Publications

• Non-Cooperative Identification of Ships with Electrooptical Data
• Digital Signal Processing Applications in Cochlear Implant Research
• Discrimination Performance Requirements for Ballistic Missile Defense
• An Automatic Face Recognition System Using the Adaptive Clustering Network

Non-Cooperative Identification of Ships with Electrooptical Data
Jerrold Baum, Eleanor Tung, and Steven Rak

A multidimensional sensor suite consisting of a laser radar and a passive-infrared (IR) sensor has been proposed and evaluated for detecting and identifying ships at long ranges from an airborne platform. The passive-IR sensor would detect targets by taking advantage of the high target-to-background contrast and the sensor's ability to track over a wide field of view. The higher information content of the ship's range profile would then be exploited for target identification.

A two-part approach was developed to evaluate the proposed system. The first part concentrated on creating synthetic signatures of naval vessels under a variety of controlled sensor operating characteristics and target scenarios. Two neural networks were used to classify the synthetic signatures. In the second part of the evaluation, active and passive-IR measurements of two naval vessels were taken with an existing airborne multidimensional sensor system. These missions demonstrated that measured range profiles were similar to the synthetic profiles. Our results, from both synthetic and measured data, indicate that range-profile and passive-IR signatures complement each other in covering all viewing aspects for long-range ship classification.

Digital Signal Processing Applications in Cochlear Implant Research
Joseph Tierney, Marc A. Zissman, and Donald K. Eddington

We have developed a facility that enables scientists to investigate a wide range of sound-processing schemes for human subjects with cochlear implants. This digital signal processing (DSP) facility—named the Programmable Interactive System for Cochlear Implant Electrode Stimulation (PISCES)—was designed, built, and tested at Lincoln Laboratory and then installed at the Cochlear Implant Research Laboratory (CIRL) of the Massachusetts Eye and Ear Infirmary (MEEI). New stimulator algorithms that we designed and ran on PISCES have resulted in speech-reception improvements for implant subjects relative to commercial implant stimulators. These improvements were obtained as a result of interactive algorithm adjustment in the clinic, thus demonstrating the importance of a flexible signal-processing facility. Research has continued in the development of a laboratory-based, software-controlled, real-time, speech-processing system; the exploration of new sound-processing algorithms for improved electrode stimulation; and the design of wearable stimulators that will allow subjects full-time use of stimulator algorithms developed and tested in a laboratory setting.

Discrimination Performance Requirements for Ballistic Missile Defense
Stephen D. Weiner and Sol M. Rocklin

A missile defense system must be able to deal with an attack containing decoys in addition to warheads. If the defense system does not have enough interceptors to shoot at all the incoming objects, it must be able to discriminate between decoys and warheads. This discrimination process is not perfect and results in two types of errors: leakage (not shooting at warheads) and false alarms (shooting at decoys). This article describes a methodology for analyzing the consequences of these discrimination errors and determining how well discrimination must perform in a variety of defense scenarios. The analysis focuses on game-theoretic solutions in which the defense can achieve its overall objective of surviving the attack regardless of the tactics used by the offense.

An Automatic Face Recognition System Using the Adaptive Clustering Network
Murali M. Menon and Eric R. Boudreaus

We have developed an Automatic Face Recognition (AFR) System that uses the Adaptive Clustering Network (ACN—-a hybrid classifier that combines neural network learning with statistical decision making. The ACN automatically groups similar faces into the same cluster and creates new clusters for novel input faces. During training, the ACN updates the clusters continuously, and multiple clusters are created for the same subject to accommodate variations in the presentation of the subject's face (for example, changes in facial expression and/or head orientation). With incremental training, new subjects and further variations of existing subjects may be added on line without retraining the classifier on previous data. During the testing process, the ACN associates an input face with the cluster that most closely matches the face. The ACN minimizes misidentifications by reporting completely novel input faces as "unknown."

In addition to the ACN classifier, the overall APR system includes a preprocessing stage that removes the background in an image and centers the face in the image frame, and a feature-extraction stage that compresses the data. The system requires relatively simple processing and has been implemented in software on a SUN workstation. The preliminary results have been encouraging. Using imagery of eight subjects taken with a video camera, the system achieved a correct-classification performance of 99% with no misidentifications.