【摘要】 视频检测是智能交通系统中一种重要的检测手段,但是运动车辆阴影的存在严重影响了检测效果。为了减少阴影对检测系统中交通参数计算的影响,采用了一种快速归一化的互相关函数(FNCC)直接对灰度视频图像检测运动阴影。通过引入三个加总表(sum-table)和设定阴影检测区使传统归一化互相关函数(NCC)算法的复杂度大大降低。实验表明该算法可以实时有效地检测出运动车辆的阴影。更多还原

【Abstract】 Video detection plays an important role in intelligent transportation system,and the shadows of moving vehicles have serious influence on object detection and segmentation.By adopting a FNCC(Fast Normalized Cross-Correlation) algorithm in this paper,moving shadows were directly detected from grayscale video sequences.By applying three sum-tables in FNCC and detecting shadows within a confined range,the computational complexity had been significantly reduced compared with the traditional NCC(Normalized Cross-Correlation) algorithm.And the experimental results have shown that this method can detect moving vehicles’ shadows efficiently and accurately.更多还原