Image Acquisition and Segmentation for Ceramic Bearing Ball Surface Inspection System

Abstract

Surface and subsurface defects in ceramic bearing balls significantly affect ball strength and lifetime. A brief review of nondestructive evaluation technologies on hybrid bearing balls was presented. An automatic ceramic bearing ball surface inspection system based on computer image technology was developed. Various surface defects were imaged by microscope and image digitizer on a novel ball surface unfolding mechanism. Top-hat and logarithm transformations were applied to eliminate uneven lighting disturbance. An automatic seeded region growing algorithm based on histogram concavity analyses and Otsu's method was developed for defects segmentation. The results show that the experimental system is effective for ceramic ball surface and subsurface defects detection.