Part Concentricity Error Measurement Based on Dimension-reducing Hough Transformation Algorithm
Guo Yan-ji · Jiguang zazhi · 2015
To overcome some limits presented in the traditional Hough transformation circle detection algorithm,which such as heavy calculation amount,low measuring efficiency and inapplicable to homocentric circles,the paper proposes the dimension- reducing Hough transformation algorithm to carry out parts concentricity error measurement with non-contact vision detection technique. The proposed method implements directional projection to the binarize object image undergoing threshold segmentation to obtain the radius parameters of circle contours. Then with the use of Hough transformation the centre coordinates of internal and external contours are calculated,and hence the concentricity error of the measured part is obtained. The proposed algorithm reduces the parameter space from 3D into 2D,and it can effectively divide the internal and the external contours characteristic points. Therefore,the computational efficiency is improved greatly. Additionally,with numerical simulation and experiment the effectivity and accuracy of the method are validated and analyzed. Moreover,the results indicate that the proposed method is a convenient,efficient,reliable and automatic non-contact measuring method.