Unlock instant, AI-driven research and patent intelligence for your innovation.

A denoising method for underwater images based on generative confrontation network

An underwater image and network technology, applied in biological neural network models, image enhancement, image analysis, etc., can solve the problems of blurred image edge outline, reduced effectiveness, loss of detail information, etc., to achieve improved denoising effect and easy production , good denoising effect

Active Publication Date: 2022-03-11
NORTHWESTERN POLYTECHNICAL UNIV
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] This method has a simple model and can effectively remove noise with known attributes, but for noise with unknown underwater attributes, the effectiveness of the method will be greatly reduced. serious loss of information

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A denoising method for underwater images based on generative confrontation network
  • A denoising method for underwater images based on generative confrontation network
  • A denoising method for underwater images based on generative confrontation network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0029] In order to overcome the problems that the underwater image noise is difficult to remove and the edge texture cannot be enhanced, the present invention proposes an underwater image denoising method based on a generative adversarial network. The method first inputs the retina-enhanced underwater image with noise into a generative network composed of several residual blocks, and obtains a feature map with three-channel (r, g, b three-channel) output; then the output The obtained feature map and the noise-free label image of Shimizu are respectively mapped to a deep feature space through the VGG-19 network (the network has been proposed and publicly used by Google), and the feature map and the noise-free label image of Shimizu are calculated in The perceptual cost in the deep feature space makes the feature map output by the generative network as cl...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention provides an underwater image denoising method based on generating confrontation networks. First, the underwater image is input into a generating network composed of several residual blocks to obtain a feature map; then the output feature map is combined with Qingshui's noise-free label images are respectively mapped to a deep feature space through the VGG-19 network, and the perceptual cost of the feature map and Qingshui's noise-free label images in the depth feature space is calculated. When the perceptual cost is calculated, the network output will be generated The feature map of the input to the confrontation network, and finally after the training is completed, the noisy underwater image is input to the generation network, and the output is the processed noise-free image. The present invention has obvious denoising effect by introducing the confrontation mechanism, especially the method can effectively preserve or even enhance the edge texture information in the image, and has better visual effect and imaging quality.

Description

technical field [0001] The invention relates to the field of underwater image processing, in particular to an underwater image denoising method. Background technique [0002] Underwater images play an important role in the development and exploration of marine resources, and are mainly used in the following aspects: (1) In the military, it can detect and identify underwater targets, realize underwater high-definition imaging, and facilitate the search and salvage of sunken ships and aircraft wreckage. (2) In terms of environmental protection, it can monitor the migration of marine species and changes in the marine ecological environment; (3) In terms of engineering, it can monitor the construction of submarine projects, realize the docking of deep-sea workstations, and facilitate the follow-up automatic maintenance of the project. Compared to imaging in the air, the environment for capturing optical images in water is more complex. Due to the scattering effects of planktoni...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/00G06N3/04
CPCG06T2207/10004G06T2207/20192G06T2207/20081G06N3/045G06T5/70
Inventor 冯晓毅蒋晓悦夏召强李磊黄东张晓彪
Owner NORTHWESTERN POLYTECHNICAL UNIV