Optical Recognition of Chinese Characters by Richard Suchenwirth, Jun Guo, Irmfried Hartmann, Georg

By Richard Suchenwirth, Jun Guo, Irmfried Hartmann, Georg Hincha, Manfred Krause, Zheng Zhang

Ebook by way of NA, NA

Show description

Read or Download Optical Recognition of Chinese Characters PDF

Similar control systems books

Adaptive Control of Ill-Defined Systems

There are a few sorts of advanced structures which are equipped like clockwork, with well-defined components that have interaction in well-defined methods, in order that the motion of the full should be accurately analyzed and expected with accuracy and precision. a few structures should not themselves so well-defined, yet they are often modeled in ways in which are like proficient pilots in well-built planes, or electrolyte stability in fit people.

Modeling and Control of Economic Systems 2001. A Proceedings volume from the 10th IFAC Symposium, Klagenfurt, Austria, 6 – 8 September 2001

This quantity includes papers provided on the IFAC symposium on Modeling and keep an eye on of monetary platforms (SME 2001), which used to be held on the collage of Klagenfurt, Austria. The symposium introduced jointly scientists and clients to discover present theoretical advancements of modeling options for monetary platforms.

Cyber Security: Analytics, Technology and Automation

The e-book, as well as the cyber threats and know-how, tactics cyber protection from many aspects as a social phenomenon and the way the implementation of the cyber safeguard method is performed. The ebook provides a profound suggestion of the main spoken phenomenon of this time. The ebook is acceptable for a wide-ranging viewers from graduate to professionals/practitioners and researchers.

Extra info for Optical Recognition of Chinese Characters

Sample text

Any two neighbouring sub-sequences must have different values. 5-4) where N5 is the number of sub-sequences, W is the number of white pixels, B is the number of black pixels and e is a threshold. When a defective edge is detected, it is smoothed out. lf the number of white pixels is lass than the number of black pixels, all white pixels are replaced with black pixels; otherwise black pixels are replaced with white pixels. Fti;. 1 Principles of Feature Extraction Feature extraction is one of the crucial stages in all pattern recognition systems.

4-2). This prevents the Suppression ofthin lines and reduces the influence of inhomogenaus lighting. Let s(i,j) be the grayscale image of a character with N rows and M columns. 5 Edge Smoothing on Binary Patterns ln the binary image of printed Chinese characters, there are often a Iot of defects on some edges of strokes, especially the horizontal and vertical strokes (see examples in fig. 5-1). g. the SDF feature), making them rather unstable. When such features are used, preprocessing must include edge smoothing.

4-3 Structure of the TECH/S recognition system recognition sottware. See [McC 87] for a recent overview of alphanumeric OCR systems that are available in the market. A prototype Kanji OCR system is described in [Hir 80]. Fig. 4-3 shows the rough structure of the experimental TECHIS system developed by the authors at the Technical University of Berlin. lf optical recognition of Chinese characters is to be used in real-life applications, the input technology must allow the reading of at least one page, or even better an opening (two facing pages), in one exposition.

Download PDF sample

Rated 4.75 of 5 – based on 50 votes