Accession Number : ADA326143
Title : MEMS-Based Diagnostics for Turbulent Shear Flows.
Descriptive Note : Final rept. 15 Aug 93-14 Feb 97,
Corporate Author : GEORGIA INST OF TECH ATLANTA
Personal Author(s) : Glezer, Ari ; Allen, Mark ; Brooke, Martin
PDF Url : ADA326143
Report Date : APR 1997
Pagination or Media Count : 11
Abstract : The objective of this section of the research is to develop hardware for fast processing of charge coupled display (CCD) images. This hardware will take digital CCD output and process it at the same output data rate of the CCD. The result is processed images at the maximum rate possible with a CCD camera. In particular, the hardware will perform convolutions of the input images with a digitally stored convolution kernal matrix, such as are needed for image spatial filtering and neural networks. A hardware test chip has been completed, characterized, and has been used to process images. This chip does not contain the digital front end necessary to connect directly to a CCD data stream but does demonstrate all of the critical analog components of the image processor. The test chip will be integrated with a shift register in a later chip fabrication to produce a fully functional image processor chip capable of connection to the data stream directly output from a CCD or similar row-wise serial output imaging device.
Descriptors : *IMAGE PROCESSING, *TURBULENT FLOW, *SHEAR FLOW, *CHARGE COUPLED DEVICES, SIGNAL PROCESSING, NEURAL NETS, DATA MANAGEMENT, CHIPS(ELECTRONICS), INTEGRATED CIRCUITS, ANALOG TO DIGITAL CONVERTERS, SPATIAL FILTERING, ANALOG SIGNALS, HYDRAULIC ACTUATORS.
Subject Categories : Electrooptical and Optoelectronic Devices
Distribution Statement : APPROVED FOR PUBLIC RELEASE