基于模板匹配的目标识别算法

来源期刊:中南大学学报(自然科学版)2011年第z1期

论文作者:由霖 修春波

文章页码:737 - 740

关键词:模板匹配;目标识别;微分;匹配准则

Key words:template matching; target recognition; differential; matching criterion

摘    要:为提高现有模板匹配算法的有效性,提出一种新的模板匹配方法。匹配模板选用圆形结构,利用圆形模板的中心对称性,可使算法对识别结果具有旋转不变性。匹配准则函数由2部分组成,一部分通过环形模板与相应跟踪区域图像的各子窗口的匹配值获得图像的灰度匹配信息,并以此减小噪声对匹配结果的影响;另一部分通过计算图像的微分值获得微分图像,通过计算各子窗口微分匹配值来增强匹配过程中图像的细节信息的作用,并以此抑制光照变化对识别结果的影响。利用该方法对复杂背景下的移动目标进行识别跟踪。仿真结果表明,该方法对噪声、光照变化等影响具有很好的抑制作用,能够提高识别的精度和准确性。在某电视跟踪系统中的应用结果表明,该方法能够满足跟踪系统实时性的要求。

Abstract: A new template matching method was proposed to improve the effectiveness. The matching template was of round structure. The center symmetric property of round template could make the algorithm have rotation invariant. The matching criterion function contained two components. One was the gray matching value between template sub-windows and image sub-windows, which could reduce the influence of noise. And the other was differential matching value of differential images, which represented texture information of images and could help to reduce the influence of illumination. Target recognition in complex background is realized by the method. Simulation results show that the method has a good recognition accuracy, and can restrain the impact of noise and illumination. Results of its application in the TV tracking system show that it can meet the demand of real time of the tracking system.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号