The novel method for automatic pattern recognition is presented. This method is based on Segment and Neighbors Matching algorithm which can be applied for recognizing of distinct well-known alphabets, complex glyphs, and Arabic scripts. In this work some different reported methods have been evaluated on Latin, Chinese characters, and Mayan glyphs with the principle objective to select those with the highest processing speed and recognition grade. The case of Mayan glyphs is more complicated due to a big number of elements in any glyph, significant variations of their representation or writing, there are more than 800 classes of glyphs and many of them with similar components and locations. The proposed method of Segments and Neighbors Matching has been developed on base of fuzzy sets and membership functions concept which can be defined during manipulation with the glyph skeleton. Next, levels of matching with predefined patterns are used for segments recognition and interpretation of whole glyph. The main characteristics of recognizing process are matching level, time of processing, grade of membership, and efficiency of interpretation that is important for incomplete glyphs images. On base of proposed method the special software RECGLYM (Mayan Glyphs Recognition) has been designed for the SUN and Intel PC computers platforms. The advantages of the proposed Segments and Neighbors Matching method are quick image processing and high probability of complex glyphs interpretation. The proposed method could be used in different applications, for example, for selection and diagnose of certain anomalies by means of processing of X-ray images or for Internet navigation and searching information searching by image similarity analysis with predefined pattern.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Oleg STAROSTENKO, Jose Antonio NEME, "Automatic Complex Glyphs Recognition and Interpretation" in IEICE TRANSACTIONS on Fundamentals,
vol. E82-A, no. 10, pp. 2154-2160, October 1999, doi: .
Abstract: The novel method for automatic pattern recognition is presented. This method is based on Segment and Neighbors Matching algorithm which can be applied for recognizing of distinct well-known alphabets, complex glyphs, and Arabic scripts. In this work some different reported methods have been evaluated on Latin, Chinese characters, and Mayan glyphs with the principle objective to select those with the highest processing speed and recognition grade. The case of Mayan glyphs is more complicated due to a big number of elements in any glyph, significant variations of their representation or writing, there are more than 800 classes of glyphs and many of them with similar components and locations. The proposed method of Segments and Neighbors Matching has been developed on base of fuzzy sets and membership functions concept which can be defined during manipulation with the glyph skeleton. Next, levels of matching with predefined patterns are used for segments recognition and interpretation of whole glyph. The main characteristics of recognizing process are matching level, time of processing, grade of membership, and efficiency of interpretation that is important for incomplete glyphs images. On base of proposed method the special software RECGLYM (Mayan Glyphs Recognition) has been designed for the SUN and Intel PC computers platforms. The advantages of the proposed Segments and Neighbors Matching method are quick image processing and high probability of complex glyphs interpretation. The proposed method could be used in different applications, for example, for selection and diagnose of certain anomalies by means of processing of X-ray images or for Internet navigation and searching information searching by image similarity analysis with predefined pattern.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e82-a_10_2154/_p
Copy
@ARTICLE{e82-a_10_2154,
author={Oleg STAROSTENKO, Jose Antonio NEME, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Automatic Complex Glyphs Recognition and Interpretation},
year={1999},
volume={E82-A},
number={10},
pages={2154-2160},
abstract={The novel method for automatic pattern recognition is presented. This method is based on Segment and Neighbors Matching algorithm which can be applied for recognizing of distinct well-known alphabets, complex glyphs, and Arabic scripts. In this work some different reported methods have been evaluated on Latin, Chinese characters, and Mayan glyphs with the principle objective to select those with the highest processing speed and recognition grade. The case of Mayan glyphs is more complicated due to a big number of elements in any glyph, significant variations of their representation or writing, there are more than 800 classes of glyphs and many of them with similar components and locations. The proposed method of Segments and Neighbors Matching has been developed on base of fuzzy sets and membership functions concept which can be defined during manipulation with the glyph skeleton. Next, levels of matching with predefined patterns are used for segments recognition and interpretation of whole glyph. The main characteristics of recognizing process are matching level, time of processing, grade of membership, and efficiency of interpretation that is important for incomplete glyphs images. On base of proposed method the special software RECGLYM (Mayan Glyphs Recognition) has been designed for the SUN and Intel PC computers platforms. The advantages of the proposed Segments and Neighbors Matching method are quick image processing and high probability of complex glyphs interpretation. The proposed method could be used in different applications, for example, for selection and diagnose of certain anomalies by means of processing of X-ray images or for Internet navigation and searching information searching by image similarity analysis with predefined pattern.},
keywords={},
doi={},
ISSN={},
month={October},}
Copy
TY - JOUR
TI - Automatic Complex Glyphs Recognition and Interpretation
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2154
EP - 2160
AU - Oleg STAROSTENKO
AU - Jose Antonio NEME
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E82-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 1999
AB - The novel method for automatic pattern recognition is presented. This method is based on Segment and Neighbors Matching algorithm which can be applied for recognizing of distinct well-known alphabets, complex glyphs, and Arabic scripts. In this work some different reported methods have been evaluated on Latin, Chinese characters, and Mayan glyphs with the principle objective to select those with the highest processing speed and recognition grade. The case of Mayan glyphs is more complicated due to a big number of elements in any glyph, significant variations of their representation or writing, there are more than 800 classes of glyphs and many of them with similar components and locations. The proposed method of Segments and Neighbors Matching has been developed on base of fuzzy sets and membership functions concept which can be defined during manipulation with the glyph skeleton. Next, levels of matching with predefined patterns are used for segments recognition and interpretation of whole glyph. The main characteristics of recognizing process are matching level, time of processing, grade of membership, and efficiency of interpretation that is important for incomplete glyphs images. On base of proposed method the special software RECGLYM (Mayan Glyphs Recognition) has been designed for the SUN and Intel PC computers platforms. The advantages of the proposed Segments and Neighbors Matching method are quick image processing and high probability of complex glyphs interpretation. The proposed method could be used in different applications, for example, for selection and diagnose of certain anomalies by means of processing of X-ray images or for Internet navigation and searching information searching by image similarity analysis with predefined pattern.
ER -