Analysis of matching algorithm for infrared images based on anisotropic diffusion equation

Ruicheng Yan, Zhiguo Cao, Yukun LiProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2009 · DOI

Abstract

We present an algorithm for infrared target recognition based on edge features, aiming to improve the performance of recognition within low contrast and low SNR infrared images. The main process including: using nonlinear anisotropic diffusion equation to smooth noise as preprocessing after enhancing the image contrast through equalizing the image gray-level histogram, so as to filter out noise as well as maintain edge and shape features; then considering the greater differences between the target template and the fragmentary edges extracted directly from infrared images, Otsu threshold segmentation algorithm is selected to extracted the region feature before we use Canny operator to extract target edge features; finally, we match the target edge template with infrared images to recognize the target, using Hausdorff distance as a matching measure. Through experiment on several infrared image sequences, the results show that comparing with other preprocessing algorithms, our method achieve better performance for target recognition.
BibTeX
@inproceedings{yan2009analysis,
  author = {Ruicheng Yan and Zhiguo Cao and Yukun Li},
  title = {{Analysis of matching algorithm for infrared images based on anisotropic diffusion equation}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2009},
  doi = {10.1117/12.831086},
}

Loading related content...

Figure