Study on the Pedigree Clustering of Peanut Pod´s Variety Based on Image Processing

Using image recognition, 20 varieties of peanut are collected through scanner. The image of 100 peanut seeds´ positive and two sides of each variety are collected; each image´s characteristics of shape, color and texture totaling up to 50 are obtained. A cluster analysis model based on these features and another cluster analysis model with PCA data optimizing is built. For further step, we have got a pedigree clustering tree for 20 peanut varieties. The first 17 principal components of PCA features have been able to fully simulate the 50 statistical features. The peanut is almost no difference in the classification categories when the cumulative contribution rate is getting 85% or more.