Font Size: a A A

A Mercury Thermometer Automatic Inspection System Based On Machine Vision

Posted on:2016-10-17Degree:MasterType:Thesis
Country:ChinaCandidate:K YaoFull Text:PDF
GTID:2284330479977450Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
With the expansion of industrial production scale and technical level unceasing enhancement, replacing manual operation with the machine in automatic production has become the trend of the development of the industry. In recent years, machine vision has been usually applied in industrial production tests because of its high efficiency intelligent technology advantage. This subject developed a mercury thermometer automatic inspection system based on machine vision according to the actual demand for medical device manufacturers. It can be used to check out the unqualified products and reject them in time so as to realize the purpose of efficient automatic production.An overall design scheme was put forward in the subject according to the requirements of the mercury thermometer automatic detection system, including the design of the hardware platform and software system. The system gets images with a line-scan digital camera, take some relevant pretreatment with the image in order to make the detection algorithm run more steadily. Firstly, stretch images lengthways to eliminate the difference between mercury columns and strengthen the consistency of calibration area in pictures. Then we highlight the difference between calibration area and background effectively through adaptive threshold operation in order to realize the recognition and location of the calibration area.Visual inspection algorithm was studied in the system, an inspection scheme based on pruning algorithm. The system computes thermometer reading with the relative position of calibration area and mercury column in the image. The image matching algorithm based on template and the local characteristics were respectively studied to realize the recognition and location of calibration area, on the other hand, in view of the recognition and location of mercury column, a longitudinal scan endpoint lookup algorithm was put forward in addition to the image recognition method based on template matching, which can effectively positioning the endpoint of the mercury column. According to the design of mercury thermometer inspection system,we programmed in VC++ compiler based on Open CV project. In addition, a simple and friendly interactive interface was designed. The system was tested in a experimental environment, the data showed that the performance of system meets the design requirements.This system can obtain the reading of product without external measuring instrument, which goes through software processing. The system is intelligent,it can improve the efficiency of inspection and reduce the labor costs, which has a good industrial application prospect.
Keywords/Search Tags:machine vision, line-scan digital camera, automatic inspection, image matching, longitudinal scan
PDF Full Text Request
Related items