1:00, SAM-L3.1
DETECTION AND IDENTIFICATION OF ODORANTS USING AN ELECTRONIC NOSE
R. POLIKAR, R. SHINAR, V. HONAVAR, L. UDPA, M. PORTER
Gas sensing systems for detection and identification of odorant molecules are of crucial importance in an increasing number of applications. Such applications include environmental monitoring, food quality assessment, airport security, and detection of hazardous gases. In this paper, we describe a gas sensing system for detecting and identifying volatile organic compounds (VOCs), and discuss the unique problems associated with the separability of signal patterns obtained by using such a system. We then present solutions for enhancing the separability of VOC patterns to enable classification. A new incremental learning algorithm that allows new odorants to be learned is also introduced.

1:20, SAM-L3.2
A NEW BEARING FAULT DETECTION AND DIAGNOSIS SCHEME BASED ON HIDDEN MARKOV MODELING OF VIBRATION SIGNALS
H. OCAK, K. LOPARO
This paper introduces a new bearing fault detection and diagnosis scheme based on hidden Markov modeling (HMM) of vibration signals. First features are extracted from amplitude demodulated vibration signals obtained from both normal and faulty bearings. The features are based on the reflection coefficients of the polynomial transfer function of the auto-regressive model of the vibration signal. These features are then used to train HMMs to represent various bearing conditions. The technique allows for online detection of faults by monitoring the probabilities of the pre-trained HMM for the normal case. It also allows for the diagnosis of the fault by the HMM that gives the highest probability. The new scheme was tested with experimental data collected from drive end ball bearing of an induction motor (Reliance Electric 2HP IQPreAlert) driven mechanical system.

1:40, SAM-L3.3
FAST KNOCK DETECTION USING PATTERN SIGNALS
S. CARSTENS-BEHRENS, J. BOEHME
In order to detect knock in spark ignition engines, usually structure-borne sound signals measured by acceleration sensors mounted on the engine housing are analyzed. Earlier investigations have shown that using linear, time variant filtered structure-borne sound signals as approximated pressure signal instead improves knock detection significantly. But this method is computationally too expensive for application in production vehicles. In this paper we propose to fit suitable pattern signals to structure-borne sound and use the estimated scaling parameters to approximate pressure. The new approach is applied to knock detection with measured data.

2:00, SAM-L3.4
BEST BANDS SELECTION FOR DETECTION IN HYPERSPECTRAL PROCESSING
N. KESHAVA
In this paper, we explore the role of best bands algorithms in the context of maximizing the performance of hyperspectral algorithms. Specifically, we first focus on creating an intuitive framework for how metrics quantify the distance between two spectra. Focusing on the Spectral Angle Mapper (SAM) metric, we demonstrate how the separability of two spectra can be increased by choosing the bands that maximize the metric. This intuition about best bands analysis for SAM is extended to the Generalized Likelihood Ratio Test (GLRT) for a practical target/background detection scenario. Results are shown for a scene imaged by the HYDICE sensor demonstrating that the separability of targets and background can be increased by carefully choosing the bands for the test.

2:20, SAM-L3.5
ADAPTIVE MATCHED SUBSPACE DETECTORS FOR HYPERSPECTRAL IMAGING APPLICATIONS
D. MANOLAKIS, C. SIRACUSA, G. SHAW
Real-time detection and identification of man-made objects or materials (``targets'') from airborne platforms using hyperspectral sensors are of great interest for civilian and military applications. Over the past several years, different algorithms for the detection of targets with known spectral signature have been developed. Most of these algorithms have been reviewed by Manolakis et al within a unified theoretical and notational framework. In this paper we study adaptive matched subspace detection algorithms for low probability, single-pixel or subpixel targets. These algorithms explore the linear mixing model to both specify the desired target and characterize the interfering background. The derived algorithms are theoretically and experimentally evaluated with regard to two desirable properties: capacity to operate in constant false alarm rate (CFAR) mode and target ``visibility'' enhancement. Furthermore, an approach for taking into account target variability, when present, to improve detection is presented.

2:40, SAM-L3.6
INFRARED LAND MINE DETECTION BY PARAMETRIC MODELING
M. LUNDBERG
A parametric model for the infrared signature caused by a buried land mine is presented. Further, a detector is calculated for the case where the background noise can be described by an autoregressive process. The detector separately estimates the parameters of the mine and the noise in an alternating fashion. The estimates are then used in the likelihood ratio test. Simulations show that significant gains in performance can be achieved as compared to the standard detector used, which correlates the infrared image with the known mine shape and thresholds the square of the output.