TY - GEN
T1 - A novel approach to cloth classification through deep neural networks
AU - Fengxin, Li
AU - Yueping, Li
AU - Xiaofeng, Zhang
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/2
Y1 - 2017/7/2
N2 - The recent development of the field of artificial intelligence makes the traditional technical recognition more accurate. An important area is commodity identification which helps to classify commodity and provide information for data-ming and commercial decision. This paper considers cloth classification by means of deep neural networks. We summarize the existing methods: the effect improvement of network can be divided into two kinds, by modifying network structure according to their priorities, i.e., increase the depth of network and enhance the performance of convolution unit. In order to further improve the performance of network model, we redesign the network structure based on AlexNet, and put forward the deep convolution neural network model. Experiments are performed on the data sets including ImageNet-1000 and cloth data sets ACS and CAPB. The results show that the proposed deep convolutional neural network is superior to the original AlexNet on these three data sets in terms of accuracy.
AB - The recent development of the field of artificial intelligence makes the traditional technical recognition more accurate. An important area is commodity identification which helps to classify commodity and provide information for data-ming and commercial decision. This paper considers cloth classification by means of deep neural networks. We summarize the existing methods: the effect improvement of network can be divided into two kinds, by modifying network structure according to their priorities, i.e., increase the depth of network and enhance the performance of convolution unit. In order to further improve the performance of network model, we redesign the network structure based on AlexNet, and put forward the deep convolution neural network model. Experiments are performed on the data sets including ImageNet-1000 and cloth data sets ACS and CAPB. The results show that the proposed deep convolutional neural network is superior to the original AlexNet on these three data sets in terms of accuracy.
KW - classification
KW - convolutional neural network
KW - deep neural networks
KW - network model
UR - https://www.scopus.com/pages/publications/85050611319
U2 - 10.1109/SPAC.2017.8304306
DO - 10.1109/SPAC.2017.8304306
M3 - 会议稿件
AN - SCOPUS:85050611319
T3 - 2017 International Conference on Security, Pattern Analysis, and Cybernetics, SPAC 2017
SP - 368
EP - 371
BT - 2017 International Conference on Security, Pattern Analysis, and Cybernetics, SPAC 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 International Conference on Security, Pattern Analysis, and Cybernetics, SPAC 2017
Y2 - 15 December 2017 through 17 December 2017
ER -