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

A machine learning-based defect classification method for injection molding products

A technology for injection molding products and defect classification, applied in instruments, computer parts, computing, etc., can solve the problems of increased time cost, inability to classify, single classification, etc., to achieve the effect of enhancing connection, improving accuracy, and effectively classifying

Active Publication Date: 2022-07-12
GUANGDONG UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this injection molding product detection method, although the secondary detection of unqualified products helps to improve the detection accuracy, it increases the time cost
Another type of classification for injection molding products is based on physical judgment of whether the shapes are consistent. Therefore, there is a shortcoming of too single classification. When there are cases where the classified products have the same shape, effective classification cannot be carried out.

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 machine learning-based defect classification method for injection molding products
  • A machine learning-based defect classification method for injection molding products
  • A machine learning-based defect classification method for injection molding products

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] This embodiment proposes a machine learning-based defect classification method for injection molding products, such as figure 1 As shown, it is a flowchart of a method for classifying defects of injection molding products based on machine learning in this embodiment.

[0042] A method for classifying defects of injection molding products based on machine learning proposed in this embodiment includes the following steps:

[0043] S1: Collect images of injection molded products with various defect types as training samples, and preprocess them.

[0044] In this step, the steps of preprocessing the training samples include:

[0045] S11: Perform manual labeling on the images [t1, t2,..., tn] in the training sample to obtain an image label set [s1, s2,..., sn];

[0046] S12: Grayscale the images [t1, t2,..., tn] in the training samples, and perform parameter transformation on the two-dimensional domain, and the transformation formula is as follows:

[0047] tn′=tn×τ

[...

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 present invention provides a method for classifying defects of injection molding products based on machine learning, which includes the following steps: collecting training samples; inputting the training samples into a classifier for training to obtain a trained classifier; The image of the product is preprocessed; the image of the product to be classified is input into the trained classifier to obtain the classification label, and then the one-hot encoding is transcoded to obtain the label matrix; the label matrix is ​​stacked to obtain the label containing the label information. Stacked matrix; separate the data of the stacked matrix to obtain a low-rank matrix and a sparse matrix, and decompose the sparse matrix into the noise of the training sample and the noise of the image of the product to be classified; iterate the low-rank matrix and take the optimal solution after iteration The corresponding label matrix is ​​used as the optimized label matrix; the injection molding product system converts the optimized label matrix into a machine signal to realize the control of the defect classification of the injection molding product by the machine on the production line.

Description

technical field [0001] The invention relates to the technical field of product defect classification, and more particularly, to a machine learning-based defect classification method for injection molding products. Background technique [0002] At present, the defects of injection molding machine products are mainly caused by the design of the mold, the manufacturing precision and the degree of wear. Since the injection molding cycle itself is very short, if the process conditions are not well mastered, a large number of waste products will be generated. [0003] An existing method for intelligent secondary inspection of injection molded products mainly judges whether the products are qualified by comparing preset image information, and then performs secondary detection on unqualified products to avoid false reductions or missed inspections. In the injection molding product inspection method, although the secondary inspection of unqualified products helps to improve the inspe...

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): G06K9/62G06V10/764G06V10/774
CPCG06F18/214G06F18/241Y02P90/30
Inventor 谭淇李建中潘嘉伟吴宗泽
Owner GUANGDONG UNIV OF TECH
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