• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Joint Conference on Web Intelligence and Intelligent Agent Technology >An Inclined Bounding Box Method for Detecting Parking Spaces in Open Parking Lots using Aerial Images
【24h】

An Inclined Bounding Box Method for Detecting Parking Spaces in Open Parking Lots using Aerial Images

机译：一种倾斜边界盒方法，用于使用空中图像检测开放式停车场中的停车位

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

With the increasing population in urban areas, finding a vacant parking space has always been the biggest problem for drivers, which causes a series of issues such as energy waste and environmental pollution. Most outdoor parking lots cannot provide details of vacant parking spaces due to limited sensor technology in outdoor environments (i.e., poor resistance to external interference due to heat, light or signal noise). Although some studies are based on images captured by cameras, the existing vacancy detection methods using fixed-position surveillance cameras are not flexible and accurate enough for wide use. We propose an Inclined Bounding Box (IBB) method for detecting vacant parking spaces using aerial images. The IBB method takes advantage of the feature-description ability of deep Convolutional Neural Network (CNN) to identify the status of parking spaces and develops an inclined bounding boxes calibration algorithm on the top layer of CNN architecture. The proposed IBB method can automatically identify parking spaces, locate vacant parking spaces with any direction and detect multiple parking spaces simultaneously. The experimental results show that the proposed IBB method is more accurate than the three counterpart methods.
机译：随着城市地区人口的增加，寻找空置停车位始终是司机最大的问题，这导致了一系列问题，如能量浪费和环境污染。由于室外环境中的传感器技术有限（即，由于热量，光线或信号噪声，由于热量，光线或信号噪声导致的外部干扰耐受耐受性，因此，大多数室外停车位无法提供空置停车位的细节。虽然一些研究基于摄像机捕获的图像，但是使用定位监控摄像机的现有空位检测方法并不灵活，足以准确，足以广泛使用。我们提出了一种倾斜的边界框（IBB）方法，用于使用空中图像检测空置停车位。 IBB方法利用了深度卷积神经网络（CNN）的特征描述能力，以识别停车位的状态，并在CNN架构的顶层开发倾斜边界框校准算法。所提出的IBB方法可以自动识别停车位，找到具有任何方向的空置停车位，并同时检测多个停车位。实验结果表明，所提出的IBB方法比三个对应方法更准确。

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