首页> 外国专利> Method for detecting and recognizing traffic lights signal based on features

Method for detecting and recognizing traffic lights signal based on features

机译:基于特征的交通信号灯检测识别方法

摘要

The present invention relates to a method detection and recognition of a traffic light based on feature extraction of a traffic light for processing an image received from the imaging unit mounted on the vehicle to determine the location and color of a traffic light in the image, a frame from the video image photographing device attention specifying the area according to the received input in sequence predetermined criteria by region of interest region setting step, noise and binarization to convert the color space, and filter the signal color, and black color within the region of interest, present in the filter frame a When the lights present in the correction step, the color of the signal, location of the signal of interest from the traffic light determination step and said light determination step of using the shape of the form and the traffic light to determine the light existence in the area of interest of the signal to the region to eliminate having a value of in the region of interest setting step, the light does not exist in the area of interest If directed to a method of detecting and recognizing a traffic light based on feature extraction of a traffic light, characterized in that to release formed by a region of interest comprises a region of interest comprising: reset performed from the region of interest setting step again.;
机译:本发明涉及一种基于交通信号灯的特征提取的交通信号灯的检测和识别方法,用于处理从安装在车辆上的成像单元接收的图像以确定交通信号灯在图像中的位置和颜色。来自视频图像摄影设备注意的帧,通过关注区域设置步骤,根据接收到的输入,按照预定的标准依次指定区域,进行噪声和二值化以转换色彩空间,并对信号区域中的信号颜色和黑色进行过滤感兴趣,存在于滤光镜架中a当校正步骤中出现灯时,信号的颜色,来自交通信号灯确定步骤的感兴趣信号的位置以及使用表格形状和交通信号灯的所述光确定步骤来确定存在于信号感兴趣区域中的光是否存在,以消除在最重要的是,在目标区域中不存在该光。如果涉及一种基于交通信号灯的特征提取来检测和识别交通信号灯的方法,其特征在于,由目标区域形成的释放包括一个区域。兴趣包括:从关注区域设置步骤再次执行重置。

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