B. Lazzerini, L. Reyneri, F. Gregoretti, A. Mariani
{"title":"High-Speed Recognition of Handwritten Amounts on Italian Checks","authors":"B. Lazzerini, L. Reyneri, F. Gregoretti, A. Mariani","doi":"10.1201/9781003069379-8","DOIUrl":null,"url":null,"abstract":"In this chapter we describe an Italian bank check processing system. The system is made up of several processing modules, including those for data acquisition, image preprocessing, character center detection, character recognition, courtesy amount recognition, and legal amount recognition. The system reads and recognizes (at the character level) the courtesy and the legal amounts separately. Then the two streams of information are sent to a context analysis subsystem, which exploits the mutual redundancy in the courtesy and legal amounts. The output of the system is a list of possible amounts, which are sorted according to a decreasing recognition con dence. The system has an accuracy of 53% and can process up to three bank checks per second. The proposed system is partially implemented on an ad-hoc VLSI chip containing an array of dedicated processors tailored to the application, and tightly interconnected to a host personal computer.","PeriodicalId":177083,"journal":{"name":"KNOWLEDGE-BASED INTELLIGENT TECHNIQUES in CHARACTER RECOGNITION","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1999-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"KNOWLEDGE-BASED INTELLIGENT TECHNIQUES in CHARACTER RECOGNITION","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1201/9781003069379-8","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In this chapter we describe an Italian bank check processing system. The system is made up of several processing modules, including those for data acquisition, image preprocessing, character center detection, character recognition, courtesy amount recognition, and legal amount recognition. The system reads and recognizes (at the character level) the courtesy and the legal amounts separately. Then the two streams of information are sent to a context analysis subsystem, which exploits the mutual redundancy in the courtesy and legal amounts. The output of the system is a list of possible amounts, which are sorted according to a decreasing recognition con dence. The system has an accuracy of 53% and can process up to three bank checks per second. The proposed system is partially implemented on an ad-hoc VLSI chip containing an array of dedicated processors tailored to the application, and tightly interconnected to a host personal computer.