Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems

碩士 === 健行科技大學 === 工業管理系碩士班 === 106 === Because people''s dependence on cars increases, the proper rate of cars becomes important. The engines, steering, brakes, transmission systems, etc. in the automotive component system are driven and controlled by the structure of the fluid circuit sys...

Full description

Bibliographic Details
Main Authors: Yan-Lan Shen, 沈燕蘭
Other Authors: Chi-Jie Lu
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/g4etak
id ndltd-TW-106CYU05041005
record_format oai_dc
spelling ndltd-TW-106CYU050410052019-10-03T03:40:49Z http://ndltd.ncl.edu.tw/handle/g4etak Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems 機器視覺於汽車零組件流體迴路系統清潔度檢測之應用 Yan-Lan Shen 沈燕蘭 碩士 健行科技大學 工業管理系碩士班 106 Because people''s dependence on cars increases, the proper rate of cars becomes important. The engines, steering, brakes, transmission systems, etc. in the automotive component system are driven and controlled by the structure of the fluid circuit system. The components of the fluid circuit system are often affected. The circuit is blocked and the sliding surface is stuck, which causes the parts to fail. The part manufacturer needs to detect the cleanliness of the automotive component fluid circuit system to avoid the failure of the parts. The detection method is based on ISO 16232, and the digital microscope is operated by personnel. The method of image capture and measurement of residual particles on the filter membrane, measuring the maximum length of the particles, particle classification, etc.; such a detection method can’t obtain a complete filter film image, and easy to detect time is too long, difficult to judge Whether the particles are repeatedly measured, the measured value is distorted and the repeatability is not good; in order to find a good detection technology to solve the problems caused by manual detection, this study constructs the system architecture and method of machine vision detection, and carries out the characteristic values of the particles. Detection. According to the experimental results, it is better to use the indoor light source brightness 60 LUX to take the image acquisition architecture, and then use the image processing program combining variance filtering, thresholding and median filtering, the resolution of the final analysis is 78.1%; The average inspection time of the machine vision method took only 22.5 seconds, which was much lower than the average time of 205 seconds using the ISO 16232 test method. The detection system of this study can improve the detection efficiency by 89%. Chi-Jie Lu 呂奇傑 2018 學位論文 ; thesis 59 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 健行科技大學 === 工業管理系碩士班 === 106 === Because people''s dependence on cars increases, the proper rate of cars becomes important. The engines, steering, brakes, transmission systems, etc. in the automotive component system are driven and controlled by the structure of the fluid circuit system. The components of the fluid circuit system are often affected. The circuit is blocked and the sliding surface is stuck, which causes the parts to fail. The part manufacturer needs to detect the cleanliness of the automotive component fluid circuit system to avoid the failure of the parts. The detection method is based on ISO 16232, and the digital microscope is operated by personnel. The method of image capture and measurement of residual particles on the filter membrane, measuring the maximum length of the particles, particle classification, etc.; such a detection method can’t obtain a complete filter film image, and easy to detect time is too long, difficult to judge Whether the particles are repeatedly measured, the measured value is distorted and the repeatability is not good; in order to find a good detection technology to solve the problems caused by manual detection, this study constructs the system architecture and method of machine vision detection, and carries out the characteristic values of the particles. Detection. According to the experimental results, it is better to use the indoor light source brightness 60 LUX to take the image acquisition architecture, and then use the image processing program combining variance filtering, thresholding and median filtering, the resolution of the final analysis is 78.1%; The average inspection time of the machine vision method took only 22.5 seconds, which was much lower than the average time of 205 seconds using the ISO 16232 test method. The detection system of this study can improve the detection efficiency by 89%.
author2 Chi-Jie Lu
author_facet Chi-Jie Lu
Yan-Lan Shen
沈燕蘭
author Yan-Lan Shen
沈燕蘭
spellingShingle Yan-Lan Shen
沈燕蘭
Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
author_sort Yan-Lan Shen
title Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
title_short Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
title_full Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
title_fullStr Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
title_full_unstemmed Application of Machine Vision in the Cleanliness Testing of Automotive Component Fluid Circuit Systems
title_sort application of machine vision in the cleanliness testing of automotive component fluid circuit systems
publishDate 2018
url http://ndltd.ncl.edu.tw/handle/g4etak
work_keys_str_mv AT yanlanshen applicationofmachinevisioninthecleanlinesstestingofautomotivecomponentfluidcircuitsystems
AT chényànlán applicationofmachinevisioninthecleanlinesstestingofautomotivecomponentfluidcircuitsystems
AT yanlanshen jīqìshìjuéyúqìchēlíngzǔjiànliútǐhuílùxìtǒngqīngjiédùjiǎncèzhīyīngyòng
AT chényànlán jīqìshìjuéyúqìchēlíngzǔjiànliútǐhuílùxìtǒngqīngjiédùjiǎncèzhīyīngyòng
_version_ 1719259788067471360