In this paper, we propose a new object modeling method for infrared (IR) image. It is based on the modeling method using Gaussian Mixture Model (GMM) that has been originally proposed for general visible image. The original method is one of effective object modeling algorithms that can describe the topological structures of the internal patterns of object. This approach can also eliminate the influences due to small differences between patterns. On the other hand, an IR image is often used instead of visible image in actual applications such as outdoor surveillance. IR images make it easy to extract foreground object regions from background scenes, but their low-contrast makes object modeling difficult. We therefore propose a modeling method using Orientation-Code for IR image. Orientation-Code of each pixel has information about the maximum-gradient orientation of image, not intensity information. Gradient orientation information does not depend on contrast and describes internal pattern structures of objects even in unclear IR images. We also applied proposed method to vehicle detection for outdoor scenes, where it extracts multiple foreground regions as vehicle candidates using background subtraction for IR image, and they are described as models by our method. Models are finally compared with standard vehicle view models pre-memorized to determine which candidate is true vehicle or not. Evaluation tests with actual IR video sequences have proved that our proposed algorithm detects objects robustly.

1. [1] C. Stauffer and W. E. L. Grimson, “Adaptive background mixture models for real-time tracking,” CVPR99, pp. 246-252, 1999.
2. [2] T. Matsuyama, T. Ohya, and H. Habe, “Background Subtraction for Non-Stationary Scenes,” Proc. of ACCV, pp. 662-667, 2000.
3. [3] M. Seki, T. Wada, H. Fujiwara, and K. Sumi, “Background subtraction based on cooccurrence of image variations,” Proc. of CVPR, Vol.II, pp. 65-72, 2003.
4. [4] S. Der, A. Chan, N. Nasrabadi, and H. Kwon, “Automated Vehicle Detection in Forward-Looking Infrared Imagery,” Applied Optics, Vol.43, Issue 2, pp. 333-348, 2004.
5. [5] M. Kagesawa, S. Ueno, K, Ikeuchi, and H. Kashiwagi, “Recognizing vehicles in infrared images using IMAP parallel visionboard,” IEEE Transactions on Intelligent Transportation Systems, Vol.2, No.1, pp. 10-17, 2001.
6. [6] S. Eminoglu, D. S. Tezcan, M. Y. Tanrikulu, and T. Akin, “Lowcost uncooled infrared detectors in CMOS process,” Sensors and Actuators A, Vol.109/1-2, pp. 102-113, Dec., 2003.
7. [7] M. Seki, K. Sumi, H. Taniguchi, and M. Hashimoto, “Gaussian Mixture Model for Object Recognition,”MIRU2004, pp. I-344-349, 2004 (in Japanese).
8. [8] A. Dempster, N. Laird, and D. Rubin, “Maximum likelihood from incomplete data via the EM algorithm,” Journal of Royal Statistical Society, Vol.39, pp. 1-38, 1977.
9. [9] F. Ullah, S. Kaneko, and S. Igarashi, “Object Search Using Orientation Histogram Intersection,” Proc. of FCV2000, pp. 110-115, 2000.
10. [10] H. Saito and M. Kimura, “Superquadrics Modeling of Multiple Objects from Shading Images Using Genetic Algorithms,” 22nd International Conference on Industrial Electronics, Control, and Instrumentation, pp. 1589-1593, Taipei, August, 1996.
11. [11] H. Makita, N. Murakami, H. Tanahashi, K. Kato, and K. Yamamoto, “Extraction of Superquadrics from Cylindrical Data using Genetic Algorithm,” Vision Interface, 1998, pp. 189-196, June, 1998.