Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Surface defect detection method and device and computer readable storage medium

A defect detection and defect technology, which is applied in the field of image recognition, can solve the problems that the detection method cannot meet the requirements and the surface defect image data is small, and achieve the effect of solving the problem of small area defect detection, increasing the amount of training data, and avoiding proportional imbalance

Pending Publication Date: 2021-04-30
ZHENGZHOU JINHUI COMP SYST ENG
View PDF10 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The purpose of the embodiments of the present invention is to provide a surface defect detection method, which aims to overcome the problem that the existing detection method cannot meet the requirements due to the lack of surface defect image data and small defects.

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
  • Surface defect detection method and device and computer readable storage medium
  • Surface defect detection method and device and computer readable storage medium
  • Surface defect detection method and device and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] For a surface defect method provided by the first embodiment of the present invention, please refer to figure 1 , the method includes step S01 to step S07:

[0050] Step S001, collecting surface images of industrial parts.

[0051]Use professional acquisition equipment to collect images on the surface of industrial parts. The surface of each part should be captured as several non-overlapping images. In order to ensure the high quality of the image, the parts should be photographed in a controlled environment, and the high resolution should be ensured to avoid the loss of tiny cracks or other defects due to pixel problems. For example, an industrial image acquisition system composed of a host computer, an industrial camera, an LED light source, and a photoelectric sensor. Its working process is: first initialize the equipment and self-check the equipment, and then the host computer drives the industrial camera (area array CCD sensor) through software, but the industria...

Embodiment 2

[0063] In another embodiment of the present invention, the images collected by collecting the surface images of industrial parts include images with defects and images without defects.

[0064] By collecting both defective and non-defective images, training on both defective and non-defective images can increase the amount of data and improve the judgment accuracy of the defect detection network.

Embodiment 3

[0066] In one embodiment of the invention, see image 3 , the construction of the defect detection network includes steps S031-S032:

[0067] Step S031, building a segmentation network, the segmentation network is composed of a convolutional layer, a pooling layer and a fully connected layer.

[0068] Such as Figure 4 Schematic diagram of the segmentation network shown, which consists of convolutional layers, pooling layers, and fully connected layers. The size of the convolution kernel and the number of convolution layers can be changed according to the size of the image block, but in order to capture the small feature details in the image, this segmentation network needs to use a large convolution kernel in a relatively deep layer, and all convolutions The step size of the product kernel is set to 1. The segmentation network of the present invention uses the pooling layer to complete the downsampling operation, and uses the pooling layer to replace the convolutional laye...

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 is applicable to the technical field of image recognition, and provides a surface defect detection method, which comprises the following steps of: acquiring a surface image of an industrial part; partitioning the original large-size image of the surface image into a plurality of sub-images to obtain a sub-image training data set; constructing a defect detection network; inputting the sub-images in the sub-image training data set into the defect detection network for training to obtain a trained defect detection network; carrying out segmentation processing on a to-be-detected image, and segmenting the to-be-detected image into a plurality of sub-image block images; and inputting the plurality of sub-image-block images into a trained defect detection network to obtain detection results of the plurality of sub-image-block images. According to the method provided by the invention, the problem that the detection difficulty is increased due to poor surface defect detection precision, small data volume and small defects in the prior art is solved.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a surface defect method, device and computer-readable storage medium. Background technique [0002] In the process of industrial production and manufacturing, quality assessment and inspection are particularly critical steps. Reducing waste products from the factory and ensuring high quality of finished products have always been the goals pursued by factories. Inspection, especially product defect inspection, has always been a legacy problem in the industrial field. The traditional quality control method is to rely on trained quality inspectors to manually screen products. However, with the continuous expansion of the factory scale and the continuous high efficiency and intelligentization of industrial product production lines, the disadvantages of labor time-consuming and low efficiency have seriously affected the factory. automation reform and rapid develo...

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): G06T7/00G06N3/04G06N3/08
CPCG06T7/0004G06N3/08G06T2207/20081G06T2207/20084G06T2207/30164G06N3/045
Inventor 徐明亮王可刘奕阳姜晓恒张晨民李丙涛栗芳
Owner ZHENGZHOU JINHUI COMP SYST ENG
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products