As the surface of steel ball has high reflectivity and defect feature is difficult to extract,a rapid surface defect detection method is proposed which is based on the laser scattering image of steel ball surface and machine vision technology.The detecting system is designed,and the structure and principle of the detecting system are introduced.The image processing algorithms of morphology and edge connecting are used to extract laser speckles of the defects,and Heywood circularity factor and elongation factor are used to judge the defect type.Based on the LabVIEW software platform,steel ball samples of different defects are tested.Test results show that the defects which include pits,hard spots and scratches can be effectively extracted by using laser scattering images of steel ball surface.Based on the integrated system,the defects on the steel ball surface can be recognized rapidly and exactly.The system has practical value and application prospect for on-line testing of steel ball surface defects.