Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing
Image dehazing represents a dynamic area of research in computer vision. With the exponential development of deep learning, particularly convolutional neural networks (CNNs), innovative and effective image dehazing techniques have surfaced. However, in stark contrast to the majority of computer visi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/24/4984 |
| _version_ | 1797381321952067584 |
|---|---|
| author | Haoyang Yu Xiqin Yuan Ruofei Jiang Huamin Feng Jiaxing Liu Zhongyu Li |
| author_facet | Haoyang Yu Xiqin Yuan Ruofei Jiang Huamin Feng Jiaxing Liu Zhongyu Li |
| author_sort | Haoyang Yu |
| collection | DOAJ |
| description | Image dehazing represents a dynamic area of research in computer vision. With the exponential development of deep learning, particularly convolutional neural networks (CNNs), innovative and effective image dehazing techniques have surfaced. However, in stark contrast to the majority of computer vision tasks employing CNNs, the output from a dehazing model is often treated as uninformative noise, even though the model’s filters are engineered to extract pertinent features from the images. The standard approach of end-to-end models for dehazing involves noise removal from the hazy image to obtain a clear one. Consequently, the model’s dehazing capacity diminishes, as the noise is progressively filtered out throughout the propagation phase. This leads to the conception of the feature reduction network (FRNet), which is a distinctive CNN architecture that incrementally eliminates informative features, thereby resulting in the output of noise. Our experimental results indicate that the CNN-driven FRNet surpasses previous state-of-the-art (SOTA) methods in terms of the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) evaluation metrics. This highlights the effectiveness of the FRNet across various image dehazing datasets. With its reduced overhead, the CNN-based FRNet demonstrates superior performance over current SOTA methods, thereby affirming the efficacy of CNNs in image dehazing tasks. |
| first_indexed | 2024-03-08T20:49:46Z |
| format | Article |
| id | doaj.art-66ab60dc04464f56a9d897045ab62d50 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-08T20:49:46Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-66ab60dc04464f56a9d897045ab62d502023-12-22T14:05:09ZengMDPI AGElectronics2079-92922023-12-011224498410.3390/electronics12244984Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image DehazingHaoyang Yu0Xiqin Yuan1Ruofei Jiang2Huamin Feng3Jiaxing Liu4Zhongyu Li5School of Cyberspace Security, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaBeijing Electronic Science and Technology Institute, Beijing 100070, ChinaInstitute of Electronic Computing Technology, China Academy of Railway Science Co., Ltd., Beijing 100081, ChinaBeijing Electronic Science and Technology Institute, Beijing 100070, ChinaBeijing Electronic Science and Technology Institute, Beijing 100070, ChinaSchool of Cyber Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaImage dehazing represents a dynamic area of research in computer vision. With the exponential development of deep learning, particularly convolutional neural networks (CNNs), innovative and effective image dehazing techniques have surfaced. However, in stark contrast to the majority of computer vision tasks employing CNNs, the output from a dehazing model is often treated as uninformative noise, even though the model’s filters are engineered to extract pertinent features from the images. The standard approach of end-to-end models for dehazing involves noise removal from the hazy image to obtain a clear one. Consequently, the model’s dehazing capacity diminishes, as the noise is progressively filtered out throughout the propagation phase. This leads to the conception of the feature reduction network (FRNet), which is a distinctive CNN architecture that incrementally eliminates informative features, thereby resulting in the output of noise. Our experimental results indicate that the CNN-driven FRNet surpasses previous state-of-the-art (SOTA) methods in terms of the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) evaluation metrics. This highlights the effectiveness of the FRNet across various image dehazing datasets. With its reduced overhead, the CNN-based FRNet demonstrates superior performance over current SOTA methods, thereby affirming the efficacy of CNNs in image dehazing tasks.https://www.mdpi.com/2079-9292/12/24/4984artificial intelligenceimage dehazingcomputer visionCNNsdeep learning |
| spellingShingle | Haoyang Yu Xiqin Yuan Ruofei Jiang Huamin Feng Jiaxing Liu Zhongyu Li Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing Electronics artificial intelligence image dehazing computer vision CNNs deep learning |
| title | Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing |
| title_full | Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing |
| title_fullStr | Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing |
| title_full_unstemmed | Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing |
| title_short | Feature Reduction Networks: A Covolution Neural Network-Based Approach to Enhance Image Dehazing |
| title_sort | feature reduction networks a covolution neural network based approach to enhance image dehazing |
| topic | artificial intelligence image dehazing computer vision CNNs deep learning |
| url | https://www.mdpi.com/2079-9292/12/24/4984 |
| work_keys_str_mv | AT haoyangyu featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing AT xiqinyuan featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing AT ruofeijiang featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing AT huaminfeng featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing AT jiaxingliu featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing AT zhongyuli featurereductionnetworksacovolutionneuralnetworkbasedapproachtoenhanceimagedehazing |