The Research On Mini-size Precision Measurement Based On Image Analysis

The development of measurement technology indicated the level of science and technology to a certain degree in one country. In the present, with the rapid development of production, science and technology, the higher demand was put forward to the measurement precision, the measurement efficiency and the measurement automation of tiny dimension. So studying on the improvement of the precision of tiny dimension's measurement has practical significance. This paper has studied the method on how to improve the measurement precision of tiny dimension from software algorithm based on image measurement technology and used the mini-size gauge blocks as experiment object.The image processing techniques involved in the research, such as image filtering and edge detection etc are discussed and studied in the thesis. And based on these techniques, this paper brings forward some methods being suitable to this topic. In the aspect of image filtering, the source of image yawp was analyzed in the first, and then the edge details are preserved while removing the impulse yawp through adopting the median filter technology according to the traits of impulse yawp of the image;In the aspect of edge detection, we bring forward a method that can make the image edge obtain a pixel wide through using SOBEL arithmetic operators combining non-maximum restraining technology to get rid of these non-maximum points of the local. And using the SOBEL arithmetic operator feature of smoothing these residual yawps in the operation process, we make the computational result more stable and accuracy.More precisely locating the image edge is the key for improving the image measurement precision. We have provided a new edge subdivision algorithm based on the theory of gauss curve fitting and interpolation after detailed discussing the sub-pixel edge subdivision algorithm in this paper. Experiment results have testified the method can improve the CCD resolution.In order to validate the precision of measurement algorithm, the paper presents the practical measurement results through using the mini-size gauge image as experiment object in the end. The back accuracy of the experimental results reach 0.1 pixel. This illuminated that the measurement algorithm is successful in design and the effect is better.