您的位置:首页  > 论文页面

复杂场景中的运动目标检测研究

发表时间:2017-08-31  浏览量:1824  下载量:381
全部作者: 郑毅,陈林,徐栋,徐先瑞,李响
作者单位: 华东师范大学地理科学学院
摘 要: 基于前人的研究,提出一种新的自适应背景差分法。该方法根据样本一致性和样本差异性的统计模版,先对图像进行初始分割,再结合错误检测抑制模型消除一些误检区域,最后基于区域边界评价的确定性,利用前景更新和概率性更新策略实现背景模型的更新。实验表明,本文的检测方法能够实现复杂场景的检测,相比于大部分检测算法,本文算法具有较好的鲁棒性,同时满足实时检测的要求。
关 键 词: 计算机应用;背景差分法;目标检测;图像处理
Title: Research on moving target detection in complex environment
Author: ZHENG Yi, CHEN Lin, XU Dong, XU Xianrui, LI Xiang
Organization: School of Geographic Sciences, East China Normal University
Abstract: We propose a newly adaptive background subtraction for moving targets detection. First, we build a sample disparity and a sample consensus statistical templates in the initial segmentation phase. Second, a false positive suppression model is added into the detection method. Last, regional boundary and probabilistic update strategy are used to implement background model update based on the regional boundary evaluation certainty. Experiment demonstrates that the proposed method is suitable for the most complex environment and outperforms the most existing detection methods. The proposed algorithm has good robustness and satisfies the requirements of real-time detection.
Key words: computer applications; background subtraction; target detection; image processing
发表期数: 2017年8月第16期
引用格式: 郑毅,陈林,徐栋,等. 复杂场景中的运动目标检测研究[J]. 中国科技论文在线精品论文,2017,10(16):1842-1849.
 
0 评论数 0
暂无评论
友情链接