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

Scratch detection method, device and electronic equipment based on image processing

An image processing and scratch technology, applied in the field of image recognition, can solve the problems of identifying weak scratches, it is difficult to segment the scratch area, and the filter template is difficult to satisfy information correlation and information accumulation, so as to improve reliability and effectiveness sexual effect

Active Publication Date: 2019-10-08
NAT UNIV OF DEFENSE TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, in the process of realizing the present invention, the inventors found that in the prior art, for products with rough surfaces such as nickel foam, the gray scale of the material image pixels itself has great fluctuations due to the surface roughness of the material. Therefore, it is generally difficult to segment the scratch area through the gray distribution of pixels, and it is usually only effective in the case of severe scratches.
In addition, the vast majority of scratches may be discrete points or lines, that is, the continuous situation of the scratch area is far less than the intermittent situation, so the combined application of standardized filter templates is difficult to meet the special needs of information correlation and information accumulation. need
Therefore, it is difficult to effectively identify weak scratches through existing image processing technology when detecting scratches on a surface with uniform roughness

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
  • Scratch detection method, device and electronic equipment based on image processing
  • Scratch detection method, device and electronic equipment based on image processing
  • Scratch detection method, device and electronic equipment based on image processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] Since it is difficult to accurately identify minor scratches on rough surfaces by simply using image pixel information, the embodiments of the present invention begin to use information other than pixels, and the most intuitive one is to use the spatial distribution characteristics of scratches to increase the possibility of scratch extraction. Specifically, such as figure 2 As shown, the embodiment of the present invention provides a scratch detection method based on image processing, including:

[0064] S1. Obtain a background image, and use the background image to cancel the original image to obtain a difference image;

[0065] S2. Using a directional filter bank to process the pixels to be analyzed in the difference image to determine suspected points and corresponding suspected directions;

[0066] S3. Using the suspected points and the suspected directions, perform spatial accumulation processing in the mapping area of ​​the directional filter bank to obtain an ...

Embodiment 2

[0069] Objects with rough surfaces are usually affected by factors such as uneven surfaces and / or uneven illumination, and the image grayscale itself may have large fluctuations. Therefore, it is difficult to implement image segmentation in the prior art only by image grayscale. In one embodiment of the present invention, a difference image with relatively consistent gray distribution is obtained by canceling the background image and the original image, thereby eliminating image inhomogeneity and enhancing the reliability and effectiveness of image segmentation.

[0070] Specifically, such as image 3 As shown, in an implementation of the present invention, step S1 of the above method (that is, obtaining a background image, and using the background image to cancel the original image to obtain a differential image) includes:

[0071] performing longitudinal mean filtering and transverse mean filtering on the original image to obtain the background image;

[0072] performing di...

Embodiment 3

[0077] For the image with weak scratches on the rough surface, on the one hand, the gray level difference between the pixels in the scratched area and the pixels in the non-scratched area is very small; on the other hand, due to the influence of the rough surface, the gray level of the pixels in the scratched area fluctuates greatly Therefore, relying solely on the gray value of a single pixel for scratch region segmentation not only cannot effectively segment the pixels in the scratch, but also may cause serious fractures, and the accuracy of scratch detection is extremely low.

[0078] In one embodiment of the present invention, the image is filtered through a directional filter bank by using the linear characteristics of the scratches, and the linear directional filtering process can effectively enhance the grayscale of the pixels in the direction of the scratches. In order to improve the distinguishability of the gray level of pixels inside and outside the scratch area, it ...

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 embodiments of the invention provide a scratch detection method, apparatus and electronic device based on image processing. The method includes the following steps: acquiring a background image, offsetting the background image and the original image to obtain a differential image; using a directional filter bank to process to-be-analyzed pixel in the differential image, determining suspicious points and corresponding suspicious directions; and using the suspicious points and the suspicious directions to conduct air space accumulation processing on a mapping region of the directional filter bank, obtaining an enhanced scratch detection image. According to the embodiments of the invention, the method uses the characteristics of spatial distribution of scratches, which refers to the fact the scratch often are distributed in linear shape to some degree, enhances the appearance of the scratched region in an image, such that the technical solution of the embodiments of the invention can greatly increase the reliability and effectiveness of scratch detection.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a scratch detection method, device and electronic equipment based on image processing. Background technique [0002] In the process of industrial automation, it has become a trend for machines to replace manpower to complete various tasks in manufacturing. How to improve the reliability and efficiency of machine processing has increasingly become the key to the development of industrial automation. For example, for the nickel foam substrate, a key raw material required in the manufacture of some new energy vehicle batteries, during the production process, due to the possible adhesion of hard particles on the rollers of the processing equipment, the nickel foam material will be scratched, while When the scratch is serious enough, it will bring potential safety hazards to the battery; in order to avoid such problems, material surface defect detection must be carried out 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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06T5/20G06T7/11G06T7/136
Inventor 刘丽华虞华戴超凡黄宏斌邓苏
Owner NAT UNIV OF DEFENSE TECH