Traffic sign recognition algorithm based on shape signature and dual-tree complex wavelet transform

A novel traffic sign recognition system is presented in this work. Firstly, the color segmentation and shape classifier based on signature feature of region are used to detect traffic signs in input video sequences. Secondly, traffic sign color-image is preprocessed with gray scaling, and normalized to 64×64 size. Then, image features could be obtained by four levels DT-CWT images. Thirdly, 2DICA and nearest neighbor classifier are united to recognize traffic signs. The whole recognition algorithm is implemented for classification of 50 categories of traffic signs and its recognition accuracy reaches 90%. Comparing image representation DT-CWT with the well-established image representation like template, Gabor, and 2DICA with feature selection techniques such as PCA, LPP, 2DPCA at the same time, the results show that combination method of DT-CWT and 2DICA is useful in traffic signs recognition. Experimental results indicate that the proposed algorithm is robust, effective and accurate.     

基于无参数形状检测子和DT-CWT的交通标志识别

针对车辆行驶环境中难以检测的交通标志,提出了一种检测和识别方法.首先分割交通标志的特征颜色区域,并扩展感兴趣区域,提取区域边缘.然后用直线分割和杂点去除粗略划分边缘,根据直线间顶点处的曲率关系,计算转向角并分类顶点的类型,用无参数形状检测子来检测图像中的圆形、三角形和矩形等.将检测到的候选区域送入形状分类器中,分类形状并排除杂质的干扰,最后通过二元树复小波变换和二维独立分量分析相结合来识别交通标志类型.实验结果表明提出的方法对交通标志被遮挡、光照不均匀、颜色部分失真的情况下,检测率和识别率均较高,并且可以达到实时处理的效果.     

混合阶多智能体无向网络的拟平均一致性

为解决混合阶动态多智能体网络的拟平均一致性,将混合阶动态多智能体网络的拟平均一致性转化为增广的一阶多智能体网络的平均一致性,并给出相应一致性问题协议的定义,根据网络的拓扑结构连通性特点,运用代数图论、矩阵理论中的圆盘定理和李雅普诺夫理论证明了定拓扑和变拓扑无向网络下一致性协议的稳定性定理,并以由4个一阶智能体和2个二阶智能体所组成的混合阶智能体网络为例进行具体说明.理论研究和仿真结果都表明所提协议的正确性.     

应用圆形度和颜色直方图的交通信号灯识别

针对智能车辆介绍了一种在城市环境中识别交通信号灯的新方法.该方法先对原图像进行预处理,过滤不符合形态学准则的区域.计算候选区域的圆形度和背板的颜色信息,把圆形度符合和背板为黑色的区域作为交通信号灯区域.并将检测到的交通信号灯区域图像从RGB空间转换到HSV空间中,用颜色直方图对图像的H分量进行分布统计,根据其分布特征来识别交通信号灯的类型.实验结果表明,该方法能有效地识别复杂环境下的交通信号灯.     

Performance assisted enhancement based on change point detection and Kalman filtering

A performance assisted enhancement Kalman filtering algorithm （PAE-KF） for GPS/INS integration navigation in urban areas was presented in this work. The aim of this PAE-KF algorithm was to prevent ＂deep contamination＂ caused by error GPS data. This filtering algorithm effectively combined fault estimation of raw GPS data and nonholonomic constraint of vehicle. In fault estimation, a change point detection algorithm based on abrupt change model was proposed. Statistical tool was then used to infer the future bound of GPS data, which can detect faults in GPS raw data. If any kinds of faults were detected, dead reckoning mechanism begins to compute current position. Nonholonomic constraint condition of vehicle was used to estimate velocity of vehicle and change point detection was added into classic Kalman filtering structure. Experiment on vehicle shows that even when the GPS signals are unavailable for a period of time, this method can also output high accuracy data.     

基于多模型表示的交通标志识别算法设计

通过变换 RGB 空间颜色值并分割图像,利用标记图为特征的形状分类器检测城市环境中的交通标志.为了提高分类准确度,用两种模型表示方法分类交通标志:1)结合二元树复小波变换和二维独立分量分析提取特征,送入最近邻分类器中分类交通标志;2)提取交通标志的内部图形,利用模板匹配进行分类.最后,将两种分类结果融合输出.实验结果表明,所提出算法的整体识别率超过91%,平均处理帧率达到6.6帧/s,系统能够鲁棒、有效和实时地识别交通标志.     

形状标记图和Gabor小波的交通标志识别

交通标志识别为智能车辆行驶提供了有价值的道路环境信息.提出一种结合形状标记图和Gabor波的交通标志识别方法,交通标志识别过程如下:1)变换图像的RGB像素值来增强交通标志主特征颜色(红,蓝,黄)区域并进行分割,用形态学操作消除噪声点的影响;2)提取感兴趣区域的标记图作为其形状特征,用Euclidean距离来对其进行初分类;3)对交通标志感兴趣区域的灰度图像进行Gabor小波变换,获得其不同角度和尺度的小波图像,用二维独立分量分析法提取其主特征,并送入线性支持向量机来判断感兴趣区域所属的交通标志类型.实验结果表明,提出的算法能够稳定、有效地检测和识别智能车辆行驶环境中的多类交通标志.     

离散时间混合多智能体的拟平均一致性控制

为了研究混合一阶和二阶异质无向多智能体网络离散时间一致性问题,提出了固定和可变拓扑结构的多智能体网络的拟平均一致性协议.根据Lyapunov函数和代数图论,分析了多智能体网络的稳定性,表明在固定和可变拓扑结构的情况下,混合阶多智能体系统可以达到拟平均一致.对6个节点组成的4种不同的无向网络拓扑结构进行了仿真,结果显示了变拓扑离散时间异质多智能体网络的位置和速度均能达到拟平均一致.仿真结果表明,该算法对混合阶多智能体网络的拟平均一致性控制有效.     

面向无人驾驶车感知系统的自动曝光方法

为了采集清晰且稳定的图像,提出了一种相机自动曝光方法.首先在初始帧图像中检测交通标志,根据交通标志的检测结果,在划分的9个区域中确定感兴趣区域.利用HSV颜色空间中V分量直方图来准确判断图像的曝光情况,并依据感兴趣区域的位置和曝光情况来选择权值矩阵.最后利用灰度值方法获取下一帧图像的曝光时间,完成车载相机的自动曝光.实验结果表明该方法能够快速、有效地完成相机的自动曝光,并对逆光和背光具有一定的适应性.