Efficient object tracking algorithm using modified colour-texture descriptor

In this paper, rotation invariant local binary pattern and local contrast measure have been joined together for describing the texture feature. In addition to this, colour features have also been included in the feature vector of the object tracking algorithm. Thus the texture can be characterised by two orthogonal properties like, the spatial structure local binary pattern and the strength of the pattern contrast. Covariance matrix of the feature vector has been taken as the object descriptor in the proposed algorithm. The performance of the proposed method has been compared by different performance measures, such as the detection rate, tracking speed and coverage test. The proposed method has also been tested for various challenging situations, such as occlusion, camera motion, non-rigidity, and changes in illumination. The experimental results show that the proposed method outperforms the existing methods in terms of performance indices.

[1]  Dorin Comaniciu,et al.  Kernel-Based Object Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Ohta based covariance technique for tracking object in video scene , 2012, 2012 IEEE Students' Conference on Electrical, Electronics and Computer Science.

[4]  Chunhua Shen,et al.  Enhanced Kernel-Based Tracking for Monochromatic and Thermographic Video , 2006, 2006 IEEE International Conference on Video and Signal Based Surveillance.

[5]  Dorin Comaniciu,et al.  Real-time tracking of non-rigid objects using mean shift , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[6]  T. Kanade,et al.  Color information for region segmentation , 1980 .

[7]  Dipti Patra,et al.  Object Tracking in Video Images Using Hybrid Segmentation Method and Pattern Matching , 2009, 2009 Annual IEEE India Conference.

[8]  Myron Flickner,et al.  Detection and tracking of shopping groups in stores , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[9]  Erkki Oja,et al.  Texture discrimination with multidimensional distributions of signed gray-level differences , 2001, Pattern Recognit..

[10]  W. Förstner,et al.  A Metric for Covariance Matrices , 2003 .

[11]  Fatih Murat Porikli,et al.  Integral histogram: a fast way to extract histograms in Cartesian spaces , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[12]  Fatih Murat Porikli,et al.  Region Covariance: A Fast Descriptor for Detection and Classification , 2006, ECCV.

[13]  Hélène Laurent,et al.  Review and evaluation of commonly-implemented background subtraction algorithms , 2008, 2008 19th International Conference on Pattern Recognition.

[14]  P. Meer,et al.  Covariance Tracking using Model Update Based on Means on Riemannian Manifolds , 2005 .

[15]  Calvin C. Gotlieb,et al.  Texture descriptors based on co-occurrence matrices , 1990, Comput. Vis. Graph. Image Process..

[16]  Matti Pietikäinen,et al.  Rotation-invariant texture classification using feature distributions , 2000, Pattern Recognit..

[17]  Matti Pietikäinen,et al.  Local Binary Pattern Descriptors for Dynamic Texture Recognition , 2006, 18th International Conference on Pattern Recognition (ICPR'06).