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

Image denoising method based on edge enhancement and convolutional neural network

A convolutional neural network and edge enhancement technology, which is applied in the field of image processing and can solve the problems of insignificant denoising effect and unclear edge texture of denoising results.

Pending Publication Date: 2021-06-08
XIAN UNIV OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide an image denoising method based on edge enhancement and convolutional neural network, which solves the problems in the prior art that the denoising effect is not obvious and the edge texture of the denoising result is not clear

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
  • Image denoising method based on edge enhancement and convolutional neural network
  • Image denoising method based on edge enhancement and convolutional neural network
  • Image denoising method based on edge enhancement and convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0038] The present invention is based on the image denoising method of edge enhancement and convolutional neural network, specifically implements according to the following steps: as figure 1 as shown,

[0039] Step 1, make the noise image data set to be trained, select n images of different scenes from the commonly used image processing data set, denoted as I(x), x=1,2,3...n, put Each selected image I(x) is cut into a fixed size of 128×128 and a certain level of Gaussian noise is added to obtain the noise image I'(x). First, the NLM (Non-local Means, non-local mean filter) denoising method is used Pre-denoise the noise image I'(x) to obtain the result M, and then perform canny edge detection on the result M to obtain the corresponding edge matrix K for backup;

[0040] Step 2, still perform parallel operation on the noise image sample in s...

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 invention discloses an image denoising method based on edge enhancement and a convolutional neural network. The method specifically comprises the following steps: step 1, selecting n images as training samples, and processing each image in the training samples; step 2, performing NSCT on the image processed in the step 1 to obtain a low-frequency sub-band graph L and a high-frequency sub-band graph; step 3, positioning the edge of the high-frequency sub-band graph J in the step 2, and performing NSCT inverse transformation on the positioned high-frequency direction sub-band graph and the low-frequency sub-band graph obtained in the step 2 to obtain a final noise image X to be trained after edge enhancement; and step 4, inputting the edge-enhanced noise image obtained in the step 3 into a residual denoising network for training and learning to obtain a final denoised clean image. According to the method, the problems that the denoising effect is not obvious and the edge texture of a denoising result is not clear in the prior art are solved.

Description

technical field [0001] The invention belongs to the technical field of image processing methods, and relates to an image denoising method based on edge enhancement and convolutional neural network. Background technique [0002] In daily life, images are often disturbed and affected by various noises during the collection and transmission process, making the image blurred, even covering up the details of the image, and bringing certain difficulties to the further processing of the image, such as image segmentation, Image classification, object recognition, etc. [0003] Denoising is the process of reconstructing the original image from a corrupted image by removing unnecessary noise. Its purpose is to suppress noise while preserving as much structure and edge texture information of the image as possible. It is critical for applications ranging from camera imaging and medical imaging to video surveillance image processing. Therefore, suppressing these noises and improving i...

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
IPC IPC(8): G06T5/00G06T7/13G06N3/04G06K9/46G06N3/08
CPCG06T7/13G06N3/08G06T2207/20192G06V10/44G06N3/045G06T5/70Y02T10/40
Inventor 刘晶董玉田冲
Owner XIAN UNIV OF TECH