TY - GEN
T1 - Design of Human Posture Recognition System Integrating Computer Vision and Deep Learning Algorithm
AU - Chu, Jinquan
AU - Wang, Haoyi
AU - Dong, Anxing
AU - Chen, Yuming
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - In this article, a human posture recognition model combining computer vision and deep learning (DL) algorithm is proposed. Under the condition of keeping the data structure and sample labels unchanged, the model randomly cuts the video frame images by using the horizontal translation matrix, which effectively expands the data set samples, thus reducing the risk of over-fitting of the model to some extent. The experimental results show that the improved human target detection algorithm in this article can effectively improve the detection performance of the algorithm, and the computational complexity of the human posture recognition algorithm is small, which increases the correlation between adjacent video frames in the recognition process, and improves the accuracy of posture recognition while ensuring real-time. The realization of this algorithm is of great significance for building a simplified human posture recognition system with high response and accuracy.
AB - In this article, a human posture recognition model combining computer vision and deep learning (DL) algorithm is proposed. Under the condition of keeping the data structure and sample labels unchanged, the model randomly cuts the video frame images by using the horizontal translation matrix, which effectively expands the data set samples, thus reducing the risk of over-fitting of the model to some extent. The experimental results show that the improved human target detection algorithm in this article can effectively improve the detection performance of the algorithm, and the computational complexity of the human posture recognition algorithm is small, which increases the correlation between adjacent video frames in the recognition process, and improves the accuracy of posture recognition while ensuring real-time. The realization of this algorithm is of great significance for building a simplified human posture recognition system with high response and accuracy.
KW - Computer vision
KW - Deep learning
KW - Human posture recognition
KW - Object detection
UR - http://www.scopus.com/inward/record.url?scp=85186111812&partnerID=8YFLogxK
U2 - 10.1109/PEEEC60561.2023.00096
DO - 10.1109/PEEEC60561.2023.00096
M3 - Conference proceedings published in a book
AN - SCOPUS:85186111812
T3 - Proceedings - 2023 International Conference on Power, Electrical Engineering, Electronics and Control, PEEEC 2023
SP - 465
EP - 469
BT - Proceedings - 2023 International Conference on Power, Electrical Engineering, Electronics and Control, PEEEC 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 International Conference on Power, Electrical Engineering, Electronics and Control, PEEEC 2023
Y2 - 25 September 2023 through 27 September 2023
ER -