针对车载智能监控的要求,设计一种嵌入式车体移动报警安全监控系统。系统以TMS320DM642芯片为核心,采用简化的Itti模型来提取监控画面的颜色和亮度特征,并检测和分割出感兴趣区域(Regions of Interest)。根据ROI质心相似度的变化来判断车辆状态。实验结果表明:该系统较为准确地检测和分割感兴趣区域,并能对其进行实时地监控;系统对车辆状态的判断和报警可靠性较高。
Due to low parameter sensitivity for balanced realiza- tions, balanced structure becomes a good candidate for an statespace adaptive infinite impluse response （IIR） filter. Here, using coefficients of the transfer function as the adaptive filtering parameters, a balanced adaptive IIR filtering algorithm is proposed for output-error minimization. The algorithm in the internally balanced realization guarantees that the adaptive IIR filter always minimizes the ratio of maximum-to-minimum eigenvalue of the Grammian matrices at the each iteration. Simulation results are provided to corroborate the proposed algorithm.
针对传统车辆检索方法中存在准确性和区分度较低的问题,提出了一个基于改进SURF(speeded up robust features)算法的视频车辆检索方法。在车辆视频关键帧提取的基础上,根据改进SURF算法完成车辆图像的特征提取及匹配,其中包含改进FAST(features from accelerated segment test)特征点检测、SURF特征向量提取以及最近邻查询方法来进行特征点的匹配;通过计算比较待检索车辆图像与数据库车辆图像的相似度,算法完成图像筛选并反馈检索结果。实验结果表明:针对交通监控视频中待检索车辆,该方法能够较为准确地进行检索并反馈结果。
