Abstract: A real-time vehicle detection and recognition system based on the fusion of data from radar and a video sensor is presented.The radar data is used both for narrowing down the size of the search area for vehicle rears on the video image and for the distance measurement of the vehicles in front.Using the video sensor a radar object is verified and the width as well as the lateral position of the vehicle are determined.The contribution of this work is threefold：At first，a methodology for developing a novel evolutionary optimized symmetry measure is proposed.Secondly，a vehicle detection and recognition algorithm is demonstrated，which consists of two steps：Hypothesis generation using a detector based on a set of Haar-like filters and an AdaBoost learning algorithm and hypothesis verification using an evolutionary optimized and biologically motivated vehicle recognition system.Finally，the performance of the system is evaluated by not only using classical confusion matrices but also giving information on the accuracy of the width and lateral position sensing.Experimental results demonstrate a robust and real-time system trained and tested on more than 30 000 images.

Key words: vehicle recognition, evolutionary optimized, symmetry measurement

摘要： 提出了一种基于雷达和视觉传感器融合的实时车辆检测和识别系统。雷达数据用来测量前方车辆的距离和缩小图像中车辆尾部的搜索区域。利用视频传感器，可以验证雷达目标和确定车辆的宽度及横向位置。所做工作如下：第一，提出了一种新的进化优化对称测度方法；第二，阐述了一种车辆检测和识别算法，此算法包括两个步骤：（1）利用基于一组Haar滤波器和AdaBoost学习算法检测器的生成假设；（2）利用进化优化与生物机制车辆识别系统的验证假设。第三，利用经典的混淆矩阵及宽度和横向位置的准确度信息来对系统的性能做出评价。实验训练和测试了超过30 000幅图片，结果表明本系统具有良好的鲁棒性和实时性。

关键词: 车辆识别, 进化优化, 对称测度