ABSTRACT
Three kinds of breeder seeds’ samples are collected to verify the performance of crop classification and variety clustering based on electrophoretogram and investigate vegetable seed genetic relationships. They are bell pepper, Chinese cabbage, and cucumber. This chapter discusses the positive significance to automatic seed inspection by computer based on electrophoresis and selection of breeding direction. Pattern recognition test based on digital image process by computer is a new method with high test speed, strong resolving ability and high repeatability and can test large quantities tirelessly. This method has been widely used in the variety recognition and quality test of corn, peanuts, wheat, rice and other varieties. The recognition pattern is support vector machine (SVM) algorithm. Preprocessing of image includes some routine methods, such as images’ enhancement, de-noising, median filtering, and gray processing. The method of SVM is on the basis of Vapnik-Chervonenkis dimension and structural risk minimum principle of statistical learning theory.
