Single picture bidirectional rain removing method based on picture block raindrop density

Abstract

The invention discloses a single picture bidirectional rain removing method based on picture block raindrop density. The method can remove rain from a picture with non-uniform raindrop distribution. The method remove rain of picture blocks by different degrees, according to raindrop intensities of different areas of a single picture predicted by a classifier, and excessive rain removal and incomplete rain removal are prevented. A plurality of picture block bidirectional rain removing modules are trained to remove rain from picture blocks with different sizes, and reinforcing the relationship between the encoder and the generator through the generated hidden layer codes to ensure that the generator can remove rain according to the raindrop intensity. A repairing module is used for repairingthe spliced image subjected to rain removal according to the unsmooth and distorted phenomena. Considering that dense raindrops generate an atomization phenomenon in a rainy day picture, a defoggingmodule is added to a defogging network. Rain removal is carried out on pictures which are distributed unevenly in rainy days, heavy rain pictures are effectively removed through a circulating structure, and meanwhile the atomization phenomenon generated in the rainy days is removed.

Description

technical field [0001] The present invention relates to computer vision technology, in particular to deep learning and neural network rain removal technology, specifically a bidirectional rain removal method for a single picture based on the raindrop density of a picture block. Background technique [0002] In rainy days, since the falling raindrops are relatively dense, and the falling raindrops are reflected by light, the visibility of the background is reduced, and the quality of the picture is seriously affected. It can be observed that in the images taken in rainy days, the density, coverage, direction and shape of raindrops are all different, which makes the pictures taken in rainy days blurred and deformed. Dense raindrops can also create fog, making rainy days blurry. In different vision applications, falling raindrops will affect the effect of monitoring or recognition. [0003] The raindrop removal algorithm can be divided into video raindrop removal algorithm an...

AI Technical Summary

Problems solved by technology

[0006] However, in real life, due to the occlusion of some objects, the density of raindrops also changes for a whole image of raindrops. De-raining a picture with a change in the density of raindrops will result in excessive or dense raindrop areas in the image. Cases that cannot be completely removed

[0007] Deng Junping and others divided the picture into 64*64 picture blocks for rain removal, and used the depth denoising encoder to remove the rain on the distribution of the divided picture blocks, failing to consider the raindrop density of different rainy day picture blocks and the picture blocks after rain removal Unsmooth and distorted after splicing, while failing to consider the fogging phenomenon caused by raindrops

Images