To Distinguish Internal Infestation in Wheat Kernels Using Biophoton Technology and CS-BP Algorithm

Stored grain insects detection is all along one of research focuses in grain storage industries. Based on biophoton analytical technology, the ultraweak photons emitted from the health and insects-contaminated wheat are measured separately. Six statistical features (median, mean, quartile deviation, mean deviation, variance, Coefficient of Variance) and 20 histogram features are extracted from the measuring datum to construct wheat feature vector. The proposed algorithm CS-BP is used as the classify algorithm, some other machine learning algorithm (KNN, SVM, BP) are also utilized to check the effective of the proposed method. The experimental results show that the proposed model can differentiate the normal wheat from the insects-contaminated one. The proposed model can provide a new thought for the detecting the wheat pests.