Study Of The Visual Online Parts Size Measurement Methods

Indirect measurement uses machine vision and image processing technology to achieve a direct measurement of small parts of the parameters, it not only reduces the intensity of the work of staff, but also the work efficiency and accuracy significantly improved. Using a computer instead of manual interpretation, reduce visual error which is due to long working hours and reading error caused by the habit of measuring themselves, reading errors of measuring instrument, aiming error and so on. On the basis of ensure the accuracy and improve measurement efficiency, to further improve the application level measurements of small parts.In this paper, the machine vision field measurement techniques is used to measure two-dimensional geometry of mechanical parts, to improve measurement accuracy and efficiency, as with area array CCD sensor, the mechanical parts on-line measurement system is designed. focusing on the key image processing technology and system calibration problem, realize a sub-pixel edge positioning part size and lay the foundation for the further realization of high-precision measurement of mechanical parts. Starting from machine Vision technology, descript the application of machine vision technology in-line measurement of the size of machining parts exhaustively, focused on the measurement process and algorithms, and software programming developed by online measurement system based on machine vision. The main contents are as follows:1. Introduced the research background and significance, the domestic and international current situation and existing problems, Clear solutions to this article. Describes the machine vision technology, give theoretical analysis and experimental verification feasibility of Online Measurement System which is based on machine vision;2. Research on camera calibration technique, use the standard calibration method for system calibration, the calibration results calculated through experiments and analyzed system measurement accuracy;3. Through the analysis of various image processing algorithms, in C++development environment combined with Opencv library, realized the image pre-processing and pixel edge detection. Depth analysis of the sub-pixel edge positioning technology and Hough circle detection technology, and proposed an improved Hough circle detection algorithm, the method is simple, fast operation. After the rough location using sub-pixel edge detection algorithm for edge refinement, improved edge positioning accuracy;...