{"title":"A data-flow processor for real-time low-level image processing","authors":"G. Quénot, B. Zavidovique","doi":"10.1109/EUASIC.1991.212887","DOIUrl":null,"url":null,"abstract":"A chip featuring two coupled data-flow processors (DFPs) has been designed. It is to be mesh-connected into large processor arrays dedicated primarily to image processing. Each processor operates on 25 MBytes/s data-flows and performs up to 50 million 8- or 16-bit arithmetic operations per second. The chip has been processed in a 1 mu m CMOS technology. It includes 160000 transistors in a 84 mm/sup 2/ die size area, its clock is at 25 MHz and it is packaged in a 144-pin PGA package. Computations are performed on the fly on a data-flow that comes from a digital video camera. One physical operator is associated to each operation involved in the algorithm. An experimental data-flow system including eight processors in a 2*2*2 3D network has been built. Edge detection, row sums, column sums and histograms have been implemented on it at digital video speed.<<ETX>>","PeriodicalId":118990,"journal":{"name":"Euro ASIC '91","volume":"245 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1991-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Euro ASIC '91","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EUASIC.1991.212887","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8
Abstract
A chip featuring two coupled data-flow processors (DFPs) has been designed. It is to be mesh-connected into large processor arrays dedicated primarily to image processing. Each processor operates on 25 MBytes/s data-flows and performs up to 50 million 8- or 16-bit arithmetic operations per second. The chip has been processed in a 1 mu m CMOS technology. It includes 160000 transistors in a 84 mm/sup 2/ die size area, its clock is at 25 MHz and it is packaged in a 144-pin PGA package. Computations are performed on the fly on a data-flow that comes from a digital video camera. One physical operator is associated to each operation involved in the algorithm. An experimental data-flow system including eight processors in a 2*2*2 3D network has been built. Edge detection, row sums, column sums and histograms have been implemented on it at digital video speed.<>