计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (8): 154-156.

• 图形、图像、模式识别 • 上一篇    下一篇

空间目标融合识别算法及实验研究

肖旭光1,2,肖 刚1,敬忠良1   

  1. 1.上海交通大学 航空航天学院,上海 200240
    2.上海交通大学 电子信息与电气工程学院,上海 200240
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-03-11 发布日期:2011-03-11

Study on space target recognition algorithm and its experiment

XIAO Xuguang1,2,XIAO Gang1,JING Zhongliang1   

  1. 1.School of Aerospace Science and Technology,Shanghai Jiaotong University,Shanghai 200240,China
    2.School of Electronic Information and Electrical Engineering,Shanghai Jiaotong University,Shanghai 200240,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-03-11 Published:2011-03-11

摘要: 提出了基于红外与可见光图像的空间目标融合识别算法。算法针对空间目标的物理特征,利用红外和可见光图像提供的互补信息进行空间目标融合识别。方法以空间卫星为识别目标,采用不变矩和仿射不变矩来描述目标特征,提出了基于红外和可见光的特征级和决策级融合识别方案,并进行了半物理仿真实验和分析。实验证明这两种基于红外和可见光的融合识别算法可明显改善目标识别的精度和可靠性。

关键词: 融合识别, 红外图像, 可见光图像, 不变矩, 仿射不变矩

Abstract: Two approaches based on fusion of infrared and visible images for target recognition are presented.In most instances IR and visible images can provide more complementary information for feature extraction.Two kinds of images are used for satellites recognition.Invariant moments and affine invariant moments are employed to describe the characteristics of targets,and different fusion rules are adopted to improve the performance of recognition.The experiments with semi-physical simulation images indicate that the recognition based on fusion of IR and visible images consistently demonstrates better performance than others solely based on either IR or visible image.

Key words: fusion recognition, infrared image, visible image, invariant moments, affine invariant moments