Etegent Technologies (United States)

Synthetic data is often leveraged for training and testing Automatic Target Recognition (ATR) systems on a variety of operating conditions (OCs). Existing mechanisms for creating the sampling distribution of OCs to generate this data are difficult to visualize, modify, and extend. To address this, we created a user interface and toolchain for multi-modal OC sampling from probabilistic graphical models (PGMs). Our web browser-based interface allows for visualizing the PGMs, modifying the conditional probability distributions of their nodes, importing and exporting their state, operation in single- or multi-modality configurations, and persisting generated samples to a relational database. The Vue-driven web interface, programmatic interface, as well as the Python machinery for OC sampling and persistence have been containerized to allow for simplified deployment and distribution. The work described here supplies a new baseline for OC generation for single- and multi-sensor simulation and fusion.

Decision level fusion algorithms combine separate classification scores of a test sample to make a unified class declaration. The aim of decision level fusion algorithms is to achieve better classification performance by combining decisions rather than picking the single best performing algorithm. Multi-modal fusion can be achieved by fusing scores of deep learning models trained on different sensing modalities and tested on a target imaged from co-located sensors. Given differences in phenomenology, fusing EO sensors with SAR may boost performance when extended operating conditions are detrimental to the performance of one modality over the other. The EO modalities discussed in this work (VNIR and MWIR) are susceptible to the time of day while SAR is robust to time of day. Conversely, SAR returns of a target can vary greatly when aspect angle changes while EO modalities are relatively robust. This work analyzes the effectiveness of decision level fusion algorithms on MWIR, VNIR, and SAR modalities given disparate times of day and collection aspects.

Objective: Adherence to prescribed medical regimens is a key behavior for managing disease burden and progression. Despite the importance of understanding a child’s adherence, many evidence-based assessment strategies have not been implemented into subspecialty clinical practice. The aim of this article was to highlight the potential ethical and regulatory challenges of using electronic monitoring (EM) of adherence in clinical care and offer practical solutions. Methods: Relevant ethical and legal standards were researched and summarized as applicable to EM of adherence. Results: Issues related to Human Relations, Privacy and Confidentiality, Record Keeping, and Assessment all apply to EM of adherence. Conclusions: Clinicians should be mindful and seek to proactively address issues related to informed consent, autonomy, privacy, and equity before starting an EM program of adherence. Multidisciplinary teams should thoroughly consider the ethical and legal aspects of implementing this form of adherence assessment into clinical practice. Implications for Impact Statement This topical review is the first to describe the practical implications of utilizing electronic monitoring of adherence in a clinical care setting with a multidisciplinary medical team. This is an important reference for any pediatric psychologist or practitioner who wants to be fully informed about the potential ethical and legal ramifications of using this technology with youth.

We describe a batched beam decoding algorithm for NMT with LMBR n-gram posteriors, showing that LMBR techniques still yield gains on top of the best recently reported results with Transformers. We also discuss acceleration strategies for deployment, and the effect of the beam size and batching on memory and speed.

In vehicle target classification, contact sensors have frequently been used to collect data to simulate laser vibrometry data. Accelerometer data has been used in numerous literature to test and train classifiers instead of laser vibrometry data [1] [2]. Understanding the key similarities and differences between accelerometer and laser vibrometry data is essential to keep progressing aided vehicle recognition systems. This paper investigates the contrast of accelerometer and laser vibrometer data on classification performance. Research was performed using the end-to-end process previously published by the authors to understand the effects of different types of data on the classification results. The end-to-end process includes preprocessing the data, extracting features from various signal processing literature, using feature selection to determine the most relevant features used in the process, and finally classifying and identifying the vehicles. Three data sets were analyzed, including one collection on military vehicles and two recent collections on civilian vehicles. Experiments demonstrated include: (1) training the classifiers using accelerometer data and testing on laser vibrometer data, (2) combining the data and classifying the vehicle, and (3) different repetitions of these tests with different vehicle states such as idle or revving and varying stationary revolutions per minute (rpm).

