Classifier-based offline feature selection and evaluation for visual tracking of sea-surface and aerial targets

Abstract

Downloaded From: http://opticalengineering.spiedigitallibrary.org/ on 05/15/2015 Terms of Use: http://spiedl.org/termsAbstract. An offline feature selection and evaluation mechanism is used in order to develop a robust visual tracking scheme for sea-surface and aerial targets. The covariance descriptors, known to constitute an efficient signature set in object detection and classification problems, are used in the feature extraction phase of the proposed scheme. The performance of feature sets are compared using support vector machines, and those resulting in the highest detection performance are used in the covariance based tracker. The tracking performance is evaluated in different scenarios using different performance measures with respect to ground truth target positions. The proposed tracking scheme is observed to track sea-surface and aerial targets with plausible accuracies, and the results show that gradient-based features, together with the pixel locations and intensity values, provide robust target tracking in both surveillance scenarios. The performance of the proposed tracking strategy is also compared with some well-known trackers including correlation, KanadeLucas-Tomasi feature, and scale invariant feature transform-based trackers. Experimental results and observations show that the proposed target tracking scheme outperforms other trackers in both air and sea surveillance scenarios. C 2011 Society of Photo-Optical Instrumentation Engineers (SPIE).

Visual object tracking using Fourier domain phase information

“…2. In order to quantify the tracking performance, objective measures described in [17][18][19] are used. These objective measures compare the bounding boxes around the target region determined by the tracker and ground-truth information.…”

“…As a strong feature descriptor for target representation, scale invariant feature transform (SIFT) [12] has also been used in target tracking and target classification [13,14]. Covariance feature descriptor [15] which is computationally more efficient than SIFT-based approaches, has been also utilized for target tracking applications [16][17][18]. Although feature descriptors and complex tracking frameworks [19] have been very successful in tracking applications, the computational complexity of the solution is important factor for an efficient implementation.…”