摘要
针对Vi Be算法在交通视频检测中出现明显鬼影区域、缓慢目标残影难以消除、检测精确度和鲁棒性不足的问题,提出改进算法,利用灰度信息为像素建立生命长度矩阵,使鬼影或残影快速融入背景样本得以消除。结合最大类间方差法设置自适应阈值,加入良好后处理抑制动态噪声。同时借鉴分类算法的统计指标,提出质量评价多个要素,以Vi Be原算法、混合高斯算法(GMM)、LBP-Otsu相结合的背景差分法和该改进算法为例,定性、定量对实验结果作出质量评价和分析。实验表明,改进算法在较少帧数内去除了鬼影,抑制了运动目标残影,提高了运动目标检测的准确度和整体性能。
There were some problems existing in the Vi Be algorithm based in traffic video processing,such as ghosting shadow appeared in video processing,difficulty of removing residual shadow of slow moving targets,and poor detection accuracy and robustness. This paper proposed an improved Vi Be algorithm. It used the gray-scale spatial information to build matrix of pixel life length to make ghosting or target's residual shadow quickly blended into the samples of the background. It also combined with the Otsu method to set the dynamic threshold. The improved algorithm took the good post-processing method to restrain dynamic noise. The paper also proposed several quality evaluation criteria based on statistics index of classification algorithms.The paper evaluated the experiment results qualitatively and quantitatively of our improved algorithm comparing with the Vi Be algorithm,the GMM algorithm,the combination of LBP-Otsu algorithm. Experiment results show that the improved algorithm removes the ghost shadows and restrains the residual shadow of moving objects within less frames. It also promoted the accuracy and overall performance of moving object detection.
出处
《计算机应用研究》
CSCD
北大核心
2017年第12期3843-3847,共5页
Application Research of Computers
基金
国家自然科学基金资助项目(61302127)
江西省自然科学基金重大项目(20161ACB20004)
关键词
运动目标检测
Vi
Be改进算法
交通视频
阈值划分
moving object detection
improved Vi Be algorithm
traffic video
threshold segmentation