G. Eberhard, H. Groeneveld, O. Schneider, P. Simons
{"title":"一种实时图像处理芯片组","authors":"G. Eberhard, H. Groeneveld, O. Schneider, P. Simons","doi":"10.1109/ISSCC.1989.48242","DOIUrl":null,"url":null,"abstract":"A chip set is described which meets the requirements of calculation-intensive image processing algorithm (e.g. correlation) for real-time execution. It calculates one million two-dimensional correlation coefficients per second on a 16*16 pixel image at 25 MHz with a 6-b input resolution. The set consists of two chips, a multiplier-accumulator unit (MAC) and an arithmetic processing unit (APU). The MAC processes image data of different formats, up to 32 by 32 pixels, with a 6- or 12-b resolution, and outputs sums of products in an 18- or 36-b integer format. The APU converts these sums to a single-precision floating-point format (the internal data format) and calculates the correlation coefficient in 1- mu s with floating-point precision using the standard formula. A system configuration with both chips, a host processor, and input/output components is shown. The chip characteristics are summarized, and chip micrographs are shown.<<ETX>>","PeriodicalId":385838,"journal":{"name":"IEEE International Solid-State Circuits Conference, 1989 ISSCC. Digest of Technical Papers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1989-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"A realtime image processing chip set\",\"authors\":\"G. Eberhard, H. Groeneveld, O. Schneider, P. Simons\",\"doi\":\"10.1109/ISSCC.1989.48242\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A chip set is described which meets the requirements of calculation-intensive image processing algorithm (e.g. correlation) for real-time execution. It calculates one million two-dimensional correlation coefficients per second on a 16*16 pixel image at 25 MHz with a 6-b input resolution. The set consists of two chips, a multiplier-accumulator unit (MAC) and an arithmetic processing unit (APU). The MAC processes image data of different formats, up to 32 by 32 pixels, with a 6- or 12-b resolution, and outputs sums of products in an 18- or 36-b integer format. The APU converts these sums to a single-precision floating-point format (the internal data format) and calculates the correlation coefficient in 1- mu s with floating-point precision using the standard formula. A system configuration with both chips, a host processor, and input/output components is shown. The chip characteristics are summarized, and chip micrographs are shown.<<ETX>>\",\"PeriodicalId\":385838,\"journal\":{\"name\":\"IEEE International Solid-State Circuits Conference, 1989 ISSCC. Digest of Technical Papers\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1989-02-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE International Solid-State Circuits Conference, 1989 ISSCC. Digest of Technical Papers\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISSCC.1989.48242\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE International Solid-State Circuits Conference, 1989 ISSCC. Digest of Technical Papers","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISSCC.1989.48242","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A chip set is described which meets the requirements of calculation-intensive image processing algorithm (e.g. correlation) for real-time execution. It calculates one million two-dimensional correlation coefficients per second on a 16*16 pixel image at 25 MHz with a 6-b input resolution. The set consists of two chips, a multiplier-accumulator unit (MAC) and an arithmetic processing unit (APU). The MAC processes image data of different formats, up to 32 by 32 pixels, with a 6- or 12-b resolution, and outputs sums of products in an 18- or 36-b integer format. The APU converts these sums to a single-precision floating-point format (the internal data format) and calculates the correlation coefficient in 1- mu s with floating-point precision using the standard formula. A system configuration with both chips, a host processor, and input/output components is shown. The chip characteristics are summarized, and chip micrographs are shown.<>