Piston surface defect detection method and system based on deep learning

A technology of defect detection and deep learning, which is applied in the field of piston surface defect detection based on deep learning, to achieve all-round detection, the best image acquisition effect, and the effect of improving detection accuracy and efficiency

Pending Publication Date: 2021-01-05
SUZHOU CASIA ALL PHASE INTELLIGENCE TECH CO LTD
View PDF2 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] In order to solve the technical problems existing in the prior art, the present invention provides a method and system for detecting defects on the piston surface based on artificial intelligence

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
  • Piston surface defect detection method and system based on deep learning
  • Piston surface defect detection method and system based on deep learning
  • Piston surface defect detection method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0156]According to a specific implementation of the piston surface defect detection system based on deep learning of the present invention, combined with the attachedPicture 8 , The present invention will be described in detail. The invention adopts a plurality of cameras and different angle waveband end light sources to collect images at different image collecting stations.

[0157]The invention provides a piston surface defect detection system based on deep learning, including:

[0158]AttachedPicture 8 Middle 1-9 are cameras, the above cameras collect images of different parts of the piston; processor 1 connects camera 2, camera 4 and camera 7, processor 2 connects camera 3 and camera 8, processor 3 connects camera 5 and camera 6, Processor 4 connects camera 1 and camera 9; the total number of connected cameras is 9. The transmission protocol between processors 1, 2, 3 and 4 and the server is Socket TCP / IP. Processors 1 to 4 communicate with the server and PLC, the camera solution is s...

Embodiment 2

[0194]According to a specific implementation of the piston surface defect detection system based on deep learning of the present invention, combined with the attachedPicture 8 , The present invention will be described in detail.

[0195]The present invention provides a method for detecting defects on the surface of a piston based on deep learning. Taking the camera 2 connected to the camera workstation 2 as an example, the camera 2 collects a smooth and clean image, including the following steps:

[0196]The camera 2 collects multiple original image samples of the piston of the camera station;

[0197]For the smooth face image, the gray-scale image algorithm is used for processing and analysis, and the sample defect feature detection result is obtained;

[0198]The body includes:

[0199]Segment the image collected by the camera into multiple sub-pictures according to the area to be detected;

[0200]Transform multiple sub-pictures into the frequency domain, and estimate the background gray value of ...

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 provides a piston surface defect detection method and system based on deep learning, and belongs to the technical field of piston detection. According to the detection method, machine vision and a deep learning neural network are combined for piston surface defect detection, a gray image processing algorithm is adopted for defect detection on a smooth surface, and an improved FASTER-RCNN deep learning neural network is adopted for defect detection on a cast surface; meanwhile, a plurality of cameras and end light sources with different angles and wavebands are adopted to performimage acquisition at different image acquisition stations, so that the optimal image acquisition effect is achieved; and segmentation processing is carried out on the high-resolution picture, so thatthe accuracy of picture detail identification is improved. According to the invention, efficient and accurate automatic detection of various defects is realized.

Description

Technical field[0001]The invention relates to the technical field of piston defect detection, in particular to a method and system for detecting piston surface defects based on deep learning.Background technique[0002]At present, most of the domestic and foreign piston defect inspections are done by manual methods. In the manual inspection process of piston products, there are problems such as missed inspections, false inspections and low efficiency, and the detection efficiency and accuracy cannot be guaranteed.[0003]With the application of traditional machine vision algorithms in piston detection, the phenomenon that piston detection only relies on humans has been changed. Traditional machine vision algorithms can be used to detect piston oil hole leakage and oil hole aluminum chip defects; piston ring ring body Defects, cleaning, inspection bruises, graphite, skirt blisters, rim blisters, ring-inlayed bubbles, no ring-inlayed, ring-inlayed position (360° full inspection), surface ...

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): G06T7/00G06T7/11G06T7/90G06N3/08G06N3/04G06K9/62G06K9/32
CPCG06T7/0004G06N3/08G06T7/11G06T7/90G06V10/25G06N3/045G06F18/23G06F18/24
Inventor 杨光于普鹤余章卫孙浩楠周豪陆峥岩黄德奔周洲
Owner SUZHOU CASIA ALL PHASE INTELLIGENCE TECH CO LTD
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