首页> 外国专利> METHOD AND APPARATUS FOR DETECTING SPEED-LIMIT SIGN AND RECOGNIZING SPEED LIMIT IN REAL TIME BY USING SPATIAL PYRAMID FEATURES

METHOD AND APPARATUS FOR DETECTING SPEED-LIMIT SIGN AND RECOGNIZING SPEED LIMIT IN REAL TIME BY USING SPATIAL PYRAMID FEATURES

机译:利用空间金字塔特征实时检测限速标志和识别限速的方法和装置

摘要

The present invention relates to a method and apparatus for detecting a speed-limit sign and recognizing a speed limit in real time by using spatial pyramid features. More specifically, the method for detecting a speed-limit sign and recognizing a speed limit in real time by using spatial pyramid features comprises: (1) a step (S100) of extracting a candidate region for a speed-limit sign from an input image; (2) a step (S200) of determining whether the extracted candidate region corresponds to a speed-limit sign; and (3) a step (S300) of recognizing a speed limit of the speed-limit sign. According to the method and apparatus for detecting a speed-limit sign and recognizing a speed limit in real time by using spatial pyramid features, which are proposed by the present invention, spatial pyramid features and a BoostRandomForest function are used to extract a speed-limit sign and to recognize a speed limit of the extracted speed-limit sign, thereby improving speed required to extract the speed-limit sign and to recognize the speed limit of the extracted speed-limit sign. Furthermore, according to the present invention, a speed-limit sign is extracted in real time, and a speed related to the speed-limit sign is rapidly recognized. Accordingly, the newest traffic information is provided to a driver, thereby improving convenience and safety of the driver.;COPYRIGHT KIPO 2017
机译:本发明涉及一种通过使用空间金字塔特征来检测限速标志并实时识别限速的方法和设备。更具体地,通过使用空间金字塔特征来检测限速标志并实时识别限速的方法包括:(1)从输入图像中提取限速标志的候选区域的步骤(S100)。 ; (2)判断提取出的候选区域是否与限速符号对应的步骤(S200)。 (3)识别限速标志的限速的步骤(S300)。根据本发明提出的利用空间金字塔特征检测限速标志并实时识别限速的方法和装置,利用空间金字塔特征和BoostRandomForest函数提取限速符号和识别提取的限速符号的限速,从而提高提取限速符号和识别提取的限速符号的限速所需的速度。此外,根据本发明,实时提取限速标志,并且快速识别与限速标志有关的速度。因此,将最新的交通信息提供给驾驶员,从而提高了驾驶员的便利性和安全性。; COPYRIGHT KIPO 2017

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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