Grayscale Optical Correlator Workbench (GOCWB) is a computer program for use in automatic target recognition (ATR). GOCWB performs ATR with an accurate simulation of a hardware grayscale optical correlator (GOC). This simulation is performed to test filters that are created in GOCWB. Thus, GOCWB can be used as a stand-alone ATR software tool or in combination with GOC hardware for building (target training), testing, and optimization of filters. The software is divided into three main parts, denoted filter, testing, and training. The training part is used for assembling training images as input to a filter. The filter part is used for combining training images into a filter and optimizing that filter. The testing part is used for testing new filters and for general simulation of GOC output. The current version of GOCWB relies on the mathematical software tools from MATLAB binaries for performing matrix operations and fast Fourier transforms. Optimization of filters is based on an algorithm, known as OT-MACH, in which variables specified by the user are parameterized and the best filter is selected on the basis of an average result for correct identification of targets in multiple test images.
This program was written by Jay Hanan, Hanying Zhou, and Tien-Hsin Chao of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free online at www.techbriefs.com/tsp under the Software category.
This software is available for commercial licensing. Please contact Karina Edmonds of the California Institute of Technology at (626) 395-2322. Refer to NPO-41021.
This Brief includes a Technical Support Package (TSP).
Grayscale Optical Correlator Workbench
(reference NPO-41021) is currently available for download from the TSP library.
Don't have an account?
Overview
The document outlines the Grayscale Optical Correlator Workbench (GOCWB), a software tool developed by NASA's Jet Propulsion Laboratory (JPL) for simulating hardware Grayscale Optical Correlators (GOCs). The primary purpose of GOCWB is to facilitate automatic target recognition (ATR) by enabling users to build, test, and optimize filters through simulation.
GOCWB is designed with a user-friendly interface that is specifically compatible with 512x512 Spatial Light Modulators (SLMs), which are integral to GOC hardware. The software features three main components: Training, Filter, and Testing. The Training module allows users to assemble training images that serve as input for filter creation. The Filter module combines these training images into a filter and optimizes it using the OT-MACH algorithm, which parameterizes user variables to select the best filter based on average results for correct target identifications across testing images. The Testing module simulates the output of the GOC, allowing users to evaluate the performance of new filters.
One of the key improvements of GOCWB over existing software applications is its enhanced user interface, which accommodates multiple files and configurations more easily. Additionally, the software includes an optimization routine and in-the-loop feedback on filter performance, making it a more effective tool for users engaged in ATR tasks.
The GOCWB relies on mathematical tools from Matlab binaries for performing matrix operations and Fast Fourier Transforms (FFTs). While users do not need to have a deep understanding of Matlab to utilize the software, those familiar with it can easily grasp the underlying logic of the source code.
The document also references several publications that detail the development and performance testing of the GOC and its algorithms, highlighting the software's relevance to aerospace applications, particularly in the context of spacecraft precision landing and other advanced recognition tasks.
In summary, the Grayscale Optical Correlator Workbench is a sophisticated simulation tool that enhances the capabilities of ATR through effective filter generation, optimization, and testing, making it a valuable asset for researchers and engineers in the field of aerospace technology.
