Detection Method Of Assembly Coaxality For Parts With High Length-diameter Ratio In Local Imaging

Rotary parts assembly quality has played an important role in the quality and the performance of the parts. Currently, assembly coaxality detection method based on image measurement is mostly through acquiring the whole contour information of parts to detect assembly coaxality error. And if it uses detection method of assembly coaxality in global imaging to detect assembly coaxality error for complex rotary parts with high length-diameter ratio or many assemble joints, the method can’t use very effectively image sensor pixel information, it causes resource waste and increases the amount of calculation, thus it affect parts assembly coaxality detection precision.So, we proposed a detection method of assembly coaxality in local imaging in this paper, that is to say, the inspected part’s assembly coaxality is detected with rapid and quantitative single perspective detection by collecting assembly area. First of all, it analyzes that the difference of part shape, the selection of window width or window level have a great influence on the assembly coaxality detection precision based on traditional Hough transform or fitting method in the process of local imaging detection. Secondly, a detection algorithm of assembly coaxality in local imaging was proposed to solve the above problem. This algorithm firstly extracts the upper and lower edge point sets and then counts the parameter space points projecting in Hough space from two point sets based on Hough linear transform, and searches the maximum point of parameter points’ cumulative number, the point corresponds to the symmetric axis which is the optimal symmetric axis, thus achieving part assembly coaxality error. Finally, through the field of experiment system of collecting standard phantom assembly area, and then using the algorithm of this paper to test assembly coaxality, it verifies that the algorithm is feasible and accurate of assembly alignment measurement for parts with high length-diameter ratio and the difference of part shape, the selection of window width or window level have little influence on the assembly coaxality detection precision based on the algorithm of this paper.