Single image deraining via decorrelating the rain streaks and background scene in gradient domain

Du, Shuangli; Liu, Yiguang; Ye, Mao; Xu, Zhenyu; Li, Jie; Liu, Jianguo

doi:10.1016/j.patcog.2018.02.016

Cited by 43 publications

(23 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When larger inclination angles are encountered, the proposed approach may still work if additional preprocessing strategies (e.g., rotating or shearing the image matrix at an angle to force the rain streaks to appear vertical) are adopted. Different methods for determining the angles of rain streaks have been proposed (Du et al, ; Jiang et al, ). It is worth mentioning that wind effects are also a serious issue in rainfall measurement by traditional rain gauges.…”

Section: The Effectiveness and Efficiency Of Rain‐streak Identificationmentioning

confidence: 99%

Advancing Opportunistic Sensing in Hydrology: A Novel Approach to Measuring Rainfall With Ordinary Surveillance Cameras

Jiang

Babovic

Zheng

et al. 2019

Water Resources Research

View full text Add to dashboard Cite

“Opportunistic sensing” represents an appealing idea for collecting unconventional data with broad spatial coverage and high resolution, but few studies have explored its feasibility in hydrology. This study develops a novel approach to measuring rainfall intensity in real‐world conditions based on videos acquired by ordinary surveillance cameras. The proposed approach employs a convex optimization algorithm to effectively decompose a rainy image into two layers: a pure rain‐streak layer and a rain‐free background layer, where the rain streaks represent the motion blur of falling raindrops. Then, it estimates the instantaneous rainfall intensity via geometrical optics and photographic analyses. We investigated the effectiveness and robustness of our approach through synthetic numerical experiments and field tests. The major findings are as follows. First, the decomposition‐based identification algorithm can effectively recognize rain streaks from complex backgrounds with many disturbances. Compared to existing algorithms that consider only the temporal changes in grayscale between frames, the new algorithm successfully prevents false identifications by considering the intrinsic visual properties of rain streaks. Second, the proposed approach demonstrates satisfactory estimation accuracy and is robust across a wide range of rainfall intensities. The proposed approach has a mean absolute percentage error of 21.8%, which is significantly lower than those of existing approaches reported in the literature even though our approach was applied to a more complicated scene acquired using a lower‐quality device. Overall, the proposed low‐cost, high‐accuracy approach to vision‐based rain gauging significantly enhances the possibility of using existing surveillance camera networks to perform opportunistic hydrology sensing.

show abstract

Section: The Effectiveness and Efficiency Of Rain‐streak Identificationmentioning

confidence: 99%

Advancing Opportunistic Sensing in Hydrology: A Novel Approach to Measuring Rainfall With Ordinary Surveillance Cameras

Jiang

Babovic

Zheng

et al. 2019

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…In general, dictionary learning-based methods are highly effective at removing rain from single images while preserving detail since they are mostly based on the processing of local features [4,12,20,21,22]. This sometimes results in some degree of detail loss when dealing with images with elements similar to rain streaks, since some non-rain features might be treated as rain [22].…”

Section: Figurementioning

confidence: 99%

“…Furthermore, processing times are often elevated and the de-raining pipeline often includes spatial and frequency domain-based algorithms which add to the computational costs [4,12]. However, unlike such methods, most dictionary learning-based proposals are evaluated with quantitative metrics such as SSIM and PSNR [4,20,21,22]. In addition, they do not require sizable datasets, unlike the machine learning approach, although this results in a lower capability for generalizing to cases such as images taken under heavy rain conditions [4,20,22].…”

Section: Figurementioning

confidence: 99%

Untitled

2020

Investig. innov. ing

View full text Add to dashboard Cite

Objective: To use computational techniques for removing rain from images. This is motivated by the fact that for many computer vision systems, correctly capturing the scene is a key need, and if such systems receive images degraded by rain as input, their performance may be compromised. Methodology: We built a dataset comprised by 11000 synthetic rain images. We resized and normalized all the images, then we employed 9000 of them for training the autoencoder architecture. The autoencoder outputs a de-rained version of the image which is then lighting corrected in order to produce the final, de-rained image. Results: We determined the best autoencoder architecture was a 6-layer autoencoder. We evaluated it on the remaining 2000 images, resulting in a Mean Squared Error of 0.61 and Structural Similarity Index of 0.8493, which means a fair amount of information from the rain-degraded images was recovered. Conclusions: The results we obtained were superior to proposals based in the spatial / frequency domain reported in the literature. However, we determined that it is possible to improve on the current results if we consider the frequency domain as part of the architecture. Thus, options for future work include combining machine learning-based approaches with frequency domain-based image processing.

show abstract

“…However, the outdoor weather is complex and changeable, especially the severe rain and snow weather. It seriously affects the image quality collected by the vision system, which can mislead the outdoor workers with inaccurate information [1,2]. erefore, it is necessary to solve the problem by removing rain from rainy images [3].…”

Section: Introductionmentioning

confidence: 99%

Research of Single Image Rain Removal Algorithm Based on LBP-CGAN Rain Generation Method

Xue

2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Rain has an undesirable negative effect on the clarity of the collected images. In situation where images are captured in rain, it can lead to a loss of information and disability in reflecting real images of the situation. Consequently, rain has become an obstacle in outdoor scientific research studies. The reason why images captured in rain are difficult to process is due to the indistinguishable interactions between the background features and rain textures. Since current image data are only processed with the CNN (convolutional neural network) model, a trained neural network to remove rain and obtain clear images, the resulted images are either insufficient or excessive from standard results. In order to achieve more ideal results of clearer images, series of additional methods are taken place. Firstly, the LBP (local binary pattern) method is used to extract the texture features of rain in the image. Then, the CGAN (conditional generative adversarial network) model is constructed to generate rain datasets according to the extracted rain characteristics. Finally, the existing clear images, rain datasets generated by CGAN, as well as the images with rain are used for convolution operation to remove rain from the images, and the average value of PSNR (peak signal to noise ratio) can reach 38.79 by using this algorithm. Moreover, a large number of experiments are done and have proven that this joint processing method is able to successfully and effectively generate clear images despite the rain.

show abstract

Single image deraining via decorrelating the rain streaks and background scene in gradient domain

Cited by 43 publications

References 32 publications

Advancing Opportunistic Sensing in Hydrology: A Novel Approach to Measuring Rainfall With Ordinary Surveillance Cameras

Advancing Opportunistic Sensing in Hydrology: A Novel Approach to Measuring Rainfall With Ordinary Surveillance Cameras

Untitled

Research of Single Image Rain Removal Algorithm Based on LBP-CGAN Rain Generation Method

Contact Info

Product

Resources

About