Author(s)

JinBo Yi

Affiliation(s)

College of Information Technology, Shanghai Ocean University, Shanghai 201306, China.

Corresponding Author

JinBo Yi

ABSTRACT

Hazy images often suffer from low quality when processed by traditional defogging algorithms, which fail to effectively remove haze. To address this issue, this article propose a novel single-image dehazing model based on the AOD-Net architecture. The model leverages depthwise separable convolutions to construct a lightweight deep neural network. Additionally, this article introduces a new conditional convolution and attention mechanism to enhance feature extraction, thereby improving the network’s ability to capture global information from hazy images. To optimize model performance, this article train the proposed model on the NYU dataset and conduct extensive experiments on the same dataset. The dehazing effectiveness is evaluated using full-reference image quality assessment metrics. Experimental results demonstrate that the improved model achieves higher accuracy in dehazing quality compared to existing methods. Furthermore, the incorporation of the new feature extraction module and attention mechanism significantly enhances performance in haze removal, color restoration, and detail preservation, outperforming the original AOD-Net and other traditional approaches.The application feasibility of this technique is extensive: In the domain of autonomous driving, it can enhance the target detection accuracy of on-board cameras in foggy weather; in remote sensing monitoring, it facilitates satellites and unmanned aerial vehicles to obtain clearer surface information; in the area of security surveillance, it can strengthen the reliability of video analysis in low-visibility circumstances. Additionally, the lightweight design of the model can be adapted to edge computing devices, providing technical support for real-time defogging and possessing significant engineering application value and commercial potential.

KEYWORDS

Deep learning; Image dehazing; Atmospheric scattering model

CITE THIS PAPER

JinBo Yi. Single image dehazing based on improved AOD-Net. Journal of Computer Science and Electrical Engineering. 2025, 7(3): 40-45. DOI: https://doi.org/10.61784/jcsee3055.

REFERENCES

[1] He K, Sun J, Tang X. Single image haze removal using dark channel prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.

[2] Jin Y Z, Chen J X, Tian F, et al. LFD-Net: Lightweight feature-interaction dehazing network for real-time remote sensing tasks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 9139-9153.

[3] Goyal B, Dogra A, Lepcha D C, et al. Recent advances in image dehazing: Formal analysis to automated approaches. Information Fusion, 2024, 104: 102151.

[4] Li Y, et al. Low-light image enhancement via deep learning-based feature attention. IEEE Access, 2020, 8: 179524-179536.

[5] Liu H Z, Hu H M. From global to local: An adaptive environmental illumination estimation for non-uniform scattering//Proceedings of the ACM Multimedia Asia 2023. 2023: 1-7.

[6] Li B, Peng X, Wang Z, et al. AOD-net: All-in-one dehazing network//2017 IEEE International Conference on Computer Vision (ICCV). Venice: IEEE, 2017: 4780-4788.

[7] Mccartney E J. Optics of the atmosphere: Scattering by molecules and particles. New York: Wiley, 1976.

[8] Narasimhan S G, Nayar S K. Contrast restoration of weather degraded images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(6): 713-724.

[9] Nayar S K, Narasimhan S G. Vision in bad weather//Proceedings of the Seventh IEEE International Conference on Computer Vision. Kerkyra: IEEE, 1999: 820-827.

[10] Li B, Ren W, Fu D, et al. Benchmarking single image dehazing and beyond. IEEE Transactions on Image Processing, 2019, 28(1): 492-505.

[11] Gao Z L, Xie F, Yin Y, et al. Image defogging algorithm based on semi-supervised learning neural network//2023 3rd International Conference on Robotics, Automation and Intelligent Control (ICRAIC). 2023: 86-89.

[12] Chen S Y, Cheng J, Huang Z Y. GADO-Net: An improved AOD-Net single image dehazing algorithm//2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST). 2021: 640-646.

[13] ARIF Z H, MAHMOUD M A, ABDULKAREEM K H, et al. Comprehensive review of machine learning (ML) in image defogging: Taxonomy of concepts, scenes, feature extraction, and classification techniques. IET Image Processing, 2021, 16(2): 289-310. DOI:10.1049/ipr2.12365.

[14] Anonymous. Research progress and applications of image defogging algorithms. Progress in Medical Devices, 2023.

[15] Agrawal S C, Jalal A S. A joint cumulative distribution function and gradient fusion based method for dehazing of long shot hazy images. Journal of Visual Communication and Image Representation, 2021, 77: 103087.

[16] Zhang J, Dong Q Q, Song W J. GGADN: Guided generative adversarial dehazing network. Soft Computing, 2021, 27(3): 1731-1741.

[17] Nair D, Sankaran P. Benchmarking single image dehazing methods. SN Computer Science, 2021, 3(1): 1.

[18] Naz B, Rahman N, Nafis M T. Enhancement of dehazing methods using DNN and filtering//Computer Networks and Inventive Communication Technologies. Singapore: Springer, 2021: 531-544.