Font Size: a A A

Developmentof Fault Light Detection System Based On Embedded System

Posted on:2020-04-08Degree:MasterType:Thesis
Country:ChinaCandidate:R B WangFull Text:PDF
GTID:2428330602953142Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of modern science and technology,a large number of intelligent instrumentation applications have laid the foundation for industrial automation and intelligence.Modern electronic science and technology has given these instruments and equipment more powerful functions and automation capabilities,and has made various system functions more diverse.The main technical means for detecting and fault diagnosis of modern instrument equipment are as follows: 1.Using the embedded software and hardware system to realize the automatic collection of operating parameters of the instrument,as well as the automatic monitoring and fault diagnosis of the operating state,and will be through the bus or network.The relevant information is submitted to the upper computer.2.The operation result and the fault information are displayed at the operation site by using various visual indicators such as a signal light or a display screen to help the technician judge the running state of the device.For example,some medical instruments,on-site monitoring of unmanned substations,etc.,reflect the operating status and operating parameters of the equipment through visual information such as various types of indicators."Development of portable fault light detection system" was proposed by a provincial medical product supervision and inspection agency.According to the technical requirements of the project party,we proposed a technical framework based on embedded system design,using high-speed industrial cameras as signal acquisition equipment.With the help of machine vision detection technology,the status of equipment operation is detected by detecting the color,frequency and duty ratio of various forms of signal lamps,which solves the problem of automatic detection of many medical instruments that need to use various light markers for online detection,and reduces the problem.Detect the labor intensity of the personnel.This topic uses AM3354 processor as the hardware platform,using Mars640-300 gc high-speed industrial camera for image acquisition,which solves the problem that ordinary camera can't perform accurate measurement of light flicker frequency and duty cycle due to low frame rate;use GigE interface for high speed The image data transmission and camera control method ensure the real-time test results;the LCD display is used to display the test results to the user,and the power signal interface is reserved to support the user to use the oscilloscope and other devices for verification.The application runs under the Linux operating system and implements functions such as GigE interface,shared resource allocation,multitasking management,etc.,and completes the color and flashing mode of various indicator lights(color LED,liquid crystal analog display,etc.)of medical instruments(The parameters such as constant light and strobe,flashing duty ratio,frequency,etc.are displayed on the LCD screen by comparing the operating status of the instrument with the fault table.Based on a detailed analysis of the embedded Linux operating system startup process task management,this project completed the Linux system migration process: BootLoader startup process analysis and porting,Linux kernel cutting and configuration,Linux root file system customization,GigE interface Work on camera driver design and installation.Aiming at the problems of large amount of video information and high redundancy,the inter-frame difference data is proposed to determine the flickering area and the flickering boundary,which reduces the storage requirement for massive image data.The application of image enhancement and edge extraction is improved.The recognition rate and contrast of the source image are used;the feature classification method of the color histogram is used to classify and recognize the fault lights of different colors,and finally the frequency and duty ratio of the fault lamp are accurately detected by the timing characteristics.The actual test of the prototype shows that the equipment developed can meet the requirements of the task book for measuring the color,frequency and duty cycle of the fault light.At the same time,it provides ideas and methods for such industrial automation production guidance and equipment testing.
Keywords/Search Tags:Cortex-A8 Microprocessor, GigE Camera, Embedded Linux Operating System, Fault Light Detection
PDF Full Text Request
Related items