21 April 1995 Image segmentation for stone-size inspection
Author Affiliations +
Proceedings Volume 2501, Visual Communications and Image Processing '95; (1995) https://doi.org/10.1117/12.206698
Event: Visual Communications and Image Processing '95, 1995, Taipei, Taiwan
Object size inspection is an important task and has various applications in computer vision, for example, automatic control stone-breaking machines. In this paper, an algorithm is proposed for image segmentation on size inspection of almost round stones with strong textures or almost no textures. We use one camera and multiple light sources at difference positions to take one image when each of the light sources is on. Then we compute the image differences and threshold them to extract edges. We will explain, step by step, picture taking, edge extraction, noise removal, and edge gap filling. Experimental results will be presented. Through various experiments, we find our algorithm robust on various stones and under noise.
© (1995) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jui-Pin Hsu, Jui-Pin Hsu, Chiou-Shann Fuh, Chiou-Shann Fuh, "Image segmentation for stone-size inspection", Proc. SPIE 2501, Visual Communications and Image Processing '95, (21 April 1995); doi: 10.1117/12.206698; https://doi.org/10.1117/12.206698


Deformable surfaces: a free-form shape representation
Proceedings of SPIE (August 31 1991)
Segmented Aperture Lenses for Imaging and Illumination
Proceedings of SPIE (January 21 1987)
Shape From Symmetry
Proceedings of SPIE (February 28 1990)
Discounting Illuminants Beyond The Sensor Level
Proceedings of SPIE (March 26 1989)

Back to Top