Single Image Dehazing via NIN-DehazeNet

Yuan, Kangle; Wei, Jianguo; Lu, Wenhuan; Xiong, Naixue

doi:10.1109/access.2019.2958607

Cited by 20 publications

(14 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the advantages of the proposed method (CISACC) over other image dehazing techniques developed by different researchers, a features comparison of all existing methods has been given in Table 6. Some current methods include dark channel prior (DCP), 33 CLAHE, 11,12,15,22 AOD‐Net, 20 FFA‐Net, 21 underwater image enhancement, 36 pyramid dehazing network, 37 NIN‐DehazeNet, 38 and so forth.…”

Section: Performance Analysis and Discussionmentioning

confidence: 99%

Artificial intelligent based smart system for safe mining during foggy weather

Kumari

Choudhary

Mishra

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

Opencast mining operations at hilly areas are usually affected during foggy weather due to the inability of drivers to operate heavy earth‐moving machinery in low visibility conditions. This article deals with an intelligent vision enhancement system for continuing opencast mining operations during foggy weather. The system integrates hardware and software to provide multistage safety features that make it unique from existing systems. The system includes hardware like thermal cameras, high definition cameras, proximity radar, wireless devices, GNSS module, graphical processing unit, display unit, and so forth, and image processing software, namely real‐time image stitching, image enhancement, and object detection using convolutional neural networks. The integrated system and algorithms display a 180° panorama field view of the vehicle's front using real‐time video stitching. The front view after image processing, rear camera view, object detection through proximity radar, and real‐time location of the vehicles on a 3D geo‐tagged mine map by GNSS modules are displayed in four splitter windows on a touch screen fitted on the dashboard in front of the driver's seat. The driver can drive the vehicle by seeing the display screen during foggy weather. The output image of the developed image‐processing algorithm has less distortion, better quality, and better depth perception than existing methods. Overall, there are significant improvements in the persistence of the color elements by 39.65%, contrast by 4.62%, and the corresponding entropy by 7.11% concerning the similar existing methods. The final system has been successfully tested in an opencast mine.

show abstract

Section: Performance Analysis and Discussionmentioning

confidence: 99%

Artificial intelligent based smart system for safe mining during foggy weather

Kumari

Choudhary

Mishra

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…In the early stage, people built various features (such as dark channel [3], color attenuation [4], ICA [5], contrast [6], etc.) to assist in estimating transmission map [7]. Later, with the development of artificial intelligence, related methods based on machine learning and deep learning were proposed to estimate transmission map.…”

Section: Introductionmentioning

confidence: 99%

CAA-Net: End-to-End Two-Branch Feature Attention Network for Single Image Dehazing

JIN

Zhai

Wei

2023

IEICE Trans. Fundamentals

Self Cite

View full text Add to dashboard Cite

In this paper, we propose an end-to-end two-branch feature attention network. The network is mainly used for single image dehazing. The network consists of two branches, we call it CAA-Net: 1) A U-NET network composed of different-level feature fusion based on attention (FEPA) structure and residual dense block (RDB). In order to make full use of all the hierarchical features of the image, we use RDB. RDB contains dense connected layers and local feature fusion with local residual learning.We also propose a structure which called FEPA.FEPA structure could retain the information of shallow layer and transfer it to the deep layer.FEPA is composed of serveral feature attention modules (FPA). FPA combines local residual learning with channel attention mechanism and pixel attention mechanism, and could extract features from different channels and image pixels. 2) A network composed of several different levels of FEPA structures.The network could make feature weights learn from FPA adaptively, and give more weight to important features. The final output result of CAA-Net is the combination of all branch prediction results. Experimental results show that the CAA -Net proposed by us surpasses the most advanced algorithms before for single image dehazing. key words: End-to-end,two-branch feature attention network,single image dehazing,residual dense block,all branch prediction results.

show abstract

“…In addition to conventional computer vision approach to solve the dehazing problem, numerous methods based on machine and deep learning have been proposed [30] [31] [32] [33]. For instance, the methods [30] and [31] dehaze a hazy image using multilayer perceptrons, and the method [31] dehazes images using a trainable end-to-end multiscale convolutional neural network. Similarly, the approach presented in [32] using NIN-DehazeNet combining Networkin-Network with MSCNN (Single Image Dehazing via Multi-Scale Convolutional Neural Networks) to estimate transmission map.…”

Section: Introductionmentioning

confidence: 99%

A Single Image Dehazing Technique Using the Dual Transmission Maps Strategy and Gradient-Domain Guided Image Filtering

et al. 2021

View full text Add to dashboard Cite

In this paper, a single image dehazing technique using dual transmission maps strategy and gradient domain guided image filtering is presented. A new strategy is adopted to compute the dual transmission maps using the dark channel and atmospheric light. Further, the transmission maps are refined to remove any remaining ill effects using the gradient-domain-guided filter. Finally, using the dark channel, atmospheric light, and refined transmission map, the haze-free image is obtained. The dual transmission maps strategy not only removes halo artifacts and reduces the saturation but also ensures the natural appearance in the recovered images. Furthermore, the proposed scheme is evaluated using a wide range of images and compared with state-of-the-art schemes. The comparison shows the superiority of the proposed technique in terms of recovering haze-free images.INDEX TERMS Image de-hazing, transmission map, gradient-domain guided image filter

show abstract

Single Image Dehazing via NIN-DehazeNet

Cited by 20 publications

References 39 publications

Artificial intelligent based smart system for safe mining during foggy weather

Artificial intelligent based smart system for safe mining during foggy weather

CAA-Net: End-to-End Two-Branch Feature Attention Network for Single Image Dehazing

A Single Image Dehazing Technique Using the Dual Transmission Maps Strategy and Gradient-Domain Guided Image Filtering

Contact Info

Product

Resources

About