An end-to-end image defogging processing method based on deep learning

Abstract

The invention relates to an end-to-end image defogging processing method based on deep learning. According to the method, a foggy image is converted into a non-foggy image through a trained deep convolutional neural network, and the deep convolutional neural network comprises a feature extraction module which comprises a plurality of convolution sub-modules and is used for carrying out convolutioncalculation on an input image and extracting a multi-dimensional feature map; the feature pooling module comprises a plurality of pooling layers, and after each pooling layer is correspondingly connected to one convolution sub-module, redundancy removal processing is carried out on the multi-dimensional feature map; the recovery module comprises a plurality of deconvolution sub-modules, is connected to the feature pooling module and then outputs an output image with the same resolution as the input image; and a plurality of inter-layer jump connection layers which are used for realizing inter-layer jump connection between the output of the pooling layer and the input of the corresponding deconvolution sub-module and fusing the multi-scale feature map. Compared with the prior art, the method has the advantages of good defogging effect, simple process and the like.

Description

technical field [0001] The present invention relates to an image processing method, in particular to an end-to-end image defogging processing method based on deep learning. Background technique [0002] Fog is a common atmospheric phenomenon over land and sea. In foggy weather there are many atmospheric microscopic particles of a certain size. They not only absorb the reflected light of the target object / scene, but also their own reflected light enters the camera together with the reflected light of the target object, which interferes with the light information acquired by the camera and makes it impossible to clearly image the target object / scene. Due to the blur and noise of imaging, it brings great difficulties and challenges to the performance of various algorithms based on computer vision, such as target recognition / tracking, scene segmentation, automatic driving, etc. [0003] With the development of image processing technology, image dehazing has received extensive ...

AI Technical Summary

Problems solved by technology

[0004] At present, the traditional image defogging processing methods have great deficiencies in the restoration accuracy and universality, and most of the existing deep learning-based methods have not achieved effective end-to-end image defogging, and need to pass the estimated transmittance and Atmospheric light intensity for postprocessing

Images