Image deraining method and device

Abstract

The present invention provides an image deraining method and device, comprising: separating the image to be detected with raindrops into a high-frequency component image and a low-frequency component image; inputting the high-frequency component image into a trained convolution with a residual structure Neural network, output the rain-free image, combine the rain-free image with the image to be detected to obtain a rain-free image; input the rain-free image to the trained convolutional neural network with a discriminative structure, and output the rain-free image corresponding to the image to be detected The category of the rain-free image comparison, input the category into the convolutional neural network with residual structure, and update the parameters of the convolutional neural network with residual structure to obtain the final rain-free image. The final rain-free image obtained by the present invention retains the texture details of the rain-free region in the image to be detected, so that the final rain-free image is close to the image without raindrops corresponding to the image to be detected, and well retains the raindrops in the image to be detected. Key factor.

Description

technical field [0001] The invention relates to the technical fields of image processing and machine learning, in particular to a method and device for removing rain from an image. Background technique [0002] With the rapid development of machine learning technology and the maturity of image processing technology, the outdoor computer vision system can obtain a large amount of rich and high-resolution image information in time due to its accuracy, speed, reliability, intuitive, real-time and comprehensive reflection of the monitored object, etc. Advantages, especially in occasions that are not easy to be directly observed by humans, it can solve the problem of manual observation, and is widely used in various fields, such as car driverless or remote monitoring systems. However, in real life, in rainy and snowy weather Under the influence of light and refraction, objects in the image are easily blurred and blocked by separate rain marks, so rain and snow have a great impact...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the current image deraining method based on deep learning will smooth the texture details of the rainless area in the rain image, and at the same time cannot cover some important factors in the real rain image well, the present invention provides an image deraining method methods, including:

Images