Traditional image contour tracking algorithm has low tracking accuracy. To solve this problem, an object contour tracking algorithm based on local significant edge features in complex background is proposed. In the algorithm, the projective invariant is firstly introduced, to construct the geometric information descriptor between the edge positions of the infrared image, and set up the histogram of the feature number of each target contour. The geometric similarity between the features is measured by the pasteurized coefficient, the edge features of the neighbourhood around the object contour are established, and the object contour with significant features in the edge of the image is searched. Combining Shape-context operator with edge feature, the feature description vector can be formed, and Euclidean distance is defined to track measurement function. Using this function, the selected object contour is tracked preliminarily. The random consistency checking algorithm is used to eliminate the false tracking feature points and obtain the best tracking value of the object contour, thus the infrared image’s object contour tracking is carried out in the complex background. Experimental simulation shows that the proposed algorithm has high tracking accuracy and effectively improves the quality of infrared image analysis.

1. [1] A. Mondal, S. Ghosh, and A. Ghosh, “Efficient silhouette-based contour tracking using local information,” Soft Computing, Vol.20, No.2, pp. 785-805, 2016.
2. [2] P. Lv, Q. Zhao, Y. Chen, Y. et al., “Multiple cues-based active contours for object contour tracking under sophisticated background,” Visual Computer, Vol.33, No.9, pp. 1-17, 2016.
3. [3] L. Fang, W. Zhao, X. Li, et al., “A convex active contour model driven by local entropy energy with applications to infrared ship target segmentation,” Optics Laser Technology, Vol.96, pp. 166-175, 2017.
4. [4] L. Fang, X. Wang, and Y. Wan, “Adaptable active contour model with applications to infrared ship target segmentation,” J. of Electronic Imaging, Vol.25, No.4, pp. 041010, 2016.
5. [5] K. Qian, H. Zhou, S. Rong, et al., “Infrared dim-small target tracking via singular value decomposition and improved Kernelized correlation filter,” Infrared Physics & Technology, Vol.82, pp. 18-27, 2017.
6. [6] W. Li, F. Pan, Y. Xiao, et al., “Infrared Target Tracking Based on Color Fusion Image and Particle Filter,” IEEE 5th Int. Conf. on Instrumentation and Measurement, Computer, Communication and Control, pp. 1424-1428, 2016.
7. [7] G. Yang, Y. Zhang, J. Yang, et al., “Automated classification of brain images using wavelet-energy and biogeography-based optimization,” Multimedia Tools & Applications, Vol.75, No.23, pp. 15601-15617, 2016.
8. [8] A. Liaghat and M. A. Masnadi-Shirazi, “Aerial target detection and tracking in infrared image sequences by using morphological operations Kalman filtering and elliptical representation,” 24th Iranian Conf. on Electrical Engineering, pp. 1841-1847, 2016.
9. [9] C. Lv, T. Zhang, and C. Liu, “An Improved Otsu’s Thresholding Algorithm on Gesture Segmentation,” J. Adv. Comput. Intell. Intell. Inform., Vol.21, No.2, pp. 247-250, 2017.
10. [10] S. Zheng, D. Fu, T. Yang et al., “A Novel Method for Eye Contour Extraction from Blurred Infrared Images,” Int. Conf. on Intelligent Human-Machine Systems and Cybernetics, pp. 103-106, 2016.
11. [11] H. Li, D. Fu, and T. Yang, “Hand target extraction from infrared images with descriptor based on pixel temporal characteristics,” Int. Congress on Image and Signal Processing, pp. 458-463, 2016.