The Application Research Of Hough Transform In Vision Measurement System

Vision measurement is an application of machine vision theory in measurement field. It researched further and applied widely abroad, and has been paid more attention as in China. With the ratio of performance to price increasing and the relative technology developing, the vision inspection technology has realized the aim to zero-wasted of mechanical product and has suited for the need of modern manufacturing progress and development by its virtues such as non-contact, on-line, real-time, rapid, suitable precision and strong scene anti-jamming. It has showed wide applied foreground in production.Combined with the technology of machine vision and image processing, with piston as measured object, the thesis researches on kinds of key technique in image measurement. For the image preprocessing, the thesis analyzes result of several kinds of smoothing filter, the median filter stands out for its good edge maintenance, adaptability and speed. This article takes fast median filter as the way to process the collected image. In edge detection part, the performance between classical edge algorithm and the Canny algorithm is compared. Canny algorithm is better than classical edge detection algorithm in the way of the precision of edge locating and anti-noise. So the Canny algorithm is used in the paper. According to the characteristic of the measurement system and measured object, an improved algorithm of Hough Transform is proposed and is applied to diameter measurement. In this thesis, using VC++ 6.0 as the development platform, it realizes the improved algorithm to measure the diameter and the experimental results prove the availability and validity of this new method. In addition, base on the measurement process and system, the thesis analyzes the source and influence of error, and then gives some effective plans to control error.The theory and the examination proved that applying Hough Transform in image measurement to measure the dimension of workpiece is feasible. According to the character of the edge image, the improved arithmetic reduces the dimensions of parameter space and the quantum of calculation, so it raises the speed of image processing and keeps the precision.Due to the expensive price of high precision optical system and image sensor, the existing laboratory equipment limited the research and real-time application. In spite of that, the total work achieved the satisfying result and made a strong foundation of the future practical application.