Detection and Recognition of 3-D Targets in Panchromatic and in Synthetic Aperture Radar Imagery Using a Map-Seeking Circuit Algorithm

作者：Murphy Patricia K; Rodriguez Pedro A; Peterson Cameron K

来源：Johns Hopkins APL Technical Digest, 2013, 31(3): 234-253.

A 3-D target detection and recognition algorithm, based on the biologically inspired map-seeking circuit (MSC), is implemented to efficiently solve the template-matching problem in synthetic aperture radar (SAR) and panchromatic grayscale imagery. Given a 3-D model of a target, this algorithm locates the target in a 2-D image and determines its pose (i.e., viewing angles, scale, and spatial translations). A key aspect of the MSC is the simultaneous forward transformation of the model to match the image coupled with a backward path to make the image match the model. The efficiency of the algorithm is a result of the decomposition of the n-dimensional pose transformation space into a series of one-dimensional searches for each of the transformation parameters. Although originally designed for panchromatic electro-optical imagery, we demonstrate that the MSC architecture can also be successfully applied to SAR by simply changing the feature-extraction preprocessing. Additionally, we introduce modifications to the MSC algorithm that increase the speed of detection and allow efficient classification when multiple targets are present in the same image. We present promising results after applying our algorithm to challenging real-world panchromatic electro-optical and SAR imagery.

• 出版日期2013