Hybrid Image Inpainting Using Wavelet and Cahn-Hilliard Model
DOI:
https://doi.org/10.56714/bjrs.51.2.9Keywords:
Image Inpainting, Frequency Domain, Discrete Wavelet Transform (DWT), Cahn-Hilliard equation, Implicit finite differences methodAbstract
With a wide range of applications in digital image processing, including multimedia editing, medical imaging, and cultural heritage preservation, image restoration is a fundamental task. In this work, we present a hybrid image restoration framework that combines a diffusion model based on the Cahn-Hilliard equation with a single-level wavelet transform to decompose an image into distinct frequency components. To maintain structural continuity, this hybrid method leverages the benefits of wavelets in the frequency domain for texture reconstruction and their propagation capabilities in the spatial domain. To achieve smooth information propagation and maintain edge sharpness, the finite implicit finite difference approach is used to solve the problem numerically. The effect of several types of wavelets was analyzed, including Haar, db1, sym1, bior1.1, db4, sym4, coif3, and others. The results indicate that the performance of wavelets varies depending on the length of the support and the smoothness of the fundamental functions, and provide practical guidance for selecting the most suitable wavelet for image restoration in modern image processing applications
Downloads
References
[1] Z. A. A. Kareem and A. K. Al-Jaberi, “A Novel Image Inpainting Technique Based on Isotropic Diffusion,” Basrah J. Sci., vol. 40, no. 2, pp. 289–305, 2022. DOI: 10.29072/basjs.20220203 DOI: https://doi.org/10.29072/basjs.20220203
[2] A. K. Al-Jaberi, A. Asaad, S. A. Jassim, and N. Al-Jawad, “Topological data analysis to improve exemplar-based inpainting,” in Mobile Multimedia/Image Processing, Security, and Applications 2018, SPIE, 2018, pp. 13–24. DOI: 10.1117/12.2309931 DOI: https://doi.org/10.1117/12.2309931
[3] Y. Huang et al., “Wavedm: Wavelet-based diffusion models for image restoration,” IEEE Trans. Multimed., vol. 26, pp. 7058–7073, 2024. DOI: 10.1109/TMM.2024.3359769 DOI: https://doi.org/10.1109/TMM.2024.3359769
[4] Z. A. A. Kareem and A. K. Al-Jaberi, “Removing Text and Inpainting Irregular Regions in Digital Images within Spatial and Frequency Domains,” Basrah J. Sci., vol. 43, no. 1, pp. 35–50, 2025. DOI: 10.29072/basjs.20250103
[5] A. Kiruluta and A. Lemos, “A hybrid wavelet-fourier method for next-generation conditional diffusion models,” arXiv Prepr. arXiv2504.03821, 2025. DOI: 10.48550/arXiv.2504.03821
[6] Y. Yu et al., “Wavefill: A wavelet-based generation network for image inpainting,” in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 14114–14123. DOI: 10.1109/ICCV48922.2021.01385 DOI: https://doi.org/10.1109/ICCV48922.2021.01385
[7] R. H. Chan, S. Setzer, and G. Steidl, “Inpainting by flexible Haar-wavelet shrinkage,” SIAM J. Imaging Sci., vol. 1, no. 3, pp. 273–293, 2008. DOI:https://doi.org/10.1137/070711499 DOI: https://doi.org/10.1137/070711499
[8] T. F. Chan, J. Shen, and H.-M. Zhou, “Total variation wavelet inpainting,” J. Math. Imaging Vis., vol. 25, no. 1, pp. 107–125, 2006) . DOI: 10.1007/s10851-006-5257-3 DOI: https://doi.org/10.1007/s10851-006-5257-3
[9] A. Averbuch, P. Neittaanmäki, V. Zheludev, M. Salhov, and J. Hauser, “Image inpainting using directional wavelet packets originating from polynomial splines,” Signal Process. Image Commun., vol. 97, p. 116334, 2021) .DOI : 10.1007/s10444-023-10024-4 DOI: https://doi.org/10.1016/j.image.2021.116334
