The measurement system of hardware's shape and dimension based on image processing method

Image measurement technology applies the digital image processing technology to the field of precision measurement. And compared to traditional ways of measurement, image measurement technology has advantages of high accuracy, large dynamic range, non-contact and flexibility, therefore, it is very suitable for a number of on-site uses that traditional ways are difficult to achieve, and it has broad application prospects and practical value. In this paper, key technologies of image measurement technology were improved to raise the measurement accuracy and measurement speed. Two-dimensional median filter was used to filter image noise and enhance filtering speed. Edge extraction algorithm based on eight-direction Sobel operator overcame the edge fracture problem and gained unilateralism one pixel width edge. Hough transform based on graphics geometric characteristics was applied to detect the edge of the geometric primitives. Quadratic polynomial interpolation was on the gradient direction of edge. Sub-pixel located the geometric primitives. Finally, curve was fitted by least squares on the edge of pixel points, and its geometric parameters were gained. The measurement results of standard gauge blocks and measuring needles reaches sub-pixel accuracy and proves the effectiveness of the method propose in this paper.