Product defect detection method, device and system

A technology for product defect and defect detection, applied in image analysis, image enhancement, instruments, etc., can solve the problems of fluctuation of inspection quality, high labor intensity of human inspection, and shortage of human resources, and achieve the effect of improving accuracy

Pending Publication Date: 2020-05-19
GOERTEK INC
View PDF10 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In general manufacturing plants, inspection manpower accounts for nearly 30%. Due to the large demand for manpower, there is often a shortage of human resources; and the labor intensity of manpower inspection is high, and it is easy to cause fluctuations in inspection quality due to operator fatigue.

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
  • Product defect detection method, device and system
  • Product defect detection method, device and system
  • Product defect detection method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] figure 1 is a block diagram of the hardware configuration of the product defect detection system 100 according to the embodiment of the present invention.

[0031] Such as figure 1 As shown, the product defect detection system 100 includes an image acquisition device 1000 and a product defect detection device 2000 .

[0032] The image capture device 1000 is used to capture product images, and provide the captured product images to the product defect detection device 2000 .

[0033] The image acquisition device 1000 may be any imaging device capable of taking pictures, such as a camera.

[0034] The product defect detection device 2000 can be any electronic equipment, such as PC, notebook computer, server and so on.

[0035] In this example, refer to figure 1 As shown, the product defect detection device 2000 may include a processor 2100, a memory 2200, an interface device 2300, a communication device 2400, a display device 2500, an input device 2600, a speaker 2700,...

Embodiment 2

[0042] In the process of product manufacturing, factors such as unstable process, insufficient mechanical positioning accuracy, and the environment in the factory building often cause the produced products to have various forms of defects or defects. In the early stage of product manufacturing, there are often very few bad samples, and classification algorithms based on big data and big models are difficult to meet the requirements of the production line.

[0043]In view of the above-mentioned small sample problem, the embodiment of the present application first uses the image segmentation algorithm to predict possible defects and their positions in the image, and generates a corresponding mask (mask) image, and then stitches the original image and the mask image, and inputs classifiers to classify different defects.

[0044] figure 2 The flow chart of the product defect detection method shown in the embodiment of the present invention, such as figure 2 As shown, the metho...

Embodiment 3

[0060] The embodiment of the present application also provides a product defect detection method. In this embodiment, in the above step S2200, using sample images of products containing different defect types to train the segmentation network respectively includes S2210-S2220:

[0061] S2210. Input multiple sample images containing different defect types to multiple segmentation networks respectively.

[0062] In this embodiment, one sample image may contain n defect types, or may contain less than n defect types, and the defect types contained in each sample image may be the same or different.

[0063] refer to image 3 , image 3 It shows that sample images containing n types of defects are input to n segmentation networks, and the n types of defects are all defects that can be segmented and located using segmentation algorithms, and each segmentation network outputs a mask image for locating the position of the defect, which is image 3 The mask images mask_1, mask_2, ....

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 a product defect detection method, a device and a system. The method comprises the steps of constructing a defect detection framework, the defect detection framework comprisessegmentation networks, splicing networks and classification networks, the number of the segmentation networks is set according to product defect types, and each segmentation network corresponds to onedefect type; respectively training the segmentation network by using the sample images of the products containing different defect types to obtain the segmentation network of the mask image capable of positioning the position of each defect; splicing the sample image with the mask image output by each segmentation network by using a splicing network to obtain a spliced image; training the classification network by using the spliced image to obtain a classification network capable of correctly identifying product defects and defect types; when product defect detection is carried out, an acquired product image is input into the defect detection framework, and defects and defect types existing in a product are detected by utilizing the segmentation network, the splicing network and the classification network.

Description

technical field [0001] The invention relates to a product defect detection method, device and system. Background technique [0002] In the traditional precision manufacturing industry, product defect detection is generally done through manual inspection. In general manufacturing plants, inspection manpower accounts for nearly 30%. Due to the large demand for manpower, there is often a shortage of human resources. Moreover, human inspection is labor-intensive, and it is easy to cause fluctuations in inspection quality due to operator fatigue. Therefore, a machine-automated inspection solution with stable inspection quality, consistent results, and no influence from human factors will surely be favored by the precision manufacturing industry. Contents of the invention [0003] The object of the present invention is to provide a product defect detection method, device and system. [0004] On the one hand, an embodiment of the present invention provides a product defect dete...

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): G06T7/00G06T7/10G06T3/40G06V10/764G06V10/774
CPCG06T7/0004G06T7/10G06T3/4038G06T2207/10004G06T2207/20081G06T2207/20084G06V10/89G06T2207/30164G06V10/764G06V10/774G06T2207/30108G06F18/243G06F18/2148G06F18/2193
Inventor 刘杰田继锋张文超张一凡
Owner GOERTEK INC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap