Enhancing Hazy Image Quality with a Modular CNN Encoder–Decoder

ANDIKA MUHAMMAD NUR KHOLIQ, ARIEF SURYADI SATYAWAN, MOKH MIRZA ETNISA HAQIQI, FAJAR RAHMAT AKBAR, IASYA FAIQOH NURROHMAH, AULIA ADAWIYAH, ESTI FITRIA WULANDARI, RENDI TRI SUGIAN

Abstract


This study develops a modular CNN encoder–decoder framework for single-image dehazing by replacing the conventional bottleneck with interchangeable token-mixing modules such as FNet, Spatial-FNet, MLP-Mixer, and gMLP-style designs. The pipeline integrates adaptive preprocessing (CLAHE and histogram matching), photometric augmentations, and training on a controlled subset of the SOTS dataset. Comprehensive quantitative and qualitative evaluations demonstrate substantial improvements over a baseline CNN, with mean PSNR increasing from approximately 18.4 dB to the 23.0–24.0 dB range and SSIM rising from about 0.75 to roughly 0.89–0.91. However, several variants require careful hyperparameter selection and loss-weight tuning to achieve stable performance. The results offer practical guidance for deployment in real-world vision systems.

Keywords


Image Dehazing; CNN; Modular Token-Mixing; Autonomous driving

Full Text:

PDF

References


Abeer Ayoub, Walid El-Shafai, Fathi E. Abd El-Samie, Ehab K. I. Hamad, & El-Sayed M. EL-Rabaie. (2025). Video and image quality improvement using an enhanced optimized dehazing technique. Multimedia Tools and Applications, 84, 22681–22699.

Alona Golts, Daniel Freedman, & Michael Elad. (2020). Unsupervised Single Image Dehazing Using Dark Channel Prior Loss. IEEE Transactions on Image Processing, 29, 2692–2701.

Dakshinamurthi, V., G.P., S., P, M., & Hussain, Sk. R. (2023). DEEP LEARNING-BASED IMAGE DEHAZING AND VISIBILITY ENHANCEMENT FOR IMPROVED VISUAL PERCEPTION. ICTACT Journal on Image and Video Processing, 14(2), 3122–3128. https://doi.org/10.21917/ijivp.2023.0444

Deshmukh, P. V., Kumar, A., & Chakrabarti, P. (2024). Review of Deep Learning Based Image Dehazing for Autonomous Vehicle. In Tuijin Jishu/Journal of Propulsion Technology (Vol. 45, Issue 4).

Filin, A., Kopylov, A., & Gracheva, I. (2023). A SINGLE IMAGE DEHAZING DATASET with LOW-LIGHT REAL-WORLD INDOOR IMAGES, DEPTH MAPS and INFRARED IMAGES. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 48(2/W3-2023), 53–57. https://doi.org/10.5194/isprs-archives-XLVIII-2-W3-2023-53-2023

Gonzalez, R. C. ., & Woods, R. E. . (2018). Digital image processing. Pearson.

Gui, J., Cong, X., Cao, Y., Ren, W., Zhang, J., Zhang, J., Cao, J., & Tao, D. (2022). A Comprehensive Survey and Taxonomy on Single Image Dehazing Based on Deep Learning. http://arxiv.org/abs/2106.03323

Gui, J., Cong, X., Cao, Y., Ren, W., Zhang, J., Zhang, J., & Tao, D. (2021). A Comprehensive Survey on Image Dehazing Based on Deep Learning.

Heaton, J. (2018). Ian Goodfellow, Yoshua Bengio, and Aaron Courville: Deep learning. Genetic Programming and Evolvable Machines, 19(1–2), 305–307. https://doi.org/10.1007/s10710-017-9314-z

Jing, H., Chen, J., Zhang, C., Wei, S., Chen, A., & Zhang, M. (2024). Adaptive Multi-Feature Attention Network for Image Dehazing. Electronics (Switzerland), 13(18). https://doi.org/10.3390/electronics13183706

Kaiming He, Xiaoou Tang, & Jian Sun. (2011). Single image haze removal using dark channel prior (Vol. 33). IEEE.

