Hyper-spectral image processing applications on the SIMD Pixel Processor for the digital battlefield

Author

Chai, Sek M. ; Gentile, Antonio ; Lugo-Beauchamp, Wilfredo E. ; Cruz-Rivera, José L. ; Wills, D. Scott

Author_Institution

Dept. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA

fYear

1999

fDate

1999

Firstpage

130

Lastpage

138

Abstract

Future military scenarios will rely on advanced imaging sensor technology beyond the visible spectrum to gain total battlefield awareness. Real-time processing of these data streams requires tremendous computational workloads and I/O throughputs. This paper presents three applications for hyper-spectral data streams, vector quantization, region autofocus, and K-means clustering, on the SIMD Pixel Processor (SIMPil). In SIMPil, an image sensor array (focal plane) is integrated on top of a SIMD computing layer to provide direct coupling between sensors and processors, alleviating I/O bandwidth bottlenecks while maintaining low power consumption and portability. Simulation results with sustained operation throughputs of 500-1500 Gops/sec support real-time performance and promote focal plane processing on SIMPil

Keywords

image processing; image sensors; military computing; object detection; parallel machines; real-time systems; vector quantisation; K-means clustering; SIMD Pixel Processor; SIMPil; computational workloads; digital battlefield; focal plane; hyperspectral image processing; image sensor array; imaging sensor; input output bandwidth bottlenecks; low power consumption; military; portability; real-time processing; region autofocus; simulation; vector quantization; visible spectrum; Bandwidth; Image processing; Image sensors; Military computing; Optical imaging; Pixel; Sensor arrays; Streaming media; Throughput; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Vision Beyond the Visible Spectrum: Methods and Applications, 1999. (CVBVS '99) Proceedings. IEEE Workshop on

Conference_Location

Fort Collins, CO

Print_ISBN

0-7695-0050-1

Type

conf

DOI

10.1109/CVBVS.1999.781102

Filename

781102