Joint Target Tracking, Recognition and Segmentation for Infrared Imagery Using a Shape Manifold-Based Level Set

Abstract

We propose a new integrated target tracking, recognition and segmentation algorithm, called ATR-Seg, for infrared imagery. ATR-Seg is formulated in a probabilistic shape-aware level set framework that incorporates a joint view-identity manifold (JVIM) for target shape modeling. As a shape generative model, JVIM features a unified manifold structure in the latent space that is embedded with one view-independent identity manifold and infinite identity-dependent view manifolds. In the ATR-Seg algorithm, the ATR problem formulated as a sequential level-set optimization process over the latent space of JVIM, so that tracking and recognition can be jointly optimized via implicit shape matching where target segmentation is achieved as a by-product without any pre-processing or feature extraction. Experimental results on the recently released SENSIAC ATR database demonstrate the advantages and effectiveness of ATR-Seg over two recent ATR algorithms that involve explicit shape matching.

Tracker-aided adaptive multi-frame recognition of a specific target

“…Using the formula in Eq. (2), and choosing T such that the probability of reporting a false target is zero in both cases, we observed that the conventional approach without adaptation yields PCT =79.5%. By comparison, the multi-frame adaptive algorithm dramatically improves the result to PCT =88%.…”

“…However this is generally an "open loop" process with memory-less target recognition algorithms that do not directly incorporate spatio-temporal information as part of the pattern matching process. The idea of joint tracking and multi-class target recognition has been recently reported in the literature [2]. We propose a "closed loop" approach (such as in Figure 1) where the ATR algorithm not only limits its attention to location of the tracked object, but also adaptively reduces any uncertainty (or errors) in the tracked location over time to improve decision confidence and the target aimpoint.…”

Two Algorithms for the Detection and Tracking of Moving Vehicle Targets in Aerial Infrared Image Sequences

“…In the field of infrared target tracking, good results have been obtained in previous research using region-based [15,16], contour-based [17,18], model-based [19,20] and feature-based [21,22] algorithms. For example, Ling et al [23] defined the evaluation criterion for the tracking effect and searched for the relatively accurate region similar to the reference region by maximizing the eigenvalues of the covariance matrix of the local complexity when the tracking error was large.…”

Kernel sparse coding method for automatic target recognition in infrared imagery using covariance descriptor

“…[15] proposed a joint view-identity manifold (JVIM) for multi-view and multi-target shape modeling that is well suited for ATR in infrared imagery, and Ref. [16] proposed a integrated target tracking, recognition and segmentation algorithm called ATR-Seg for infrared imagery, which is formulated in a probabilistic shape-aware level set framework incorporating a JVIM for target shape modeling.…”