arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computer Vision and Pattern Recognition

Deep Learning Approaches to Image Classification

M. Shanmuganathan, C. Nalini

· 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA)

See on the map

Abstract

Face recognition methodology (human face) have made many improvements in the emerging world have made many face identification is a technic of detecting a face through a technicality. It likens the image with a dataset of known images to encounter a match. Face identification authenticates personal identity, as well as raising secrecy challenges. Artificial neural networks (ANN) are used extensively in facial processing of images. Back-propagation neural networks (BPNNs) are mainly used for classification of facial images and also it achieves high accuracy. Nevertheless, there is a variation due to different neural designs and implementations. Accordingly, investigators usually need to do several observational trials before they can finalize the neural network design, but this is a slow process which considerably mitigates the efficiency of using the above-said method (BPNN) and the final pattern still not be quintessential. Despite the precision of image identification system is a biometric mechanism is lower than iris or fingerprint identification, it is commonly accepted its noncontact procedures. Newly, it had also been made widespread as a commercial and a business tool. As a result, there are certain guidelines for a valid design and operationalization of BPNNs. This work endeavours to find and abbreviate certain guidelines recommended by various researchers through specialized literature and observations. To contribute academicians with history and framework, somewhat ANN basics are also initiated.

More like this