Improved Multiple Target Tracking via Global Motion Compensation and Optoelectronic Correlation

Camera motion estimation in image sequences generally focuses on the recovery of the frames when the camera is mounted on a moving platform. Global motion in video sequences is more complex and involves camera operation, camera motion, and other nontarget motions. Global motion compensation is usually handled by compensating the dominant motion using estimation and segmentation techniques. To enhance tracker performance and accuracy, frame recovery operation plays a crucial role by estimating undesired motion. In this paper, a normalized correlation-based regional template-matching (TM) algorithm has been developed to accurately recover frames before the application of the tracking algorithm. Then, a robust multiple-target-tracking system has been developed using a combination of fringe-adjusted joint transform correlator and TM techniques. Joint transform correlation detects a target optoelectronically, while TM technique is performed digitally. To increase the tracking system speed and decrease the effects of clutter, a subframe segmentation and local deviation-based image-preprocessing algorithm has been proposed. The improved performance of multiple-target-tracking system is tested using real-life forward-looking infrared (IR) imagery video sequences obtained from IR sensors mounted on an airborne platform

[1]  Mohammad S. Alam,et al.  Phase-encoded fringe-adjusted joint transform correlation , 2000 .

[2]  D A Gregory,et al.  Effects of fringe binarization of multiobject joint transform correlation. , 1989, Applied optics.

[3]  D A Gregory,et al.  Multitarget detection using spatial synthesis joint transform correlation. , 1993, Applied optics.

[4]  Jake K. Aggarwal,et al.  MODEEP: a motion-based object detection and pose estimation method for airborne FLIR sequences , 2000, Machine Vision and Applications.

[5]  Marcelo G. S. Bruno,et al.  Sequential importance sampling filtering for target tracking in image sequences , 2003, IEEE Signal Processing Letters.

[6]  Mohammad S. Alam,et al.  Invariant fringe-adjusted joint transform correlation-based target tracking in FLIR sequences , 2004, SPIE Defense + Commercial Sensing.

[7]  Eliezer Oron,et al.  Precision tracking with segmentation for imaging sensors , 1993 .

[8]  B Javidi,et al.  Multiple-object detection with a chirp-encoded joint transform correlator. , 1993, Applied optics.

[9]  B. N. Chatterji,et al.  Design of a Real-Time Tracking System for Fast-moving Objects , 1997 .

[10]  Abdullah Bal,et al.  Automatic target tracking in FLIR image sequences , 2004, SPIE Defense + Commercial Sensing.

[11]  Ashraf Elnagar,et al.  Motion detection using background constraints , 1995, Pattern Recognit..

[12]  Mark B. Sandler,et al.  Regularized CDWT optical flow applied to moving-target detection in IR imagery , 2000, Machine Vision and Applications.

[13]  I. Reed,et al.  A Detection Algorithm for Optical Targets in Clutter , 1987, IEEE Transactions on Aerospace and Electronic Systems.

[14]  Karl S Gudmundsson,et al.  Sub-imaging technique to improve phase only filter search capability. , 2003, Applied optics.

[15]  Mohammad S. Alam.  Deblurring using fringe-adjusted joint transform correlation , 1998 .

[16]  B. Javidi,et al.  Joint transform image correlation using a binary spatial light modulator at the Fourier plane. , 1988, Applied optics.

[17]  Mohammad S. Alam,et al.  Efficient multiple target recognition using a joint wavelet transform processor , 2000 .

[18]  Mohammad S. Alam,et al.  Infrared image registration and high-resolution reconstruction using multiple translationally shifted aliased video frames , 2000, IEEE Trans. Instrum. Meas..

[19]  Yong-Hoon Kim,et al.  An efficient multitarget tracking algorithm for car applications , 2003, IEEE Trans. Ind. Electron..

[20]  M S Alam,et al.  Heteroassociative multiple-target tracking by fringe-adjusted joint transform correlation. , 2004, Applied optics.

[21]  Paul C. Miller,et al.  Evaluation of an optical correlator automatic target recognition system for acquisition and tracking in densely cluttered natural scenes , 1999 .

[22]  Uwe D. Hanebeck,et al.  Template matching using fast normalized cross correlation , 2001, SPIE Defense + Commercial Sensing.

[23]  Fabrice Heitz,et al.  Unsupervised statistical detection of changing objects in camera-in-motion video , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[24]  Mubarak Shah,et al.  Target tracking in airborne forward looking infrared imagery , 2003, Image Vis. Comput..

[25]  Ying Wu,et al.  Nonstationary color tracking for vision-based human-computer interaction , 2002, IEEE Trans. Neural Networks.

[26]  Majid Mirmehdi,et al.  Detection and Tracking of Very Small Low Contrast Objects , 1998, BMVC.

[27]  Mohammad A. Karim,et al.  Improved correlation discrimination in a multi-object bipolar joint transform correlator , 1992 .

[28]  J. Aggarwal,et al.  Detecting moving objects in airborne forward looking infra-red sequences , 1999, Proceedings IEEE Workshop on Computer Vision Beyond the Visible Spectrum: Methods and Applications (CVBVS'99).

[29]  Ying Wu,et al.  For Vision-Based Human Computer Interaction , 2001 .

[30]  Mohammad A. Karim,et al.  Multiple target detection using a modified fringe-adjusted joint transform correlator , 1994 .