Date Fruit Sorting Based on Deep Learning and Discriminant Correlation Analysis

Date fruit is among the major crops in middle-east region, where millions of tons are harvested every year. a healthy fruit, which involves sugars, minerals and vitamins. In addition, it helps preventing human body from several diseases such as cancer heart diseases. sorting fundamental step date industry. However, manually conducting an operation, by labors, expensive time-consuming. this paper, we propose method for classifying type incorporating supervised unsupervised deep networks. Specifically, use discriminant correlation analysis (DCA) algorithm to fuse features learned convolution neural networks (VGG-F) network called PCANet. DCA jointly performs feature fusion dimensionality reduction with low computational complexity. To carry out experiments, introduce new benchmark dataset images 20 varieties. Our is, best our knowledge, largest one terms number Note that publicly available at https://unsat.000webhostapp.com/dataset. Experimental results demonstrate utility well complementarity fused features. It has also been shown effectiveness proposed compared relevant methods.