文献类型：期刊文献

英文题名：Adaptive Multi-Feature Attention Network for Image Dehazing

作者：Jing, Hongyuan[1,2];Chen, Jiaxing[1,2];Zhang, Chenyang[2];Wei, Shuang[2];Chen, Aidong[1,2,3];Zhang, Mengmeng[1,2,3]

第一作者：Jing, Hongyuan

通讯作者：Jing, HY[1];Zhang, MM[1];Jing, HY[2];Zhang, MM[2];Zhang, MM[3]

机构：[1]Beijing Key Lab Informat Serv Engn, Coll Robot, Beijing 100101, Peoples R China;[2]Beijing Union Univ, Coll Robot, 4 Gongti North Rd, Beijing, Peoples R China;[3]Beijing Union Univ, Multiagent Syst Res Ctr, 97 Beisihuan East Rd, Beijing 100101, Peoples R China

第一机构：Beijing Key Lab Informat Serv Engn, Coll Robot, Beijing 100101, Peoples R China

通讯机构：[1]corresponding author), Beijing Key Lab Informat Serv Engn, Coll Robot, Beijing 100101, Peoples R China;[2]corresponding author), Beijing Union Univ, Coll Robot, 4 Gongti North Rd, Beijing, Peoples R China;[3]corresponding author), Beijing Union Univ, Multiagent Syst Res Ctr, 97 Beisihuan East Rd, Beijing 100101, Peoples R China.|[11417]北京联合大学;[1141739]北京联合大学机器人学院;

年份：2024

卷号：13

期号：18

外文期刊名：ELECTRONICS

收录：;Scopus(收录号：2-s2.0-85205054967);WOS:【SCI-EXPANDED(收录号:WOS:001323959700001)】；

基金：This work is mainly supported by National Natural Science Foundation of China (NSFC) under Grant 62473055 and the Beijing Nova Program under Grant 20230484477, and the Beijing Natural Science Foundation Program under Grant L223022, and it is partly supported by Beijing Municipal Education Commission Research Foundation under Grant KM202111417008 and the Beijing Union University under Grant ZK20202401.

语种：英文

外文关键词：dehazing; deep learning; attention mechanism; adaptive feature fusion

摘要：Currently, deep-learning-based image dehazing methods occupy a dominant position in image dehazing applications. Although many complicated dehazing models have achieved competitive dehazing performance, effective methods for extracting useful features are still under-researched. Thus, an adaptive multi-feature attention network (AMFAN) consisting of the point-weighted attention (PWA) mechanism and the multi-layer feature fusion (AMLFF) is presented in this paper. We start by enhancing pixel-level attention for each feature map. Specifically, we design a PWA block, which aggregates global and local information of the feature map. We also employ PWA to make the model adaptively focus on significant channels/regions. Then, we design a feature fusion block (FFB), which can accomplish feature-level fusion by exploiting a PWA block. The FFB and PWA constitute our AMLFF. We design an AMLFF, which can integrate three different levels of feature maps to effectively balance the weights of the inputs to the encoder and decoder. We also utilize the contrastive loss function to train the dehazing network so that the recovered image is far from the negative sample and close to the positive sample. Experimental results on both synthetic and real-world images demonstrate that this dehazing approach surpasses numerous other advanced techniques, both visually and quantitatively, showcasing its superiority in image dehazing.