Improving Diabetic Retinopathy Fundus Image Quality Using Dynamic Sparse Representation and Dilated Convolutional Networks
Received: 23 September 2025 | Revised: 17 December 2025 | Accepted: 29 December 2025 | Online: 4 April 2026
Corresponding author: Veerendra Basreddy
Abstract
In recent years, there has been an increase in Diabetic Retinopathy (DR). In addition, the images used for the detection of DR are not of good quality. This study focuses on improving the quality of DR fundus images by proposing an image enhancement approach. Specifically, this work presents a Dynamic Sparse Representation (DSR) approach that combines Convolution layers (Conv), Rectified-Linear-Unit (ReLU), and Dilated-Convolution methods for preprocessing. To further improve image quality, DSR utilizes a dictionary approach during the training phase. Before testing, Additive White Gaussian Noise (AWGN) is induced in the input images to simulate real-world noise conditions. In the testing phase, noisy images are processed to remove noise and enhance image quality. The proposed DSR method was evaluated on the MESSIDOR and EyeQ datasets using the following metrics: Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Mean Squared Error (MSE). For the EyeQ and MESSIDOR datasets, DSR achieved 37.82 PSNR, 0.94 SSIM, 0.0316 MSE, and 36.85 PSNR, 0.93 SSIM, and 0.0431 MSE, respectively. The findings show that the proposed DSR approach achieved better performance compared to other approaches.
Keywords:
diabetic retinopathy, fundus images, image enhancement, convolutional neural network, rectified linear unitDownloads
References
K. Ogurtsova et al., "IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040," Diabetes Research and Clinical Practice, vol. 128, pp. 40–50, June 2017. DOI: https://doi.org/10.1016/j.diabres.2017.03.024
V. T. H. Tuyet, N. T. Binh, and D. T. Tin, "Improving the Curvelet Saliency and Deep Convolutional Neural Networks for Diabetic Retinopathy Classification in Fundus Images," Engineering, Technology & Applied Science Research, vol. 12, no. 1, pp. 8204–8209, Feb. 2022. DOI: https://doi.org/10.48084/etasr.4679
Z. Shen, H. Fu, J. Shen, and L. Shao, "Modeling and Enhancing Low-Quality Retinal Fundus Images," IEEE Transactions on Medical Imaging, vol. 40, no. 3, pp. 996–1006, Mar. 2021. DOI: https://doi.org/10.1109/TMI.2020.3043495
C. Wan et al., "Retinal Image Enhancement Using Cycle-Constraint Adversarial Network," Frontiers in Medicine, vol. 8, Jan. 2022, Art. no. 793726. DOI: https://doi.org/10.3389/fmed.2021.793726
M. König et al., "Quality assessment of colour fundus and fluorescein angiography images using deep learning," British Journal of Ophthalmology, vol. 108, no. 1, pp. 98–104, Jan. 2024. DOI: https://doi.org/10.1136/bjo-2022-321963
D. Shenkut and V. Bhagavatula, "Fundus GAN - GAN-based Fundus Image Synthesis for Training Retinal Image Classifiers," in 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), July 2022, pp. 2185–2189. DOI: https://doi.org/10.1109/EMBC48229.2022.9871771
S. H. Abbood, H. N. A. Hamed, M. S. M. Rahim, A. Rehman, T. Saba, and S. A. Bahaj, "Hybrid Retinal Image Enhancement Algorithm for Diabetic Retinopathy Diagnostic Using Deep Learning Model," IEEE Access, vol. 10, pp. 73079–73086, 2022. DOI: https://doi.org/10.1109/ACCESS.2022.3189374
T. Palaniswamy and M. Vellingiri, "Internet of Things and Deep Learning Enabled Diabetic Retinopathy Diagnosis Using Retinal Fundus Images," IEEE Access, vol. 11, pp. 27590–27601, 2023. DOI: https://doi.org/10.1109/ACCESS.2023.3257988
Q. Hou, P. Cao, L. Jia, L. Chen, J. Yang, and O. R. Zaiane, "Image Quality Assessment Guided Collaborative Learning of Image Enhancement and Classification for Diabetic Retinopathy Grading," IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 3, pp. 1455–1466, Mar. 2023. DOI: https://doi.org/10.1109/JBHI.2022.3231276
K. G. Lee, S. J. Song, S. Lee, H. G. Yu, D. I. Kim, and K. M. Lee, "A deep learning-based framework for retinal fundus image enhancement," PLOS ONE, vol. 18, no. 3, Mar. 2023, Art. no. e0282416. DOI: https://doi.org/10.1371/journal.pone.0282416
W. Zhu, P. Qiu, M. Farazi, K. Nandakumar, O. M. Dumitrascu, and Y. Wang, "Optimal Transport Guided Unsupervised Learning for Enhancing Low-Quality Retinal Images," in 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), Apr. 2023, pp. 1–5. DOI: https://doi.org/10.1109/ISBI53787.2023.10230719
F. Liu and W. Huang, "ESDiff: a joint model for low-quality retinal image enhancement and vessel segmentation using a diffusion model," Biomedical Optics Express, vol. 14, no. 12, Dec. 2023, Art. no. 6563. DOI: https://doi.org/10.1364/BOE.506205
A. Q. Saeed, S. N. H. Sheikh Abdullah, J. Che-Hamzah, A. T. Abdul Ghani, and W. A. K. Abu-ain, "Synthesizing Retinal Images using End-To-End VAEs-GAN Pipeline-Based Sharpening and Varying Layer," Multimedia Tools and Applications, vol. 83, no. 1, pp. 1283–1307, Jan. 2024. DOI: https://doi.org/10.1007/s11042-023-17058-2
A. E. Ilesanmi and T. O. Ilesanmi, "Methods for image denoising using convolutional neural network: a review," Complex & Intelligent Systems, vol. 7, no. 5, pp. 2179–2198, Oct. 2021. DOI: https://doi.org/10.1007/s40747-021-00428-4
H. Fu et al., "Evaluation of Retinal Image Quality Assessment Networks in Different Color-Spaces," in Medical Image Computing and Computer Assisted Intervention – MICCAI 2019, vol. 11764, D. Shen, T. Liu, T. M. Peters, L. H. Staib, C. Essert, S. Zhou, P. T. Yap, and A. Khan, Eds. Springer International Publishing, 2019, pp. 48–56.
"Messidor," ADCIS. https://www.adcis.net/en/third-party/messidor/.
E. Decencière et al., "Feedback on a Publicly Distributed Image Database: The MESSIDOR Database," Image Analysis & Stereology, vol. 33, no. 3, pp. 231-234, Aug. 2014. DOI: https://doi.org/10.5566/ias.1155
Downloads
How to Cite
License
Copyright (c) 2026 Veerendra Basreddy, Sachinkumar Veerashetty
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