[10] S. He, G. Lin, T. Li, and Y. Chen, “Frequency-Domain Fusion Transformer for Image Inpainting,” arXiv Prepr. arXiv2506.18437, 2025. DOI: https://doi.org/10.48550/arXiv.2506.18437
[11] B. Dong, Q. Jiang, and Z. Shen, “Image restoration: Wavelet frame shrinkage, nonlinear evolution pdes, and beyond,” Multiscale Model. Simul., vol. 15, no. 1, pp. 606–660, 2017.DOI: https://doi.org/10.1137/15M1037457 DOI: https://doi.org/10.1137/15M1037457
[12] H. Zhang and S. Dai, “Image inpainting based on wavelet decomposition,” Procedia Eng., vol. 29, pp. 3674–3678, 2012) .DOI: 10.1016/j.proeng.2012.01.551 DOI: https://doi.org/10.1016/j.proeng.2012.01.551
[13] R. H. Chan, J. Yang, and X. Yuan, “Alternating direction method for image inpainting in wavelet domains,” SIAM J. Imaging Sci., vol. 4, no. 3, pp. 807–826, 2011. DOI: https://doi.org/10.1137/100807247 DOI: https://doi.org/10.1137/100807247
[14] W. Guo and L.-H. Qiao, “Inpainting based on total variation,” in 2007 International Conference on Wavelet Analysis and Pattern Recognition, IEEE, 2007, pp. 939–943. DOI: https://doi.org/10.1109/ICWAPR.2007.4420804 DOI: https://doi.org/10.1109/ICWAPR.2007.4420804
[15] Y. Cao, R. Ma, K. Zhao, and P. An, “WFIL-NET: image inpainting based on wavelet downsampling and frequency integrated learning module,” Multimed. Syst., vol. 31, no. 1, p. 21, 2025. DOI: https://doi.org/10.1007/s00530-024-01609-0 DOI: https://doi.org/10.1007/s00530-024-01609-0
[16] D. O. Kolodochka and M. V Polyakova, “LAMA-WAVELET: IMAGE IMPAINTING WITH HIGH QUALITY OF FINE DETAILS AND OBJECT EDGES,” Radio Electron. Comput. Sci. Control, no. 1, p. 208, 2024.DOI: https://doi.org/10.15588/1607-3274-2024-1-19 DOI: https://doi.org/10.15588/1607-3274-2024-1-19
[17] A. L. Bertozzi, S. Esedoglu, and A. Gillette, “Inpainting of binary images using the Cahn–Hilliard equation,” IEEE Trans. image Process., vol. 16, no. 1, pp. 285–291, 2006. DOI: https://doi.org/10.1109/TIP.2006.887728 DOI: https://doi.org/10.1109/TIP.2006.887728
[18] J. W. Cahn and J. E. Hilliard, “Free energy of a nonuniform system. I. Interfacial free energy,” J. Chem. Phys., vol. 28, no. 2, pp. 258–267, 1958. DOI: https://doi.org/10.1109/TIP.2006.887728 DOI: https://doi.org/10.1063/1.1744102
[19] Y. Zhang, H. Zheng, X. Xu, and H. Zhu, “HVI-Based Spatial–Frequency-Domain Multi-Scale Fusion for Low-Light Image Enhancement.,” Appl. Sci., vol. 15, no. 19, 2025. DOI: https://doi.org/10.3390/app151910376 DOI: https://doi.org/10.3390/app151910376
[20] X. Guan, R. He, L. Wang, H. Zhou, Y. Liu, and H. Xiong, “DWTMA-Net: Discrete Wavelet Transform and Multi-Dimensional Attention Network for Remote Sensing Image Dehazing,” Remote Sens., vol. 17, no. 12, p. 2033, 2025. DOI: https://doi.org/10.3390/rs17122033 DOI: https://doi.org/10.3390/rs17122033
[21] K. E. C. Vidyasagar, K. R. Kumar, G. N. K. A. Sai, M. Ruchita, and M. J. Saikia, “Signal to image conversion and convolutional neural networks for physiological signal processing: A review,” Ieee Access, vol. 12, pp. 66726–66764, 2024. DOI: https://doi.org/10.1109/ACCESS.2024.3399114 DOI: https://doi.org/10.1109/ACCESS.2024.3399114
[22] A. K. Al-Jaberi, S. A. Jassim, and N. Al-Jawad, “Colourizing monochrome images,” in Mobile Multimedia/Image Processing, Security, and Applications 2018, SPIE, 2018, pp. 25–37. DOI: https://doi.org/10.3390/rs17122033 DOI: https://doi.org/10.1117/12.2309938
