TY - GEN
T1 - Scene recognition model in underground mines based on CNN-LSTM and spatial-temporal attention mechanism
AU - Zheng, Tianwei
AU - Li, Yuancheng
AU - Liu, Chi
AU - Wang, Pai
AU - Liu, Beizhan
AU - Qin, Xuebin
AU - Wang, Mei
AU - Guo, Yuan
N1 - Publisher Copyright:
© 2020 IEEE
PY - 2020/11
Y1 - 2020/11
N2 - Based on the convolutional neural network (CNN) and long short-term memory neural network (LSTM), combined with data enhancement technology and spatial-temporal attention mechanism, a scene recognition model is established. In order to balance the difference in the amount of data between different samples, data enhancement technology based on video data samples is introduced. Aiming at improving the performance of the model, a spatial-temporal attention mechanism is used to improve the accuracy of scene recognition. The model sample scenes used in this paper include three types: fully mechanized coal face, coal mine roadway and mining machinery. The experimental verification shows that: compared with the traditional convolutional neural network, the accuracy of the CNN-LSTM model is improved by 2.136%, and the CNN-LSTM model with spatial-temporal attention mechanism is improved by 2.921%. The accuracy of the deep CNN-LSTM model with spatial-temporal attention mechanism is about 93.063%.
AB - Based on the convolutional neural network (CNN) and long short-term memory neural network (LSTM), combined with data enhancement technology and spatial-temporal attention mechanism, a scene recognition model is established. In order to balance the difference in the amount of data between different samples, data enhancement technology based on video data samples is introduced. Aiming at improving the performance of the model, a spatial-temporal attention mechanism is used to improve the accuracy of scene recognition. The model sample scenes used in this paper include three types: fully mechanized coal face, coal mine roadway and mining machinery. The experimental verification shows that: compared with the traditional convolutional neural network, the accuracy of the CNN-LSTM model is improved by 2.136%, and the CNN-LSTM model with spatial-temporal attention mechanism is improved by 2.921%. The accuracy of the deep CNN-LSTM model with spatial-temporal attention mechanism is about 93.063%.
KW - CNN-LSTM
KW - Data enhancement
KW - Scene recognition
KW - Spatial-temporal attention mechanism
UR - https://www.scopus.com/pages/publications/85104893219
U2 - 10.1109/IS3C50286.2020.00139
DO - 10.1109/IS3C50286.2020.00139
M3 - Conference contribution
AN - SCOPUS:85104893219
T3 - Proceedings - 2020 International Symposium on Computer, Consumer and Control, IS3C 2020
SP - 513
EP - 516
BT - Proceedings - 2020 International Symposium on Computer, Consumer and Control, IS3C 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 International Symposium on Computer, Consumer and Control, IS3C 2020
Y2 - 13 November 2020 through 16 November 2020
ER -