Color image denoising method based on bit layering and attention fusion mechanism

Abstract

The invention relates to a color image denoising method based on a bit layering and attention fusion mechanism. The method comprises the steps: 1, carrying out the bit layering of a color image with noise, and dividing the color image into 24 channels; 2, inputting the image subjected to bit layering into a noise estimation network, and adding a channel attention module to each layer of the network to obtain noise estimation of the noisy color image; 3, splicing the images subjected to noise estimation and bit layering, inputting the spliced images into a coding network, adding a channel attention module into each layer of the network, and endowing different weight values to the characteristics of different channels; 4, inputting the output of the coding network into the decoding network,splicing the output of each layer in the network with the corresponding layer of the spatial attention module and the decoding network, and performing convolution and deconvolution operation; and step5, sending the output of the decoding network to a convolution layer, and finally outputting a denoised clean image.

Description

technical field [0001] The invention relates to the technical field of computer vision images, in particular to a color image denoising method based on a bit layering and attention fusion mechanism. Background technique [0002] Image is an indispensable information carrier in daily life and plays an important role in the process of people acquiring, storing and transmitting information. With the continuous development of digital multimedia technology, computer images have been widely used in medical imaging, pattern recognition, and target detection. However, in the process of image acquisition, transmission and storage, it is inevitable to be disturbed by various noises. Image noise refers to unnecessary or redundant interference information existing in image data. People refer to various redundant signals that affect people's reception of image information as noise signals. Image noise blurs the image, and sometimes even conceals image features, and at the same time adv...

AI Technical Summary

Problems solved by technology

However, the above deep learning method does not take into account: the difference of high and low frequency information storage data in the image and the influence of the noise value in the image on the denoising effect. Therefore, the above deep learning based method has achieved relatively good results in image denoising. limited effect

Images