Article Details

Abstract

At present, the existing communication technologies for the deaf-mute individuals have problems such as low recognition accuracy and cumbersome equipment. This paper proposes a sign language gesture recognition system based on YOLOv5. Firstly, preprocessing such as grayscale conversion is carried out on the collected image information. The built-in functions of the OpenCV library are utilized to divide the images into local blocks and extract features, and RGB conversion is performed on the extracted samples. Secondly, the processed images are input into the model. The Adam optimizer is used to dynamically adjust the learning rate, and the K-means algorithm is adopted to calculate the appropriate anchor box values, thus improving the training speed of the model. Finally, prediction correction is carried out on the images output by the model. Redundant bounding boxes are removed through non-maximum suppression, and the final prediction results are retained. Experiments show that the training model based on YOLOv5 has a significant improvement in the accuracy and recognition rate of sign language gesture recognition.

Keywords

YOLOv5; gesture recognition; OpenCV; Adam

References

1. [1] Luo Longyu, Wang Yuexin Goods Identification System Based on YOLOv5 [J]. Journal of Longyan University, 2024, 42 (5): 31-37
2. [2] Zhang Siyu Research on Gesture Recognition Technology Based on Deep Learning [D]. Xi'an: Xijing University, 2023
3. [3] Wang Ge, Qian Xueming, Luo Zhengang, etc Computer Vision based Sign Language Bi directional Communication System [J]. Internet of Things Technology, 2023, 13 (10): 59-62
4. [4] Gao Qizhi Research on Sign Language Recognition Algorithm Based on Deep Learning [D]. Nanjing: Nanjing University of Posts and Telecommunications, 2023
5. [5] Liu Xingzheng Research on Gesture Recognition Algorithm Based on Improved YOLOv7 [D]. Huainan: Anhui University of Science and Technology, 2023
6. [6] Wang Rui Research on Traffic Sign Recognition Algorithm Based on Machine Vision [D]. Changsha: Changsha University of Science and Technology, 2013
7. [7] Yang Haohao Application and Comparison of Several Machine Learning Algorithms and Their Ensemble Models in Regression Problems [D]. Lanzhou: Lanzhou University, 2018
8. [8] Li Fangjie Research on Track Obstacle Recognition Algorithm Based on Deep Learning [D]. Zhengzhou: Zhengzhou University, 2021
9. [9] Xu Changkai Research on Object Detection Based on Optimized Boundary Box Regression [D]. Qingdao: Qingdao University of Science and Technology, 2021
10. [10] Zhang Dongyang, Lu Zixuan, Liu Junmin, etc A review of continuous learning of deep models: theory, methods, and applications [J]. Journal of Electronics and Information Technology, 2024, 46 (10): 3849-3878
11. [11] Wang Jing, Du Xu, Li Hao, etc Automatic annotation algorithm for knowledge points in image based educational resources for knowledge scenarios [J]. Computer Engineering and Applications, 2024, 60 (24): 119-130
12. [12] Yu Changfu Research on PCB Defect Detection Algorithm Based on Improved YOLOv7 [D]. Yantai: Yantai University, 2024
13. [13] Lu Yang, Chen Linhui, Jiang Xiaoheng, etc SDENet: A Synthetic Defect Data Evaluation Network Based on Multi scale Attention Quality Perception [J]. Journal of Cartography, 2025, 46 (1): 94-103
14. [14] Zhang Yi Research on Adaptive Reasoning Techniques for Deep Learning Models in Dynamic Edge Environments [D]. Beijing: Beijing University of Posts and Telecommunications, 2023
15. [15] Hao Wangli, Wei Peiyan, Hao Fei, etc Grain ear detection method based on YOLOv4 and adaptive anchor frame adjustment [J]. Smart Agriculture: Chinese English, 2021, 3 (1): 63-74
16. [16] Wang Bin Research and Implementation of Web Instant Messaging System Based on WebSocket Protocol [J]. Wireless Internet Technology, 2023, 20 (24): 4-7

About the authors

Qin Boming (2002-), male, Zhuang nationality, from Laibin, Guangxi, undergraduate student, research direction: visual application of machine learning; Fu Wensi (2003-), female, Han ethnicity, from Yulin, Guangxi, is currently pursuing a bachelor's degree. Her research interests include digital signal processing, communication system design, and wireless communication; Tang Zihang (2004-), male, Han ethnicity, from Zhengzhou, Henan Province, is currently pursuing a bachelor's degree with a research focus on multi scene applications of machine vision.