.1109/tpami.2019.2913372 doi: 10.1109/tpami.2019.2913372
[34]	Z. Cui, N. Wang, Y. Su, W. Zhang, Y. Lan, A. Li, ECANet: Enhanced context aggregation network for single image dehazing, Signal Image Video P., 17 (2023), 471–479. https://doi.org/10.1007/s11760-022-02252-w doi: 10.1007/s11760-022-02252-w
[35]	J. Xu, Z. Li, B. Du, M. Zhang, J. Liu, Reluplex made more practical: Leaky ReLU, In: 2020 IEEE Symposium on Computers and Communications (ISCC), 2020, 1–7. https://doi.org/10.1109/ISCC50000.2020.9219587
[36]	F. Nie, H. Huang, X. Cai, C. Ding, Efficient and robust feature selection via joint ℓ_{2, 1}-norms minimization, Adv. Neural Inform. Processing Syst., 2010.
[37]	A. Hore, D. Ziou, Image quality metrics: PSNR vs. SSIM, In: 2010 20th International Conference on Pattern Recognition, 2010, 2366–2369. https://doi.org/10.1109/ICPR.2010.579
[38]	D. Han, Comparison of commonly used image interpolation methods, In: Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013), 2013, 1556–1559. https://doi.org/10.2991/iccsee.2013.391

This article has been cited by:

1.	Mostafa M.A. Khater, Thongchai Botmart, Unidirectional shallow water wave model; Computational simulations, 2022, 42, 22113797, 106010, 10.1016/j.rinp.2022.106010
2.	Alina Alb Lupas, Characteristics of a Subclass of Analytic Functions Introduced by Using a Fractional Integral Operator, 2022, 8, 2409-5761, 75, 10.15377/2409-5761.2021.08.5
3.	Maysaa Al Qurashi, Saima Rashid, Sobia Sultana, Fahd Jarad, Abdullah M. Alsharif, Fractional-order partial differential equations describing propagation of shallow water waves depending on power and Mittag-Leffler memory, 2022, 7, 2473-6988, 12587, 10.3934/math.2022697
4.	Mostafa M. A. Khater, Computational Traveling Wave Solutions of the Nonlinear Rangwala–Rao Model Arising in Electric Field, 2022, 10, 2227-7390, 4658, 10.3390/math10244658
5.	Vediyappan Govindan, Samad Noeiaghdam, Unai Fernandez-Gamiz, Sagar Ningonda Sankeshwari, R. Arulprakasam, Bing Zhao Li, Shehu Integral Transform and Hyers-Ulam Stability of nth order Linear Differential Equations, 2022, 18, 24682276, e01427, 10.1016/j.sciaf.2022.e01427
6.	Rashid Nawaz, Nicholas Fewster-Young, Nek Muhammad Katbar, Nasir Ali, Laiq Zada, Rabha W. Ibrahim, Wasim Jamshed, Haifa Alqahtani, Numerical inspection of (3 + 1)- perturbed Zakharov–Kuznetsov equation via fractional variational iteration method with Caputo fractional derivative, 2024, 85, 1040-7790, 1162, 10.1080/10407790.2023.2262123
7.	M. L. Rupa, K. Aruna, K. Raghavendar, Insights into the time Fractional Belousov-Zhabotinsky System Arises in Thermodynamics, 2024, 63, 1572-9575, 10.1007/s10773-024-05770-0
8.	Saumya Ranjan Jena, Itishree Sahu, A novel approach for numerical treatment of traveling wave solution of ion acoustic waves as a fractional nonlinear evolution equation on Shehu transform environment, 2023, 98, 0031-8949, 085231, 10.1088/1402-4896/ace6de
9.	Yong Zhang, Graham E. Fogg, HongGuang Sun, Donald M. Reeves, Roseanna M. Neupauer, Wei Wei, Adjoint subordination to calculate backward travel time probability of pollutants in water with various velocity resolutions, 2024, 28, 1607-7938, 179, 10.5194/hess-28-179-2024

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)