Sengvieng A. Amphay

Superresolution of millimeter-wave images by iterative blind maximum likelihood restoration

The need for superresolution processing of images in multispectral seeker environments for facilitating smart munition guidance is being increasingly recognized, particularly when the sensor suite includes Millimeter-Wave (MMW) sensors with rather poor inherent resolution capabilities. Despite the technological breakthroughs being made in advanced radiometer designs, the inherent problems associated with diffraction limited imaging impose limitations on the resolution of acquired imagery thus necessitating efficient post-processing to achieve resolution improvements needed for reliable target detection, classification and aimpoint selection. Quantitative results from a recent project directed to superresolution processing of passive MMW images obtained from a 95 GHZ 1-foot diameter aperture radiometer are presented in this paper. The spectral extrapolation performance resulting from the implementation of an iterative Maximum Likelihood restoration algorithm is demonstrated and the robustness of the algorithm that facilities a blind implementation useful in scenarios characterized by an incomplete knowledge of sensor point spread function is highlighted.

ATA algorithm suite for co-boresighted pmmw and ladar imagery

The need for air-to-ground missiles with day/night, adverse weather and pinpoint accuracy Autonomous Target Acquisition (ATA) seekers is essential for todays modern warfare scenarios. Passive millimeter wave (PMMW) sensors have the ability to see through clouds; in fact they tend to show metallic objects in high contrast regardless of weather conditions. However, their resolution is very low when compared with other ATA sensor such as laser radar (LADAR). We present an ATA algorithm suite that combines the superior target detection potential of PMMW with the high-quality segmentation and recognition abilities of LADAR. Preliminary detection and segmentation results are presented for a set of image-pairs of military vehicles that were collected for this project using an 89 Ghz, 18 inch aperture PMMW sensor from TRW and a 1.06 (mu) high-resolution LADAR.

Optimized maximum-likelihood algorithms for superresolution of passive millimeter-wave imagery

Iterative image restoration algorithms developed using a maximum likelihood (ML) estimation framework are attaining considerable significance in recent times for super-resolution processing of passive millimeter wave (PMMW) images. In this paper we offer a processor requirements analysis for implementing these algorithms, which provides assurance on the feasibility of their implementation using commercially available microprocessors, even for applications where processing time may be of critical importance. Two optimized versions of these algorithms, one developed by augmenting each iterative estimation step with a post-filtering operation and the other developed by incorporating a background-detail separation approach in the estimation process, are developed which provide superior resolution enhancement performance while simultaneously suppressing noise-induced and ringing artifacts in the restored images. Results of processing data acquired from a 95 GHz 1 foot diameter aperture radiometer are included to demonstrate that these algorithms offer significant superresolution capabilities for processing PMMW imagery.

Feature based Target classification in laser radar

Numerous feature detectors have been defined for detecting military vehicles in natural scenes. These features can be computed for a given image chip containing a known target and used to train a classifier. This classifier can then be used to assign a label to an un-labeled image chip. The performance of the classifier is dependent on the quality of the set of features used. In this paper, we first describe a set of features commonly used by the Automatic Target Recognition (ATR) community. We then analyze feature performance on a vehicle identification task in laser radar (LADAR) imagery. Our features are computed over both the range and reflectance channels. In addition, we perform feature subset selection using two different methods and compare the results. The goal of this analysis is to determine which subset of features to choose in order to optimize performance in LADAR Autonomous Target Acquisition (ATA).

Sensor fusion with passive millimeter-wave and laser radar for target detection

Advanced sensors and guidance techniques are required in killing mobile offensive and defensive systems. Many different sensors such as radar, video camera, laser radar (LADAR), millimeter wave systems, infrared imagers, acoustic sensors, etc. are available for such usage. However, no single sensor provides completely satisfactory capabilities. Since some sensors have complementary capabilities, integration of multiple sensors for kill can relax the task difficulty and provide more reliable results. The use of multiple sensors can also reduce the possibility of being defeated by countermeasures. In this study, we investigated the framework and investigated potential techniques for integration and fusion of information from passive millimeter wave (PMMW) and LADAR systems. The focus has been on target detection. The PMMW is used to detect metal objects and the LADAR examines those regions of interest for other evidence of existence of a target. Advances obtained by integrating these two sensors include reduction of task complexity and improvement of reliability, both due to efficient localization of regions of interest from the PMMW. Since PMMW possesses weather penetration capabilities through fog, cloud, smoke, etc., the combined system has a near-all-weather capability. A LADAR provides 3D information, and it should be used as the primary sensor for target acquisition upon target detection. The framework of the fusion is based on the Dempster-Shafer decision method. The fusion may be done in the algorithm level and sensor level. With the Dempster-Shafer method as the framework, new sensors or new decision components can be easily integrated.

