The Corn Seed Image Segmentation and Measurement of the Geometrical Features Based on Image Analysis
In this article, separation of touching grain kernels and measurement of geometric features in an image are presented. The objective of this work is to discriminate single corn kernel and some broken kernels, which are difficult to achieve on the existing machinery and equipment, especially for the counted number and quality inspection process. The digital image was obtained from flatted conveyor belt; the geometrical features were analyzed using Matlab R2009a software. It was used for recognizing multiple kernels for once. Intact and broken kernels were characterized by the changed geometrical features and combining SVM for the classification of them with accuracy 95%.Which meet the requirement of corn quality inspection comparable to subjective human inspection.
M. Zhao et al., "The Corn Seed Image Segmentation and Measurement of the Geometrical Features Based on Image Analysis", Applied Mechanics and Materials, Vols. 66-68, pp. 1100-1105, 2011