【24h】

Accident Elements Detection Based on Improved DPM

机译:基于改进DPM的事故元素检测

获取原文

摘要

Deformable Part Models (DPM) has high detection accuracy in the image detection algorithm, while it is at the expense of complex calculation and slowly running speed. In order to improve the development of its application, the algorithms of the weighted PCA (MPCA) and fast level locating are also proposed, MPCA can reduce the model parameters by reducing the dimensions of HOG features and the fast level locating can locate the object's level in the feature pyramid fast. This paper also presents the optical accidents elements detection for the scene of the accident which can be used in varying environment. Our results show that the detection of the accidents elements obtained are sufficiently robust to the changing scene of accident and its speed increases significantly.
机译:可变形零件模型(DPM)在图像检测算法中具有高检测精度,而其牺牲复杂计算和缓慢运行速度为代价。为了改善其应用的发展,还提出了加权PCA(MPCA)和快速定位的算法,MPCA可以通过减少猪特征的尺寸来减少模型参数,并且快速定位可以定位对象的级别在特征金字塔快速。本文还介绍了可用于不同环境的事故现场的光学事故元素检测。我们的结果表明,获得的事故元素的检测对于发生变化的事故现场并且其速度显着增加了足够强大。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号