Paper Title
Shallow and Deep Convolutional Neural Network Models for Classification of VNIR Wheat Samples

Abstract
Recently, the use of machine learning systems in the agricultural industry has increased tremendously for particular tasks. In this study, we have performed Shallow and VGG16 deep learning models on a new VNIR dataset in order to classify wheat samples. There are 40 classes and 200 instances per each class. While the performance shallow model has reached an accuracy of 80.13%, the VGG16 model is 91.13%. According to the results of obtained by simulations, deep learning methods were found to be more successful than parametric dependent based ones. Index Terms - Deep Learning Models, Wheat Classification, VNIR Imaging