Object Detection Combining CNN and Adaptive Color Prior Features
When compared with the traditional manual design method, the convolutional neural network has the advantages of strong expressive ability and it is insensitive to scale, light, and deformation, so it has become the mainstream method in the object detection field. In order to further improve the accu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/8/2796 |
| _version_ | 1797537588036239360 |
|---|---|
| author | Peng Gu Xiaosong Lan Shuxiao Li |
| author_facet | Peng Gu Xiaosong Lan Shuxiao Li |
| author_sort | Peng Gu |
| collection | DOAJ |
| description | When compared with the traditional manual design method, the convolutional neural network has the advantages of strong expressive ability and it is insensitive to scale, light, and deformation, so it has become the mainstream method in the object detection field. In order to further improve the accuracy of existing object detection methods based on convolutional neural networks, this paper draws on the characteristics of the attention mechanism to model color priors. Firstly, it proposes a cognitive-driven color prior model to obtain the color prior features for the known types of target samples and the overall scene, respectively. Subsequently, the acquired color prior features and test image color features are adaptively weighted and competed to obtain prior-based saliency images. Finally, the obtained saliency images are treated as features maps and they are further fused with those extracted by the convolutional neural network to complete the subsequent object detection task. The proposed algorithm does not need training parameters, has strong generalization ability, and it is directly fused with convolutional neural network features at the feature extraction stage, thus has strong versatility. Experiments on the VOC2007 and VOC2012 benchmark data sets show that the utilization of cognitive-drive color priors can further improve the performance of existing object detection algorithms. |
| first_indexed | 2024-03-10T12:17:18Z |
| format | Article |
| id | doaj.art-5eb378ab4cd74bceb489c749fcb4ac8e |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T12:17:18Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-5eb378ab4cd74bceb489c749fcb4ac8e2023-11-21T15:45:03ZengMDPI AGSensors1424-82202021-04-01218279610.3390/s21082796Object Detection Combining CNN and Adaptive Color Prior FeaturesPeng Gu0Xiaosong Lan1Shuxiao Li2School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, ChinaInstitute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaSchool of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, ChinaWhen compared with the traditional manual design method, the convolutional neural network has the advantages of strong expressive ability and it is insensitive to scale, light, and deformation, so it has become the mainstream method in the object detection field. In order to further improve the accuracy of existing object detection methods based on convolutional neural networks, this paper draws on the characteristics of the attention mechanism to model color priors. Firstly, it proposes a cognitive-driven color prior model to obtain the color prior features for the known types of target samples and the overall scene, respectively. Subsequently, the acquired color prior features and test image color features are adaptively weighted and competed to obtain prior-based saliency images. Finally, the obtained saliency images are treated as features maps and they are further fused with those extracted by the convolutional neural network to complete the subsequent object detection task. The proposed algorithm does not need training parameters, has strong generalization ability, and it is directly fused with convolutional neural network features at the feature extraction stage, thus has strong versatility. Experiments on the VOC2007 and VOC2012 benchmark data sets show that the utilization of cognitive-drive color priors can further improve the performance of existing object detection algorithms.https://www.mdpi.com/1424-8220/21/8/2796convolutional neural networkcolor prior modelobject detection |
| spellingShingle | Peng Gu Xiaosong Lan Shuxiao Li Object Detection Combining CNN and Adaptive Color Prior Features Sensors convolutional neural network color prior model object detection |
| title | Object Detection Combining CNN and Adaptive Color Prior Features |
| title_full | Object Detection Combining CNN and Adaptive Color Prior Features |
| title_fullStr | Object Detection Combining CNN and Adaptive Color Prior Features |
| title_full_unstemmed | Object Detection Combining CNN and Adaptive Color Prior Features |
| title_short | Object Detection Combining CNN and Adaptive Color Prior Features |
| title_sort | object detection combining cnn and adaptive color prior features |
| topic | convolutional neural network color prior model object detection |
| url | https://www.mdpi.com/1424-8220/21/8/2796 |
| work_keys_str_mv | AT penggu objectdetectioncombiningcnnandadaptivecolorpriorfeatures AT xiaosonglan objectdetectioncombiningcnnandadaptivecolorpriorfeatures AT shuxiaoli objectdetectioncombiningcnnandadaptivecolorpriorfeatures |