Image Classification using Shifted Legendre-Fourier Moments and Deep Learning

Image classification using deep learning has been of great interest to researchers in recent years due to the increasing use of digital images in various fields. In this paper, we present a new image classification system using image moments and deep learning. The proposed system is derived by combining Shifted Legendre-Fourier moment’s invariants and convolutional neural network (CNN). This method allows us to obtain an image classification model with high accuracy using small image datasets during the training phase. The validity of this proposed method has been provided under different transformations.