A Survey on Image Classification and Activity Recognition using Deep Convolutional Neural Network Architecture

Deep learning, over a decade it becomes the booming field for researchers since the technique has the capability to overcome the drawbacks of already used traditional algorithms which is dependent on hand designed features. Currently four different type of architecture used in deep learning which is an Autoencoder, Deep Belief Network, Convolutional Neural Network and Restricted Boltzmann Machine. According to the reported research, Convolutional Neural Network is very efficient on image classification and speech recognition. The main aim of this survey is to broadly cover the applications of convolutional networks in the field of computer-aided diagnosis for the dreadful diseases and also in the field of agriculture. Finally, the limitations of Convolutional network and expected future research topics to be done using this network have been discussed.