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

Image denoising method based on convolution templates

A convolution template and image technology, applied in the field of image processing, can solve problems such as unfavorable hardware implementation, poor weakening effect of isolated noise points, and complicated calculation process.

Active Publication Date: 2018-11-06
XIAN UNIV OF SCI & TECH
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, using a bilateral filter to denoise, although the edge of the image can be preserved, it does not fit the actual image edge gradient at the upper and lower junctions of the edge, the weakening effect on isolated noise points is not good, and the calculation process of bilateral filtering is complicated. Not conducive to hardware implementation

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 convolution templates
  • Image denoising method based on convolution templates
  • Image denoising method based on convolution templates

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] See figure 1 , figure 1 It is a schematic flowchart of an image denoising method based on a convolution template provided by an embodiment of the present invention. The image denoising method based on the convolution template includes:

[0044] S1. Obtain a preprocessed image by preprocessing the original image;

[0045] S2. Obtain an output image by performing convolution and denoising on the preprocessed image;

[0046] S3. Obtain a denoised image by correcting the output image.

[0047] Wherein, for step S1, may include:

[0048] S11. Perform mirror extension on the outermost circle data of the original image to obtain a preprocessed image.

[0049] Wherein, for step S2, may include:

[0050] S21. Calculate the weight coefficient;

[0051] S22. Obtain a filtered image by using the filtering template and the preprocessed image;

[0052] S23. Acquire the output image according to the weight coefficient and the filtered image.

[0053] Wherein, for step S21, ma...

Embodiment 2

[0075] Please refer to figure 2 , figure 2 It is a schematic flowchart of another image denoising method based on a convolution template provided by an embodiment of the present invention. This embodiment further describes the image denoising method in detail on the basis of the above embodiments, wherein the image denoising method takes a preset template with a size of 3*3 as an example, and specifically includes the following steps:

[0076] Step 1: Perform mirror expansion on the outermost circle data of the original image;

[0077] The original image containing noise is recorded as I, and the outermost circle data (the first row, the first column, the last row, and the last column) of the original image are mirrored and extended to obtain a preprocessed image, which is recorded as I_input.

[0078] Step 2: Extract edge information of the preprocessed image;

[0079] Convolute with I_input through four preset templates to extract the edge information of the preprocessed ...

Embodiment 3

[0122] Please continue to see figure 2 , figure 2 It is a schematic flowchart of another image denoising method based on a convolution template provided by an embodiment of the present invention. This embodiment further describes the image denoising method in detail on the basis of the above embodiments, wherein the image denoising method takes a preset template with a size of 5*5 as an example, and specifically includes the following steps:

[0123] Step 1: Perform mirror expansion on the outermost circle data of the original image;

[0124] The original image containing noise is recorded as I, and the outermost circle data (the first row, the first column, the last row, and the last column) of the original image are mirrored and extended to obtain a preprocessed image, which is recorded as I_input.

[0125] Step 2: Extract edge information of the preprocessed image;

[0126] Convolute with I_input through eight preset templates to extract the edge information of the pre...

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 relates to an image denoising method based on convolution templates. The method comprises the steps that an original image is preprocessed to acquire a preprocessed image; convolution denoising is performed on the preprocessed image to acquire an output image; and the output image is corrected to acquire a denoised image. According to the embodiment, feature information of the imageis extracted by performing convolution on the image through the convolution templates; and in the filtering denoising process, edge information is reserved more completely, the process is completed through convolution operation without complicated formula calculation, and therefore the method is overall simple and facilitates hardware implementation.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to an image denoising method based on a convolution template. Background technique [0002] Image denoising refers to the process of reducing noise in an image. In the process of image transmission and acquisition, often due to factors such as working environment conditions, the image is disturbed by noise, so that part of the information of the image is destroyed, and the information that humans can extract from the image is also limited. These noises may be generated in transmission, or in processing such as quantization. An image may have various noises in practical applications, and noise is an important cause of image interference. [0003] The widely used denoising method is bilateral filter denoising. The bilateral filter uses the product of the Gaussian kernel function of the Euclidean distance and the Gaussian kernel function of the pixel value difference as t...

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 Applications(China)
IPC IPC(8): G06T5/00G06T7/13
CPCG06T7/13G06T5/70
Inventor 周筱媛
Owner XIAN UNIV OF SCI & TECH