Deep Classification of Vehicle Makers and Models：The Effectiveness of Pre-teaining and DataEnhancement
文献类型：会议论文
| 作者 | Feiyun Zhang; Xiao Xu; Yu Qiao |
| 出版日期 | 2015 |
| 会议名称 | 2015 IEEE Conference on Robotics and Biomimetics（ROBIO） |
| 会议地点 | 广东珠海 |
| 英文摘要 | Vision-based vehicle detection and classification is an important problem in machine vision and received extensive interests due to its wide applications in intelligent traffic system. Most of current vehicle type classification methods require to precisely locate car positions and use the cropped car regions as input. In this paper, we propose deep convolutional neural networks for vehicle makers and models classification, which can take whole image as input without detecting car regions. Our main contributions are two-fold. Firstly, we find pre-train the deep CNN in the task to identify whether vehicle exists in input image can boost the performance of vehicle type classification. Secondly, we show data enhancement can further improve the classification accuracy. Our methods achieve the accuracy of 79% on a large scale Cars dataset, which is comparable with the recent state of the art which requires cropped car image as input. |
| 收录类别 | EI |
| 语种 | 英语 |
| 源URL | [http://ir.siat.ac.cn:8080/handle/172644/6707]  |
| 专题 | 深圳先进技术研究院_集成所 |
| 作者单位 | 2015 |
| 推荐引用方式 GB/T 7714 | Feiyun Zhang,Xiao Xu,Yu Qiao. Deep Classification of Vehicle Makers and Models：The Effectiveness of Pre-teaining and DataEnhancement[C]. 见:2015 IEEE Conference on Robotics and Biomimetics（ROBIO）. 广东珠海. |
入库方式： OAI收割
来源：深圳先进技术研究院
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。