Study on origin traceability of peanut pods based on image recognition

In order to investigate variety-specific reflected by peanut from different origin, we use a scanner to capture the images of the same species (Huayu 22) from three different regions, Each variety includes one front and two side images of 100 peanuts respectively. For each image, we have acquired 50 characteristics including shape, color and texture characteristics. We built artificial neural network model for identification based on these characteristics and those optimized by PCA, Result shows that for different species origin the maximal detectable rate reaches 100%. Methods and conclusions of this paper have positive significance to the DUS testing of peanut.