期刊名称：Signal & Image Processing : An International Journal (SIPIJ)
印刷版ISSN：2229-3922
电子版ISSN：0976-710X
出版年度：2011
卷号：2
期号：1
页码：205
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Image matching is a key component in almost any image analysis process. Image matching iscrucial to a wide range of applications, such as in navigation, guidance, automatic surveillance, robotvision, and in mapping sciences. Any automated system for three-dimensional point positioning mustinclude a potent procedure for image matching. Most biological vision systems have the talent to cope withchanging world. Computer vision systems have developed in the same way. For a computer vision system,the ability to cope with moving and changing objects, changing illumination, and changing viewpoints isessential to perform several tasks. Object detection is necessary for surveillance applications, for guidanceof autonomous vehicles, for efficient video compression, for smart tracking of moving objects, forautomatic target recognition (ATR) systems and for many other applications. Cross-correlation and relatedtechniques have dominated the field since the early fifties. Conventional template matching algorithmbased on cross-correlation requires complex calculation and large time for object detection, which makesdifficult to use them in real time applications. The shortcomings of this class of image matching methodshave caused a slow-down in the development of operational automated correlation systems. In theproposed work particle swarm optimization & its variants based algorithm is used for detection of object inimage. Implementation of this algorithm reduces the time required for object detection than conventionaltemplate matching algorithm. Algorithm can detect object in less number of iteration & hence less time &energy than the complexity of conventional template matching. This feature makes the method capable forreal time implementation. In this paper a description of particle Swarm optimization algorithm is given &then formulation of the algorithm for object detection using PSO & its variants is implemented forvalidating its effectiveness