[23] K. Gu, S. Wang, G. Zhai, W. Lin, X. Yang, and W. Zhang, “Analysis of distortion distribution for pooling in image quality prediction,” IEEE Trans. Broadcast., vol. 62, no. 2, pp. 446–456, 2016. DOI: https://doi.org/10.1109/TBC.2015.2511624 DOI: https://doi.org/10.1109/TBC.2015.2511624
[24] A. C. Bovik, Handbook of image and video processing. Academic press, 2010. DOI: https://doi.org/10.1016/B978-0-12-119792-6.x5062-1 DOI: https://doi.org/10.1016/B978-0-12-119792-6.X5062-1
[25] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. image Process., vol. 13, no. 4, pp. 600–612, 2004. DOI: https://doi.org/10.1109/TIP.2003.819861 DOI: https://doi.org/10.1109/TIP.2003.819861
[26] A. Singhal, R. P. Singh, and M. Tenguria, “Comparison of different wavelets for watermarking of colored images,” in 2011 3rd International Conference on Electronics Computer Technology, IEEE, 2011, pp. 187–191. DOI: https://doi.org/10.1109/ICECTECH.2011.187 DOI: https://doi.org/10.1109/ICECTECH.2011.5941828
[27] H. Wu, F. He, Y. Duan, X. Yan, and B. Fan, “Haar-wavelet based texture inpainting for human pose transfer,” Inf. Process. Manag., vol. 61, no. 3, p. 103612, 2024. DOI: https://doi.org/10.1016/j.ipm.2023.103612 DOI: https://doi.org/10.1016/j.ipm.2023.103612
[28] P. Viswanathan and K. Palanisamy, “An Empirical Selection of Wavelet for Near-lossless Medical Image Compression,” 2024. DOI: https://doi.org/10.2174/1573405620666230330113833 DOI: https://doi.org/10.2174/1573405620666230330113833
[29] H. Li and K. Xu, “Innovative adaptive edge detection for noisy images using wavelet and Gaussian method,” Sci. Rep., vol. 15, no. 1, p. 5838, 2025. DOI: https://doi.org/10.1038/s41598-025-86860-9 DOI: https://doi.org/10.1038/s41598-025-86860-9
[30] W. Wronka, “Optimal mother wavelet selection for a stochastic resynthesis of the sound textures,” Vib. Phys. Syst., vol. 34, no. 2, 2023. DOI: https://vibsys.put.poznan.pl/article/optimal-mother-wavelet-selection-for-a-stochastic-resynthesis-of-the-sound-textures/
[31] T. Xu et al., “Uncovering the over-smoothing challenge in image super-resolution: Entropy-based quantification and contrastive optimization,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 46, no. 9, pp. 6199–6215, 2024. DOI: https://doi.org/10.1109/TPAMI.2024.3378704 DOI: https://doi.org/10.1109/TPAMI.2024.3378704
[32] Y. Awad, H. Fakih, and Y. Alkhezi, “Numerical Linear Algebra for the Two-Dimensional Bertozzi–Esedoglu–Gillette–Cahn–Hilliard Equation in Image Inpainting,” Mathematics, vol. 11, no. 24, p. 4952, 2023. DOI: https://doi.org/10.3390/math11244952 DOI: https://doi.org/10.3390/math11244952
[33] I. H. Latif, S. H. Abdulredha, and S. K. A. Hassan, “Discrete wavelet transform-based image processing: A review,” Al-Nahrain J. Sci., vol. 27, no. 3, pp. 109–125, 2024. DOI: https://doi.org/10.22401/ANJS.27.3.13 DOI: https://doi.org/10.22401/ANJS.27.3.13
[34] Y. Guo et al., “A Novel and Optimized Sine–Cosine Transform Wavelet Threshold Denoising Method Based on the Sym4 Basis Function and Adaptive Threshold Related to Noise Intensity,” Appl. Sci., vol. 13, no. 19, p. 10789, 2023. DOI: https://doi.org/10.3390/app131910789 DOI: https://doi.org/10.3390/app131910789
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Basrah Researches Sciences
This work is licensed under a Creative Commons Attribution 4.0 International License.
