• L. Maria Anthony Kumar
• Dr. S. Murugan
• A. Therasa Alphonsa

Human Activity Recognition (HAR) is a vital task in the domain of computer vision and pattern recognition, with applications classifying healthcare to human-computer interaction. It contains the classification and detection of many human activities depending on visual input or sensor data. HAR has accomplished vital attention because of its potential in several domains comprising fitness tracking, surveillance, human-computer interaction, and healthcare. The deep learning (DL) structures excel in learning difficult features and temporal patterns in raw sensor data, allowing the formation of sophisticated methods that accurately identify a widespread of human activities. This study presents a novel approach to HAR by combining crow search algorithm with hybrid DL (HAR-CSAHDL) technique. The HDL technique involves Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. Additionally, to enhance the model's performance, hyperparameter tuning is employed using the CSA, an optimization technique simulated by the foraging behavior of crows. The HAR-CSAHDL architecture leverages the strengths of CNNs in feature extraction from raw sensor data and the sequential learning capabilities of LSTM networks to capture temporal dependencies in human activities. The performance of the HAR-CSAHDL approach is demonstrated through comprehensive experiments on benchmark HAR datasets. Comparative analyses with recent approaches highlight the superior performance of the CNN-LSTM hybrid model.

"Modeling of Human Activity Recognition using Crow Search Algorithm with Hybrid Deep Learning Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 9, page no.a360-a372, September-2023, Available :http://www.jetir.org/papers/JETIR2309044.pdf

"Modeling of Human Activity Recognition using Crow Search Algorithm with Hybrid Deep Learning Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 9, page no. ppa360-a372, September-2023, Available at : http://www.jetir.org/papers/JETIR2309044.pdf