Target recognition method based on template matching in downward-looking infrared image

Automatic ground target recognition technology in downward-looking infrared imagery is challenging problems due to the complexity of real-world. A robust ground target recognition method is proposed for downward-looking infrared (DLIR) image sequences in this paper. A complete model of the proposed method is designed firstly and the principles of some key technologies are introduced as following. Perspective transformation is used to solve the projective problems of landmark translation, rotation and scale variance in real-time image sequence. The size of landmark is estimated real-timely by using flight parameters and imaging parameters for getting the model with an appropriate scale. Based on the matching results of landmark and flight parameters, target position technology is proposed to identify the position of target by using the position relation between landmark and target. Experimental results using real-world image data with complicated background showed that the proposed method not only causes high precisely locating results, but also has good robustness for target occlusion.

[1]  Chen Wang,et al.  A Kernel-Based Nonparametric Regression Method for Clutter Removal in Infrared Small-Target Detection Applications , 2010, IEEE Geoscience and Remote Sensing Letters.

[2]  Soo-Chang Pei,et al.  A morphological approach of target detection on perspective plane , 2001, Signal Process..

[3]  Clark F. Olson,et al.  Automatic target recognition by matching oriented edge pixels , 1997, IEEE Trans. Image Process..

[4]  B. D. Guenther,et al.  Aided and automatic target recognition based upon sensory inputs from image forming systems , 1997 .

[5]  Khan M Iftekharuddin,et al.  Detection and tracking of rotated and scaled targets by use of Hilbert-wavelet transform. , 2003, Applied optics.

[6]  Francesca Bovolo,et al.  Semisupervised One-Class Support Vector Machines for Classification of Remote Sensing Data , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[7]  Mark E. Oxley,et al.  Neural networks for automatic target recognition , 1995, Neural Networks.

[8]  Min-Seok Choi,et al.  A novel two stage template matching method for rotation and illumination invariance , 2002, Pattern Recognit..

[9]  Zhang Peng,et al.  The design of Top-Hat morphological filter and application to infrared target detection , 2006 .

[10]  Qun Zhao,et al.  Support vector machines for SAR automatic target recognition , 2001 .

[11]  Bir Bhanu,et al.  A system for model-based object recognition in perspective aerial images , 1998, Pattern Recognit..

[12]  Yi-Hsien Lin,et al.  Template matching using the parametric template vector with translation, rotation and scale invariance , 2008, Pattern Recognit..

[13]  Christopher E. Hann,et al.  Fast normalized cross correlation for motion tracking using basis functions , 2006, Comput. Methods Programs Biomed..

[14]  Daniel P. Huttenlocher,et al.  Comparing Images Using the Hausdorff Distance , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Tianxu Zhang,et al.  Robust and fast Hausdorff distance for image matching , 2012 .