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.
Highlights
As a challenging problem in pattern recognition and machine learning for decades, automatic target tracking and recognition (ATR) has been an important topic for many military and civilian applications.Infrared (IR) ATR is a more challenging problem due to two main reasons
M, to represent the foreground/background models M = {Mf, Mb }; the original graphical model of ATR-Seg in Figure 1 will become the one in Figure 4c. which defines a joint distribution of all parameters for each pixel (xi, yi ) as p(xi, yi, p, Θ, Φ, M ) = p(xi |p, Φ, M )p(yi |M )p(Φ|Θ)p(M )p(Θ)p(p) where Φ is a shape represented by the level set embedding function shown in Figure 4b and p(Φ|Θ) corresponds to joint view-identity manifold (JVIM)-based shape interpolation via Gaussian Process (GP) mapping
We present the results of the particle filter-based infrared ATR algorithm, where four shape models (JVIM, couplet of view and identity manifolds (CVIM), LL-Gaussian Process Latent Variable Models (GPLVM), nearest neighbor (NN)) are compared in the case of explicit shape matching
Summary
As a challenging problem in pattern recognition and machine learning for decades, automatic target tracking and recognition (ATR) has been an important topic for many military and civilian applications. Infrared (IR) ATR is a more challenging problem due to two main reasons. An IR target’s appearance may change dramatically under different working conditions and ambient environment. There are two important and related research issues in ATR research, appearance representation and motion modeling [1]. The former one focuses on capturing distinct and salient features (e.g., edge, shape, texture) of a target, and the latter one tries to predict the target’s state (e.g., position, pose, velocity) during sequential estimation
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
