基于压缩感知的遥感图像小运动目标检测技术研究  

ON DETECTION TECHNIQUE OF SMALL MOVING OBJECTS IN REMOTE SENSING IMAGE BASED ON COMPRESSIVE SENSING

作　　者：修晓玉 刘玉喜[1] 周国辉[1] 

机构地区：[1]哈尔滨师范大学计算机科学与信息工程学院,黑龙江哈尔滨150025

出　　处：《计算机应用与软件》2014年第5期219-222,共4页Computer Applications and Software

基　　金：黑龙江省教育厅面上项目(12511149);智能教育与信息工程重点实验室基金项目(081203);机器视觉智能检测校企共建工程研发中心基金项目((2011)238)

摘　　要：天文遥感图像中小运动目标的检测是空中监测任务的一个重要问题。压缩感知理论的出现为在线天文遥感图像处理提供了一种简单高效的编码方法。提出一种基于压缩感知理论的天文遥感图像小运动目标的检测方法。这种算法通过比较当前图像测量值和背景图像测量值之间的不同来确定目标的测量值。与其他方法对整个场景成像不同,这种方法只对最显著的位置成像,有效地节约了计算资源。同时这种方法能够得到精确的定位结果。仿真结果表明,所提出的方法能够准确地检测小运动目标,同时实现精确定位。To detect small moving objects in astronomical remote sensing images is an important issue in space surveillance. The emergence of compressive sensing theory provides a simple and efficient coding approach for onboard astronomical remote sensing image processing. We propose a detection algorithm for small moving objects in astronomical remote sensing image which is based on compressive sensing theory. The algorithm determines the measurements of the objects by comparing the difference between the measurements of current image and the measurements of background scene. In contrast to other methods in imaging the whole scene,this algorithm only imaging the foreground position,which leads to effective computational performance. At the same time it can achieve accurate positioning results. Simulation experimental results show that the proposed algorithm can accurately detect small moving objects and realise precise localisation as well.

关 键 词：压缩感知 运动目标 遥感图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]