INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Vision-based human activity recognition in privately owned spaces and public places has become a significant issue in terms of developing safety feature to avoid theft like crimes. This technology improves security of privately owned spaces and public places. Deep learning models aim to automate extraction of low-level to high-level features of input data using convolutional feature instead of using aged complicated extraction methods. Convolutional feature has achieved significant improvements in classification of large amount of data especially in vision-based datasets. In order to recognize human action in smart environment, DMLSmartActions dataset is used for monitoring human activities. In deep learning, convolutional neural networks (CNNs) architecture is a class of deep neural networks commonly used to analyze visual imagery. The performance of the proposed method has been compared with previous methods that have used traditional machine learning methods on the same dataset. Experimental results demonstrated that the proposed deep learning model has achieved 82.41% accuracy rate in the classification of human activity which is far better than traditional machine learning method.
Keywords:
DMLSmart Actions, Human Activity Recognition, Convolutional Neural Network, Human Image Threshing, Facial Image Threshing.
Cite Article:
"Human Activity Recognition using Deep Learning", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.8, Issue 1, page no.1295 - 1297, January-2023, Available :http://www.ijsdr.org/papers/IJSDR2301209.pdf
Downloads:
000337212
Publication Details:
Published Paper ID: IJSDR2301209
Registration ID:203822
Published In: Volume 8 Issue 1, January-2023
DOI (Digital Object Identifier): http://doi.one/10.1729/Journal.32956
Page No: 1295 - 1297
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn