Convolutional Neural Networks and their Applications: A Review
Abstract
This manuscript presents the current techniques used in Facial recognition and their applications. Deep Learning Convolution Neural Networks (CNN) which are an aid to identify a human face using technology are considered here and works relating to their applications are reviewed. Latest works culled from international journals relating to Face recognition have been gathered and surveyed to find which procedure is more common than the others in Face recognition and also in which area these techniques are applied. Those works cited in journals from the year 1993 onwards have been considered. This paper establishes the fact that presently, deep CNNs are the widely used approach in Face recognition. Also it is investigated as to which approach of the face recognition matches the data with a digital repository of known faces to find a match. Facial identification could help verify individual identity, but it also raises privacy issues. Though the exactness of facial recognition system as a biometric technology is lower than that of iris and finger print identification, it is widely adopted due to its contactless and non-invasive process. Recently, it has also become well-known as a business tool for recognition. Human face recognition pattern has brought numerous enhancements in the recent environment of detecting a face through technicality. It likens the image with a gallery of known images to encounter a match. Face identification authenticates personal identity as well as raising secrecy challenges. Artificial neural networks (ANNs) are extensively used in the course of treatment of images. Back-propagation neural networks (BPNNs) are extensively employed in the classification of facial images besides accomplishing high accuracy. Despite everything, there is a variation due to different neural designs and implementations. Deep CNN is a neural network designed for processing structured arrays of data such as images which are widely used in computer vision and have become the state-of-the-art for many visual applications such as image classification. Deep CNN is applied in different fields such as Intrusion Detection and Prevention Systems, Recommender systems, Sentiment analysis, Social Networks Dealing with Malware, Spam and Social Engineering Detection, Network Traffic Analysis, User Behavior Analytics etc. This manuscript comprises: I.Introduction, II. CNN III.Approaches and its Applications, IV.Types of Face Recognition Approaches, V.Types of Face Recognition Applications, VI.Observations, VII .Conclusion and References.
Keywords - Convolutional Neural Network (CNN), Face recognition, Back Propagation Neural Network (BPNN), Artificial Neural Networks (ANN), Multilayer Perceptron.