DUS (distinctness, uniformity, and stability) testing of new peanut varieties is an important method for peanut germplasm evaluation and identification of varieties. In order to verify the feasibility of DUS testing for identification of peanut varieties based on image processing, 2,000 pod images of 20 peanut varieties were obtained by a scanner. First, six DUS testing traits were quantified successfully using a mathematical method based on image processing technology, and then, size, shape, color, and texture features (totally 31) of peanut pods were also extracted. On the basis of these 37 features, the Fisher algorithm was used as feature selection method to select “good” features to expand DUS testing traits set. Then, support vector machine (SVM) and K-means algorithm were used as model of varieties recognition and clustering analysis method, respectively, to study the problems of varieties identification and pedigree clustering comprehensively. It is found that, by the Fisher feature selection method, a number of significant candidate features for DUS testing have been selected, which can be used in the DUS testing further; using the top half of these features (about 16) after ordered by the Fisher discrimination, the recognition rate of SVM model is more than 90% in identification of varieties, which is better than unordered features. Besides this, a pedigree clustering tree of 20 peanut varieties has been built based on K-means clustering method, which can be used in in-depth study of the genetic relationship between different varieties. In summary, the results of this chapter may provide a novel reference method for future DUS testing, peanut varieties identification, and study of peanut pedigree.