Laser & Optoelectronics Progress, Volume. 59, Issue 8, 0811005(2022)
Infrared Human Gait Recognition Method Based on Long and Short Term Memory Network
Fig. 1. Original infrared image and human contour image after binarization. (a) Original image; (b) binarized image; (c) centered and clipped image
Fig. 2. Proportion of each part of human body
Fig. 3. Gait image sequence after occlusion processing
Fig. 4. Gait cycle of each individual in different walking states. (a) Backpack walking; (b) normal walking; (c) fast walking; (d) slow walking
Fig. 5. Multi-branch convolutional neural network
Fig. 6. Standard cyclic neural network structure
Fig. 7. LSTM gait recognition model
Fig. 8. Comparison of recognition accuracy between CNN and LSTM
Get Citation
Copy Citation Text
Jianhua Mei, Lijun Yun, Xiaopeng Zhu. Infrared Human Gait Recognition Method Based on Long and Short Term Memory Network[J]. Laser & Optoelectronics Progress, 2022, 59(8): 0811005
Category: Imaging Systems
Received: Mar. 22, 2021
Accepted: Apr. 28, 2021
Published Online: Apr. 11, 2022
The Author Email: Yun Lijun (yunlijun@ynnu.edu.cn)