Lockheed Martin Corporation

Barry G Mattox

Gene D Tener

Hai-Wen Chen

Timothy Alderson

Jennifer K Park

James P Kane, Iii

David J Huber

William R Herzog

John Cowan

Methods and systems for image processing are provided. A method for processing images of a scene includes receiving image data of a reference and a current frame; generating N motion vectors that describe motion of the image data within the scene by computing a correlation function on the reference and current frames at each of N registration points; registering the current frame based on the N motion vectors to produce a registered current frame; and updating the image data of the scene based on the registered current frame. Optionally, registered frames may be oversampled. Techniques for generating the N motion vectors according to roll, zoom, shift and optical flow calculations, updating image data of the scene according to switched and intermediate integration approaches, re-introducing smoothed motion into image data of the scene, re-initializing the process, and processing images of a scene and moving target within the scene are provided.

Continuous extended range image processing

This invention solves a problem of registration and improves signal-to-noise ratio (SNR) when using division-by-focal-plane array to produce multiple polarization images. This is achieved by processing a sequence of angular-position-dithered frames to generate a high-definition, Nyquist-sampled, integrated image for each of the polarizations. The integration method transforms individually under-sampled, high-resolution image frames into resultant high-resolution frames that meet the Nyquist sampling criterion. During the resampling transformation, each polarization or waveband is resampled to produce precise registration to the other polarizations, since registration offsets are fixed and defined by the arrangement of the polarized pixels in the focal-plane-array. In the most straight-forward implementation, these offsets would be integer pixel shifts in X and Y.

Method for enhancing polarimeter systems that use micro-polarizers

A method and apparatus for accurately and precisely controlling the frequency (wavelength) and periodic frequency modulation of a laser are provided. An ADC (Analog to Digital Converter) is used to sample the output of a modified interferometer used as a delay line discriminator, and quadrature components of the sampled output are generated. An arctangent function (e.g., atan ) is applied to convert the quadrature components to a phase measure that is proportional to the laser frequency. Correlator circuits (e.g., cost-efficient correlator circuits) are provided to extract average frequency, modulation peak deviation, and modulation phase error signals. Control-loop feedback using the extracted signals is used to adjust the average frequency, modulation deviation, and modulation phase to respective set points.

Method for producing highly accurate frequency and FM of a laser

A method and apparatus for accurately and precisely controlling the frequency (wavelength) and periodic frequency modulation of a laser are provided. An ADC (Analog to Digital Converter) is used to sample the output of a modified interferometer used as a delay line discriminator, and quadrature components of the sampled output are generated. An arctangent function (e.g., atan2) is applied to convert the quadrature components to a phase measure that is proportional to the laser frequency. Correlator circuits (e.g., cost-efficient correlator circuits) are provided to extract average frequency, modulation peak deviation, and modulation phase error signals. Control-loop feedback using the extracted signals is used to adjust the average frequency, modulation deviation, and modulation phase to respective set points.

In a homodyne-receiver radar system used for ranging and/or target detection, the frequency-modulated (FM) transmit phase are removed from the received signal. Removal of the known FM transmit phase from the received signal reduces the bandwidth of the received signal to half that of a system that does not remove the transmit phase. This allows the sampling rate, and thus the processor throughput, to be cut in half. With the FM transmit phase removed, the phase sequence of the processed signal is akin to a delayed version of the transmit signal phase history. This allows the range processing to use segments of a single phase sequence for processing all range gates, resulting in a large reduction in the amount of coefficient storage used for the matched-phase sequences required to process the set of range gates.

Transmit phase removal in FM homodyne radars

A radar signal processing method and system for detecting target objects of unknown acceleration and having low SNRs which reduces the computational burdens and provides a more efficient way of performing the operation of non-coherent integration. Radar signal processing is conducted according to a predetermined scheme in which partially processed received signal data is selectively stored and reused, reducing redundant processing. The radar system receives return signals frequency shifted from a predetermined frequency scheme by unknown amounts. The received signals are coherently integrated transforming them into frequency domain templates which are non-coherently arranged into an array matrix. The data of the frequency domain templates are processed to form presums which are stored for use in forming higher level presums and for forming acceleration bins. Once the acceleration bins have been formed, they are analyzed to detect the presence of target object return signals.

Integrator for radar return signals with high acceleration uncertainty

Dual path detection processing in which a low SNR signal processor detects signals over a limited range of low acceleration values and a high SNR signal processor detects signals over a wider range of acceleration values. The low SNR signal processor uses acceleration bins formed from a noncoherent FFT array to detect low SNR signals of far away objects which tend to have lower angular acceleration. Because close proximity target objects tend to have higher SNR return signals, it is not necessary to rely on acceleration bins formed from an FFT array for signal detection. Close proximity targets with high SNR can often be detected in individual coherently integrated FFT templates, despite the likelihood of large acceleration uncertainty from higher angular acceleration rates. Since signal detection of higher SNR signals is much less computationally burdensome than signal detection using a noncoherent FFT array, the present inventor is able to realize computational efficiencies by using dual path detection processing.

Growing community of inventors

Orlando, FL, United States of America

Barry G Mattox