Robust Sonar ATR Through Bayesian Pose-Corrected Sparse Classification

Abstract

Sonar imaging has seen vast improvements over the last few decades due in part to advances in synthetic aperture Sonar (SAS). Sophisticated classification techniques can now be used in Sonar automatic target recognition (ATR) to locate mines and other threatening objects. Among the most promising of these methods is sparse reconstruction-based classification (SRC) which has shown an impressive resiliency to noise, blur, and occlusion. We present a coherent strategy for expanding upon SRC for Sonar ATR that retains SRC's robustness while also being able to handle targets with diverse geometric arrangements, bothersome Rayleigh noise, and unavoidable background clutter. Our method, pose corrected sparsity (PCS), incorporates a novel interpretation of a spike and slab probability distribution towards use as a Bayesian prior for class-specific discrimination in combination with a dictionary learning scheme for localized patch extractions. Additionally, PCS offers the potential for anomaly detection in order to avoid false identifications of tested objects from outside the training set with no additional training required. Compelling results are shown using a database provided by the United States Naval Surface Warfare Center. Index Terms-Sonar ATR, sparse reconstruction-based classification, anomaly detection. J. McKay and V. Monga are with the

Multi-Channel Synthetic Aperture Radar Based Classification of Maritime Scenes

“…We answer these questions by demonstrating, for the first time, an efficient and flexible classification algorithm that utilizes the in-scene MSAR-derived motion information. This method complements existing approaches to classification [11][12][13][14][15][16][17][18][19] and image segmentation [20][21][22][23][24][25][26][27] that exploit the spatial structure of static amplitude images. Our experimental results, performed on imagery captured by the U.S.…”

Target Classification in Synthetic Aperture Radar Images Using Quantized Wavelet Scattering Networks

“…In particular, image classification is the process of organizing images into different classes based on the output of feature extraction operators applied to images. There are innumerable approaches to feature extraction, a necessary precursor to classification, including decision-theoretic approaches using quantitative descriptors such as length, area, and texture [ 1 , 2 ]; structural approaches using qualitative descriptors, such as relational descriptors [ 3 ]; projection of data into fixed basis sets, such as wavelets [ 4 ] and Zernike polynomial moments [ 5 ], or adaptive basis sets [ 6 ]. Other examples include robust edges and corners that are popular in computer vision, blind synthesis of template classes by using singular value decomposition, Karhunen–Loeve Transform [ 7 , 8 ] and estimation theoretic templates [ 9 ], motion-based covariance matrix-based features for multi-sensor architectures [ 10 ], and finally micro-Doppler- [ 11 ] and vibrometry-based [ 12 ] features that have applications in radar-based sensing systems.…”

BP algorithm for the multireceiver SAS

“…The high resolution image is very useful for many applications, e.g. to find objects such as mines [4, 5] and wrecks [6] and to produce maps of the seafloor [7]. Synthetic aperture image formation [3] plays an important role in the multireceiver SAS system.…”