Abstract: The yellow corn which is popular in China was selected for the study, then the ratio of white part(embryo) to yellow part(crown) of corn seed was found out. Their color and shape are useful to identify corn vatieties by analyzing and contrasting the images and data of different varieties of corn seeds. Experiments were conducted using support vector machine (SVM) combined with genetic algorithm(GA) and Statistical Program for Social Sciences (SPSS). Experimental results indicate that the ratio of white part to yellow part is valuable and the color characteristics of the yellow area are better than that of the whole corn seeds to identify corn varieties.