Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network
The interference of the complex background and less information of the small targets are two major problems in vehicle attribute recognition. In this paper, two cascaded networks of vehicle attribute recognition are established to solve the two problems. For vehicle targets with normal size, the mul...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2019-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2019/6409630 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832552021037678592 |
|---|---|
| author | Fang Liu Yong Zhang Hua Gong Ke Xu Ligang Cai |
| author_facet | Fang Liu Yong Zhang Hua Gong Ke Xu Ligang Cai |
| author_sort | Fang Liu |
| collection | DOAJ |
| description | The interference of the complex background and less information of the small targets are two major problems in vehicle attribute recognition. In this paper, two cascaded networks of vehicle attribute recognition are established to solve the two problems. For vehicle targets with normal size, the multitask cascaded convolution neural network MC-CNN-NT uses the improved Faster R-CNN as the location subnetwork. The vehicle targets in the complex background are extracted by the location subnetwork to the classification subnetwork CNN for the classification. The implementation of this task decomposition strategy effectively eliminates the interference of the complex background in target detection. For vehicle targets with small size, the multitask cascaded convolution neural network MC-CNN-ST applies the network compression strategy and the multilayer feature fusion strategy to extract the feature maps. These strategies enrich the location information and semantic information of the feature maps. In order to optimize the nonlinear mapping ability and the hard-to-detect samples mining ability of the networks, the activation function and the loss function in the two cascaded networks are improved. The experimental results show that MC-CNN-NT for the normal targets and MC-CNN-ST for the small targets achieve the state-of-the-art performance compared with other attribute recognition networks. |
| format | Article |
| id | doaj-art-7b0058ed1661414ca59b938eba92ca21 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2019-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-7b0058ed1661414ca59b938eba92ca212025-02-03T05:59:41ZengWileyComplexity1076-27871099-05262019-01-01201910.1155/2019/64096306409630Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded NetworkFang Liu0Yong Zhang1Hua Gong2Ke Xu3Ligang Cai4College of Science, Shenyang Ligong University, Shenyang 110159, ChinaTechnology on Electro-Optical Information Security Control Laboratory, Tianjing 300308, ChinaCollege of Science, Shenyang Ligong University, Shenyang 110159, ChinaCollege of Science, Shenyang Ligong University, Shenyang 110159, ChinaCollege of Science, Shenyang University of Technology, Shenyang 110178, ChinaThe interference of the complex background and less information of the small targets are two major problems in vehicle attribute recognition. In this paper, two cascaded networks of vehicle attribute recognition are established to solve the two problems. For vehicle targets with normal size, the multitask cascaded convolution neural network MC-CNN-NT uses the improved Faster R-CNN as the location subnetwork. The vehicle targets in the complex background are extracted by the location subnetwork to the classification subnetwork CNN for the classification. The implementation of this task decomposition strategy effectively eliminates the interference of the complex background in target detection. For vehicle targets with small size, the multitask cascaded convolution neural network MC-CNN-ST applies the network compression strategy and the multilayer feature fusion strategy to extract the feature maps. These strategies enrich the location information and semantic information of the feature maps. In order to optimize the nonlinear mapping ability and the hard-to-detect samples mining ability of the networks, the activation function and the loss function in the two cascaded networks are improved. The experimental results show that MC-CNN-NT for the normal targets and MC-CNN-ST for the small targets achieve the state-of-the-art performance compared with other attribute recognition networks.http://dx.doi.org/10.1155/2019/6409630 |
| spellingShingle | Fang Liu Yong Zhang Hua Gong Ke Xu Ligang Cai Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network Complexity |
| title | Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network |
| title_full | Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network |
| title_fullStr | Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network |
| title_full_unstemmed | Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network |
| title_short | Vehicle Attribute Recognition for Normal Targets and Small Targets Based on Multitask Cascaded Network |
| title_sort | vehicle attribute recognition for normal targets and small targets based on multitask cascaded network |
| url | http://dx.doi.org/10.1155/2019/6409630 |
| work_keys_str_mv | AT fangliu vehicleattributerecognitionfornormaltargetsandsmalltargetsbasedonmultitaskcascadednetwork AT yongzhang vehicleattributerecognitionfornormaltargetsandsmalltargetsbasedonmultitaskcascadednetwork AT huagong vehicleattributerecognitionfornormaltargetsandsmalltargetsbasedonmultitaskcascadednetwork AT kexu vehicleattributerecognitionfornormaltargetsandsmalltargetsbasedonmultitaskcascadednetwork AT ligangcai vehicleattributerecognitionfornormaltargetsandsmalltargetsbasedonmultitaskcascadednetwork |