Vibration signatures sensed from distant vehicles using laser vibrometry systems provide valuable information that may be used to help identify key vehicle features such as engine type, engine speed, and number of cylinders. Through the use of physics models of the vibration phenomenology, features are chosen to support classification algorithms. Various individual exploitation algorithms were developed using these models to classify vibration signatures into engine type (piston vs. turbine), engine configuration (Inline 4 vs. Inline 6 vs. V6 vs. V8 vs. V12) and vehicle type. The results of these algorithms will be presented for an 8 class problem. Finally, the benefits of using a factor graph representation to link these independent algorithms together will be presented which constructs a classification hierarchy for the vibration exploitation problem.

This paper evaluates and expands upon the existing end-to-end process used for vibrometry target classification and identification. A fundamental challenge in vehicle classification using vibrometry signature data is the determination of robust signal features. The methodology used in this paper involves comparing the performance of features taken from automatic speech recognition, seismology, and structural analysis work. These features provide a means to reduce the dimensionality of the data for the possibility of improved separability. The performances of different groups of features are compared to determine the best feature set for vehicle classification. Standard performance metrics are implemented to provide a method of evaluation. The contribution of this paper is to (1) thoroughly explain the time domain and frequency domain features that have been recently applied to the vehicle classification using laser-vibrometry data domain, (2) build an end-to-end classification pipeline for Aided Target Recognition (ATR) with common and easily accessible tools, and (3) apply feature selection methods to the end-to-end pipeline. The end-to-end process used here provides a structured path for accomplishing vibrometry-based target identification. This paper will compare with two studies in the public domain. The techniques utilized in this paper were utilized to analyze a small in-house database of several different vehicles.

Several important applications in optical communications require high power fiber coupled laser sources operating at 1.3 /spl mu/m, 1.48 /spl mu/m and 1.55 /spl mu/m. Semiconductor diode lasers for these wavelengths can be fabricated using the InP/lnGaAsP material system. High power operation requires careful selection of the active region, p and n doping and the lateral waveguide structure. Additionally, distributed feedback (DFB) lasers require appropriate choice of grating feedback. High power, narrow linewidth and low noise single mode diode laser sources operating at 1.3 /spl mu/m and 1.55 /spl mu/m wavelengths have applications in cable television transmission systems, microwave photonics, interferometric sensors and as an alternative to diode-pumped solid state lasers. High output power is desirable to overcome losses in external modulators and for large scale distribution systems. It is particularly important for 1.3 /spl mu/m applications as reliable and efficient fiber amplifiers are not commercially available at this wavelength. 1.3 /spl mu/m DFB lasers with 440 mW chip power, over 150 mW fiber coupled power and 100 kHz linewidth have been demonstrated.

Highly efficient, 0.5 W, single-mode 980 nm pump modules for erbium-doped optical amplifiers have been developed. The laser diode has been optimized to eliminate the optical kink associated with spatial mode instability thereby allowing a significant increase of the chip power. The optimized chip has a lower series and thermal resistance resulting in a more efficient operation with high reliability. The chip has been incorporated into a newly developed, highly efficient module platform. This module utilizes the industry standard 14-pin 'butterfly' format, and has optimized fiber coupling and thermal design. The combination of an excellent chip and an optimized module design has lead to single mode coupled powers in excess of 0.5 W for telecommunication applications that require high reliability.

We implemented various laser structures in the InGaAlP material system for operation in the 635 - 680 nm wavelength range. By incorporating a grating reflector within the device we can stabilize the lasing frequency, or maintain single spatial mode operation in a broad area device. Such devices include DBRs, tunable DBRs, monolithic MOPAs, DFBs, and angled-grating DFBs. Such performance can also be obtained without a buried grating layer by using an external fiber grating to stabilize the frequency or with a flared structure for high power single mode operation.