Single Image Dehazing Using Attention Enhanced Net

Minghua Zhu

《World Scientific Research Journal》 9卷5期 (2023/05) Pp. 240-244

https://doi.org/10.6911/WSRJ.202305_9(5).0029

摘要

Producing clear and haze‐free images is crucial for many computer vision systems and algorithms. Therefore, single image dehazing is a vital area of research in this field. In the past, prior‐based methods have shown promising results. However, these methods tend to produce unwanted artifacts in their outputs since their priors cannot account for all possible scenarios. Conversely, learning‐based methods have emerged as a more natural and effective approach. In this paper, we proposed an attention enhanced net for single image dehazing, which incorporates a channel attention branch and a spatial attention branch to enable the network to focus on the most informative parts of the high-dimensional feature map, thereby improving the performance of the subsequent layers in our neural network.

關鍵字

Single image dehazing ； Attention mechanism ； Deep learning

參考文獻

W. E. K. Middleton and V. Twersky, “Vision through the atmosphere,” Phys. Today, vol. 7, no. 3, p. 21, Mar. 1954.

Qin X, Wang Z, Bai Y, et al. FFA-Net: Feature fusion attention network for single image dehazing[C]//Proceedings of the AAAI conference on artificial intelligence. 2020, 34(07): 11908-11915.

Kaiming He, Jian Sun and Xiaoou Tang, "Single image haze removal using dark channel prior," 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, 2009, pp. 1956-1963.

B. Cai, X. Xu, K. Jia, C. Qing and D. Tao, "DehazeNet: An End-to-End System for Single Image Haze Removal," in IEEE Transactions on Image Processing, vol. 25, no. 11, pp. 5187-5198, Nov. 2016.

B. Li, X. Peng, Z. Wang, J. Xu and D. Feng, "AOD-Net: All-in-One Dehazing Network," 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 2017, pp. 4780-4788.