Image restoration in multisensor missile seeker environments for design of intelligent integrated processing architectures

Two major factors that could limit successful implementations of image restoration and superresolution algorithms in missile seeker applications are, (i) lack of accurate knowledge of sensor point spread function (PSF) parameters, and (ii) noise-induced artifacts in the restoration process. The robustness properties of a recently developed blind iterative Maximum Likelihood (ML) restoration algorithm to inaccuracies in sensor PSF are established in this paper. Two modifications to this algorithm that successfully equip it to suppress artifacts resulting from the presence of high frequency noise components are outlined. Performance evaluation studies with 1D and 2D signals are included to demonstrate that these algorithms have superresolution capabilities while possessing also attractive robustness and artifact suppression properties. The algorithms developed here hence contribute to efficient designs of intelligent integrated processing architectures for smart weapon applications.

Multisensor segmentation using LADAR and PMMW

Fusing information from sensors with very different phenomenology is an attractive and challenging option for autonomous target acquisition (ATA) systems because correct target detections should correlate between sensors while false alarms might not. In this paper, we present a series of algorithms for detecting and segmenting targets from their background in passive millimeter wave (PMMW) and laser radar (LADAR) data. PMMW sensors provide a consistent signature for metallic targets. They also can effectively operate under adverse weather conditions, however they exhibit poor angular resolution. LADAR sensors produce high-resolution range and reflectance images, but are sensitive to adverse weather conditions. Sensor fusion techniques are applied with the goal of maintaining high probability of detection while decreasing the false alarm rate.

Superresolved imaging sensors with field-of-view preservation

To recover spatial information from bandlimited images using maximum likelihood (ML) and constrained least squares techniques it is necessary that the image plane be oversampled. Specifically, oversampling allows the blur component induced by spatial integration of the signal over the finite size of the detector element(s) to be reduced. However, if oversampling in the image plane is achieved with a fixed array, the field of view (FOV) is proportionately reduced. Conversely, if the FOV is to be preserved then proportionately more samples are required implying the requirement for additional detector elements. An effective solution to obtaining oversampling in the image plane and subsequently preserving the FOV, is to use either controlled or uncontrolled microscanning. There are a number of methods to achieve microscanning including translation of the sensor array in the image plane and exploitation of airframe jitter. Three unique sixteen-times-Nyquist oversampled passive millimeter wave (PMMW) images; a point source, an extended source, and an M48 tank were carefully obtained. Both ML and constrained least squares (CLS) algorithms were used for restoration of spatial information in the images. Restoration of known extended source object functions (contained in the extended source image) resulted in resolution gains of 1.47 and 3.43 using the CLS and ML methods respectively, as measured by increase in effective aperture.

Estimating the ground plane in ladar three-dimensional imagery for target detection

A common approach to detecting targets in laser radar (LADAR) 3-dimensional x, y and z imagery is to first estimate the ground plane. Once the ground plane is identified, the regions of interest (ROI) are segmented based on height above that plane. The ROIs can then be classifed based on their shape statistics (length, width, height, moments, etc.) In this paper, we present an empirical comparison of three different ground plane estimators. The first estimates the ground plane based on global constraints (a least median squares fit to the entire image). The second two are based on progressively more local constraints: a least median squares fit to each row and column the image, and a local histogram analysis of the re-projected range data. These algorithms are embedded in a larger system that first computes the target height above the ground plane and then recognizes the targets based on properties within the target region. The evaluation is performed using 98 LADAR images containing eight different targets and structured clutter (trees). Performance is measured in terms of percentage of correct detection and false alarm.

General-purpose performance evaluation tool for analysis and comparison of ATA algorithms

To support Autonomous Target Acquisition (ATA) evaluation and trades analysis, the Air Force Research Laboratory, Advanced Guidance Division (AFRL/MNG) located at Eglin AFB has incorporated a general-purpose performance evaluation system into its Modular Algorithm Concept Evaluation Tool (MACET). The MACET performance evaluation system may be used for active, passive, or multi-sensor ATA analysis. It consists of two main elements: a relational, multi-user database engine and a database client application, the Performance Evaluation Tool (PET). The database engine serves a set of databases that are used to capture, catalog, and archive test results for various algorithms under varying condition and environments. The MACET PET client application is a data mining tool for exploring the ATA test results in the databases, computing standard ATA detection and classification performance metrics (e.g., detection probability, detection reliability, false alarm rate, probability of correct classification, confusion matrices) on user defined subsets of data, calculating test case parameter statistics, and generating performance comparison plots.