A Study of Improvements for the Winbank Courtesy Amount Recognition System
نویسنده
چکیده
Four separate improvements to the Winbank Courtesy Amount Recognition System are presented in this paper. The first alteration is a grammar engine to allow for improved recognition and handling of punctuation. An adjustment to the pre-processing steps is also presented which deals with the problem of non-uniformity in character size. A system for feedback between the segmentation and recognition algorithms is introduced which allows the system to handle splitting and merging of unrecognized characters. Finally, solutions to two specific problematic character recognitions are presented. Tests with experimental data from bank-check images show that these alterations improve the recognition rates of the system. Thesis Supervisor: Amar Gupta Title: Senior Research Scientist, MIT Sloan School of Management
منابع مشابه
Adding Feedback to Improve Segmentation and Recognition of
WinBank is a system which performs automated reading of handwritten documents, particularly of strings of handwritten numerals on bank checks. A commonly-used strategy for reading handwritten strings is a combination of segmentation and recognition. In the WinBank program, segmentation involves both the separation of touching characters and the merging of character fragments to other pieces. An...
متن کاملRecognition of courtesy amounts on bank checks based on a segmentation approach
Recognition of Courtesy Amounts on Bank Checks based on a Segmentation Approach Li Qiang Shang X segmentation based courtesy amount recognition (CAR) system is presented in this thesis. This system consists of four modules, which are preprocessing, segmentation, recognition and syntacticai analysis. The noise is fht removed in the preprocessing module. Then, the segmentation module seeks to ext...
متن کاملHigh-speed Recognition of Handwritten Amounts on Italian Checks
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. Th...
متن کاملA knowledge-based segmentation algorithm for enhanced recognition of handwritten courtesy amounts
A knowledge based segmentation critic algorithm to enhance recognition of courtesy amounts on bank checks is proposed in this paper. This algorithm extracts the context from the handwritten material and uses a syntax parser based on a deterministic finite automaton to provide adequate feedback to enhance recognition. The segmentation critic presented is capable of handling a number of commonly ...
متن کاملSpeaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...
متن کامل