Tire laser speckle defect identification method based on improved residual network

A laser speckle and defect recognition technology, applied in the field of defect recognition based on deep learning, can solve the problems of low recognition accuracy and low missed judgment rate, and achieve the effect of improving safety, reducing pressure and avoiding inefficiency

Pending Publication Date: 2022-05-31
SHENYANG LIGONG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: Aiming at the above-mentioned problems existing in the existing quality inspection methods, the present invention provides a tire laser speckle defect identification method based on the improved residual network, and uses the improved Resnet-50 model to solve the problems of low defect identification accuracy and missing defects. Judgment rate is still not low
Realized that the computer replaces the human eye to judge whether the current tire laser speckle image is a defect image

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
  • Tire laser speckle defect identification method based on improved residual network
  • Tire laser speckle defect identification method based on improved residual network
  • Tire laser speckle defect identification method based on improved residual network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] In order to describe the present invention more specifically, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] This example aims to realize defect identification of tire laser speckle images according to the present invention. The method process includes image preprocessing, image enhancement, improving Resnet-50 model training, inputting the image to be tested into the model, and obtaining results. figure 1 The specific implementation process is as follows:

[0026] (1) Pass the tire through the tire laser speckle detector to obtain the original tire laser speckle image.

[0027] (2) Image preprocessing. A tire can be divided into 16 original pictures with a size of 1360*1024 after the tire laser speckle detector, and the original pictures are converted into the format and format size of the input model: 224*224, and the defects and normal xml files are recorded. The preprocessed i...

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 tire laser speckle defect identification method based on an improved residual network, and the method is based on an improved Resnet50 model, and utilizes the characteristics that the structure is simple, and the network is basically not reduced due to the increase of equal mapping. A computer is used for replacing human eyes to judge whether a current tire laser speckle image has defects or not, and binary classification is achieved through a classifier. According to the method, low tire detection efficiency caused by human factors can be avoided, and the precision ratio of defects is increased. Therefore, the safety of the tire is greatly improved, and the working pressure of quality inspection personnel can be relieved.

Description

technical field [0001] The invention belongs to the technical field of computer vision and industrial detection, and relates to a tire laser speckle defect identification method based on an improved residual network, which is a defect identification method based on deep learning, so that a computer can replace manual identification of tire laser speckle images. defects in. Background technique [0002] The tire laser speckle detector belongs to the last quality monitoring stage of the tire manufacturing process. Now major manufacturers generally adopt a three-shift system to control the quality of this stage. The specific process is that the tire is sent to the tire laser speckle detection channel, and the tire laser speckle detector takes the tire laser speckle image for the tire. The quality monitoring personnel see the tire laser speckle image in front of the display, and then the quality inspection personnel It will judge whether there is a defect according to the tire ...

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): G06T7/00G06T5/00G06T3/40G06T3/60G06N3/04G06N3/08G01N21/88
CPCG06T7/0004G06T5/007G06T3/4007G06T3/4046G06T3/60G06N3/08G01N21/8851G06T2207/20081G06T2207/20084G01N2021/8887G01N2021/8883G06N3/048Y02P90/30
Inventor 刘韵婷葛忠文郭辉
Owner SHENYANG LIGONG UNIV
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