• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Advanced Computing and Communication Systems >Traffic Sign Detection using Clara and Yolo in Python
【24h】

Traffic Sign Detection using Clara and Yolo in Python

机译：在Python中使用克拉拉和yolo的交通标志检测

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

摘要

Research in the field of self-driving and autonomous vehicles is continuously growing. Many researchers are working to make self-driving cars more secure. Researchers work diligently to make cars as safe as possible to minimize fatal injuries on our roads. In about 30 percent of these incidents, there is a central factor: speed. For several decades, speeding has been a widely discussed subject for major automotive companies. The purpose of the project is to contribute to this research by implementing a driving simulator for a device that can understand speed limit signs and make decisions that make the driver more comfortable and safer to drive. The CARLA Learning to Act (CARLA), an open-source autonomous test simulator consisting primarily of two modules, the CARLA simulator and the Python API module, is used as a simulator in this analysis. The algorithm You Look Just Once (YOLO) is used to classifying road signs. Yolo sees the whole picture during training and testing, encoding contextual information about the groups of objects and their appearances instead of a sliding window over several places in an image. This feature makes it extremely easy to analyze an image. Several utilities may be used to identify the speed signs of a road. There are two applications for this project: a request for a notice warning the driver that the vehicle’s speed is above the maximum allowed speed of that traffic, and a request for a rule to reduce the vehicle’s speed when the traffic limit is reached.
机译：自动驾驶和自治车辆领域的研究不断增长。许多研究人员正在努力使自动驾驶汽车更加安全。研究人员努力使汽车尽可能能够尽量减少道路上的致命伤害。在大约30％的这些事件中，有一个核心因素：速度。几十年来，Speeding一直是主要的汽车公司讨论的主题。该项目的目的是通过为能够理解速度限制迹象的设备实现驾驶模拟器来促进这项研究，使得使驾驶员更舒适并且更安全驱动。 Carla学习到Act（Carla），一个主要是两个模块，Carla模拟器和Python API模块组成的开源自主测试模拟器，用作该分析中的模拟器。您只需一次（YOLO）的算法用于对道路标志进行分类。 YOLO在培训和测试期间看到整个图片，编码有关物体组的上下文信息及其外观，而不是图像中的几个位置的滑动窗口。此功能使其非常容易分析图像。可用于识别道路的速度迹象。该项目有两个应用程序：请求通知警告，警告车辆的速度高于该流量的最大速度速度，以及在达到流量限制时降低车辆速度的规则请求。

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