https://scholars.lib.ntu.edu.tw/handle/123456789/607196
標題: | All Characteristics Preservation: Single Image Dehazing based on Hierarchical Detail Reconstruction Wavelet Decomposition Network | 作者: | Chen W.-T Fang H.-Y Tsai C.-C JIAN-JIUN DING SY-YEN KUO |
關鍵字: | Demulsification;Discrete wavelet transforms;Economic and social effects;Image reconstruction;Learning systems;Wavelet decomposition;Background reconstruction;Decomposition networks;Dehazing;Detail preservation;Haze removal;Learning-based methods;Single image dehazing;Single images;Trade off;Wavelets decomposition;Image quality | 公開日期: | 2021 | 起(迄)頁: | 5649-5656 | 來源出版物: | IEEE International Conference on Intelligent Robots and Systems | 摘要: | Single image haze removal is crucial in computer vision. In open literatures, two kinds of dehazing strategies (prior-based and learning-based methods) have been developed. However, they have a trade-off between detail preservation and the image quality. Prior-based methods reconstruct the detail well but have lower image quality while learning-based methods achieve better recovered quality but lose the detail. In this paper, to mitigate this dilemma, a hierarchical architecture using the discrete wavelet transform (DWT) is proposed. It divides the dehazing problem into two parts: detail and background reconstruction. Based on investigating how haze affects the image in the wavelet domain, two networks for detail and background reconstruction are proposed. To avoid color distortion and the detail loss, the anti-vanish wavelet loss and the bound penalty are proposed. The multi-level wavelet component discriminator is proposed for further improvement. Experiments show that the proposed network can achieve superior performance in all metrics. ? 2021 IEEE. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85124341581&doi=10.1109%2fIROS51168.2021.9636373&partnerID=40&md5=658603d17351656f6862b2677d4df569 https://scholars.lib.ntu.edu.tw/handle/123456789/607196 |
ISSN: | 21530858 | DOI: | 10.1109/IROS51168.2021.9636373 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。