基于二维主成分分析的运动目标检测

运动目标检测是计算机视觉研究的重要领域,在视频监控和智能交通等领域应用广泛.本文提出了一种自适应的运动目标检测方法.该方法采用二维主成分分析建立背景模型,通过比较重建图像和原图像的差异来检测运动目标.为了自适应动态变化的复杂背景,该模型由增量算法在线更新.实验结果表明,提出的方法可以在复杂变化的动态背景环境下进行有效的运动目标检测.

Moving Object Detection Based on Two-dimensional PCA

Moving object detection is a basic component of many computer vision applications.It has a critical impact on the performance of visual surveillance and intelligent transportation.This paper proposes an adaptive moving object detection approach that can deal with dynamic scenes.This approach uses 2D-PCA to describe the background model and on-line updates it using incremental learning method.Experimental results demonstrate that the proposed method is able to detect moving objects under various types of dyn...