Supercharge Your Innovation With Domain-Expert AI Agents!

A crack detection system for automobile sheet metal parts based on CNN and LR

A technology for automobile sheet metal parts and crack detection, which is applied in measuring devices, optical testing flaws/defects, image data processing, etc. The effect of accuracy

Inactive Publication Date: 2018-12-28
柳州市木子科技有限公司
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

No testing process data, unable to reflect production process data

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
  • A crack detection system for automobile sheet metal parts based on CNN and LR
  • A crack detection system for automobile sheet metal parts based on CNN and LR

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] Such as Figure 1-2 As shown, the present invention provides a kind of automobile sheet metal parts crack detection system based on CNN and LR, and concrete detection process is as follows:

[0033] 1) Manually mark positive and negative samples;

[0034] 2) Use the model for training: first use CNN to map the image into a one-dimensional vector of fixed dimensions, use the vector as the feature, and use LR to train the model parameters;

[0035] 3) Predict the new product image to determine whether it is qualified.

[0036] The process of CNN mapping an input image to a fixed-dimensional vector is as follows:

[0037] The input layer consists of 32×32 perceptual nodes, receiving the original image, and then, the computational pipeline alternates between convolution and subsampling, as follows:

[0038] The first hidden layer is convolved, which consists of 6 feature maps, each feature map consists of 28×28 neurons, and each neuron specifies a 5×5 receptive field;

...

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 an automobile sheet metal part crack detection system based on CNN and LR. The specific detection flow is as follows: firstly, positive and negative samples are labeled manually; then the model is used to train; CNN is used to map the image to one-dimensional vector of fixed dimension, and LR is used to train the parameters of the model with the feature of vector; finally,the new product image is predicted to determine whether it is qualified or not. The input layer of CNN consists of 32*32 perceptual nodes, which receive the original image. Then, the computational flow alternates between convolution and sub-sampling to obtain the output vector. The final output vector is predicted by LR and label is judged. The technical scheme realizes an automobile sheet metal part detection system based on CNN and LR. The CNN neural network has strong generalization ability, can ensure high accuracy, solves the problems of low manual detection accuracy and high labor intensity, and reduces the cost of the company.

Description

technical field [0001] The invention relates to a crack detection system for automobile sheet metal parts, in particular to a CNN and LR-based crack detection system for automobile sheet metal parts, belonging to the technical field of automobile detection. Background technique [0002] Automobile sheet metal parts: Automobile sheet metal parts are essential parts of the car. Sheet metal parts are a part of the car body in white. Every car uses at least hundreds of different sheet metal parts, which play a role in the car. It plays a very important role, and its quality problems directly affect the safety of the car, the quality of the whole vehicle and the production cycle of the OEM. [0003] Automobile sheet metal parts detection: For each automobile sheet metal part product, it is necessary to detect the outer contour size, hole size, angle and surface cracks, etc., all of which can be classified by detecting the image of the sheet metal part to determine its Eligibilit...

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
IPC IPC(8): G06T7/00G06N3/04G06N3/08G01N21/88
CPCG06N3/08G06T7/0004G01N21/8851G06T2207/30164G06T2207/30136G06T2207/10004G06N3/048G06N3/045
Inventor 李振书
Owner 柳州市木子科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com