Target detection utilizing image array comparison

摘要

A method and system for forming an image comprising at least one processor for performing the following: initializing an N by P image array IO by setting all values to a large number; inputting at least one image frame; randomly selecting and removing a subset of pixel locations from the total number of available pixel locations to form a preliminary image array IC; for each pixel location comparing the complex magnitude of each pixel in the image array IO with the magnitude of the corresponding pixel in image array IC, and if the pixel value in the image array IC is smaller than the corresponding pixel in the image array IO or if the IC value equals 0, then the current pixel value in image array IO is replaced by the pixel value in the image array IC; and repeating for a number of iterations to form an image.

主权项

1. A method for generating images comprising:
receiving, via a radar, radar return signals from an area of interest at a number of radar positions; acquiring radar return data and positional data from the radar; each of the positional data forming a positional record in an array of positional records; each of the radar data comprising a first number of radar data points in the frequency domain; forming an image comprising an array of pixel values comprising:
a) randomly setting a percentage of the first number of radar data points to zero; thereby selecting subsets of receive channels for removal from the complete first number of radar data points;b) transforming the first number of radar data points into time domain signature data;c) differentiating the signature data in the time domain to create a new set of signature data indicating the location of moving targets;d) forming an image by processing the signature data utilizing an imaging subsystem that combines the positional and signature data to form the moving target indication image utilizing one of a back-projection algorithm or range migration algorithm;e) saving the pixel values for each of the image pixels into a composite image;f) repeating the step of randomly setting a percentage of the radar data points to zero;g) transforming the first number of radar data points into time domain signature data;h) differentiating the signature data in the time domain to create a new set of signature data indicating the location of moving targets;i) forming a second image by processing the signature data utilizing an imaging subsystem that combines the positional and signature data to form SAR imagery utilizing one of a back-projection algorithm or range migration algorithm;j) comparing the pixel values at each of the pixel locations in the second image to the corresponding pixel values of the composite image;k) replacing the pixel values in the composite image if the pixel value is less in the second image than it is in the composite image;l) repeating steps i) through k) for a pre-determined number of iterations; whereby the composite image is outputted with reduced sidelobes and noise.