Lee-Thorp, J., Ainslie, J., Eckstein, I., & Ontanon, S. (2022). FNet: Mixing Tokens with Fourier Transforms. http://arxiv.org/abs/2105.03824

Lestari, I. S., Jumadi, J., & Lukman, N. (2024). IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK DENGAN PRE-TRAINED MODEL MOBILENETV2 UNTUK DETEKSI KOLESTEROL. Rabit : Jurnal Teknologi Dan Sistem Informasi Univrab, 9(2), 173–183. https://doi.org/10.36341/rabit.v9i2.4732

Liu, H., Dai, Z., So, D. R., & Le, Q. V. (2021). Pay Attention to MLPs. http://arxiv.org/abs/2105.08050

Pavethra, M., & Uma Devi, M. (2024). A cross layer graphical neural network based convolutional neural network framework for image dehazing. Automatika, 65(3), 1139–1153. https://doi.org/10.1080/00051144.2024.2346964

Sharma, T., & Verma, N. K. (2024). Artificial Intelligent Algorithms for Image Dehazing and Non-Uniform Illumination Enhancement. Springer Nature Singapore. https://books.google.co.id/books?id=3wIPEQAAQBAJ

Sopian, S. M., Satyawan, A. S., Haqiqi, M. M. E., Susilawati, H., Wijaya, B., Artemysia, K. A., Firman, & Samie, M. I. (2025). Development of a Modified CycleGAN Model with Residual Blocks and Perceptual Loss for Image Dehazing. Teknika, 14(2), 232–238. https://doi.org/10.34148/teknika.v14i2.1235

Stimper, V., Bauer, S., Ernstorfer, R., Schölkopf, B., & Xian, R. P. (2019). Multidimensional Contrast Limited Adaptive Histogram Equalization. IEEE Access, 7, 165437–165447. https://doi.org/10.1109/ACCESS.2019.2952899

Supandi S, Ghofur S, & Santoso F. (2025). Deteksi Kualitas Rumput Laut Menggunakan Metode Convelution Neural Network (Cnn) Berdasarkan Citra Digital (Studi Kasus : Desa Alasmalang Kecamatan Raas Sumenep). Rabit: J. Teknol. Dan Sistem Informasi Univrab, 10, 604–614. https://doi.org/10.36341/rabit.v10i2.7273

Szeliski, R. (2021). Computer Vision: Algorithms and Applications 2nd Edition. https://szeliski.org/Book,

Tolstikhin, I., Houlsby, N., Kolesnikov, A., Beyer, L., Zhai, X., Unterthiner, T., Yung, J., Steiner, A., Keysers, D., Uszkoreit, J., Lucic, M., & Dosovitskiy, A. (2021). MLP-Mixer: An all-MLP Architecture for Vision. http://arxiv.org/abs/2105.01601

Tyas, D. A., & Ratnaningsih, T. (2023). Implementasi Mask-Rcnn Pada Dataset Kecil Citra Sel Darah Merah Berdasarkan Kriteria Warna Sel. Rabit : Jurnal Teknologi Dan Sistem Informasi Univrab, 8(1), 100–104. https://doi.org/10.36341/rabit.v8i1.3026

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2023). Attention Is All You Need. http://arxiv.org/abs/1706.03762

Zheng, C., Ying, W., & Hu, Q. (2025). Comparative analysis of dehazing algorithms on real-world hazy images. Scientific Reports, 15(1). https://doi.org/10.1038/s41598-025-95510-z




DOI: https://doi.org/10.26760/elkomika.v14i1.69

Refbacks

  • There are currently no refbacks.


 

_______________________________________________________________________________________________________________________

ISSN (print) : 2338-8323 | ISSN (electronic) : 2459-9638

Publisher:

Department of Electrical Engineering Institut Teknologi Nasional Bandung, Indonesia

Address: 20th Building  Institut Teknologi Nasional Bandung PHH. Mustofa Street No. 23 Bandung 40124, Indonesia

Contact: +627272215 (ext. 206)

Email: jte.itenas@itenas.ac.id________________________________________________________________________________________________________________________


Free counters!

Web Analytics Made Easy - StatCounter

Statistic Journal

Jurnal ini terlisensi oleh Creative Commons Attribution-ShareAlike 4.0 International License.

Creative Commons License