Abstract: The planar three-point circle center positioning method,based on machine vision technology,is proposed for the problem that the irregular-shaped shaft test hole for mechanical characteristics of an intelligent molded case circuit breaker is difficult to position and machine.After the captured images are analyzed by probability area histogram,the images will be pre-processed by median filtering,gray stretching and linear transformation algorithm.Then,the target areas are identified by circular template matching algorithm.Finally,the interest points of the images are detected by improved Harris corner detection algorithm after eliminating the burrs of the target areas and regional small points by opening algorithm,and then the drilling position coordinate is determined after classification and conversion.The position and process results of the 20test holes show that 85% of the hole concentricity is within the range of IT11grade,and 90%of the hole straightness is within the range of IT12grade,which satisfies the positioning accuracy requirement of the intelligent molded case circuit breaker main shaft test hole for mechanical properties.