一种序列图像中运动点目标的检测方法

序列图像中运动点目标的检测按图像的成像系统不同，可分为红外图像中运动点目标的检测和可见光图像中运动点目标的检测，而现有检测算法多是针对前者。为寻找一种适用于两种类型图像的运动点目标的检测方法，以多云天空为研究背景，提出了一种新的运动点目标检测算法。采用高通滤波和形态学滤波相结合的方法进行背景抑制，基于检测前跟踪（TBD）的基本思想，根据相邻三帧进行目标分割，利用轨迹能量累积方法完成目标检测。理论分析和仿真结果表明，该算法简单易行．既适用于红外图像又适用于可见光图像的运动点目标，而且对目标的运动速度和方向无任何限制。

Method of moving point target detection in image sequences

According to different imaging system,moving point target detection in image sequences can be divided into two categories：moving point target detection in infrared image sequences and moving point target detection in visible image sequences. Most of existing arithmetics are aimed at infrared image sequences,so the main purpose of this article is to look for a new detection method which can be used in the two kinds of image sequences. A new arithmetic of moving point target detection for cloudy sky background is proposed.In the stage of background suppressing, a method that combines high-pass filter with mathematic morphology filter is applied. Based on the idea of TBD, three frames are conjointed to accomplish target segmentation. Track energy cumulation is used to detect the targets.Theoretic analysis and simulation results prove that this arithmetic is simple and easy to be realized, it can be used in infrared image and visible image ,furthermore it has no restriction on target＇s velocity and direction.