An improved algorithm for segmenting and recognizing connected handwritten characters

Xiaoyu Zhao; Zheru Chi; Dagan Feng

doi:10.1109/ICARCV.2010.5707382

An improved algorithm for segmenting and recognizing connected handwritten characters

Xiaoyu Zhao, Zheru Chi, Dagan Feng

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

4 Citations (Scopus)

Abstract

In this paper, an improved algorithm is proposed for the segmentation and recognition of handwritten character strings. In the method, a gradient descent mechanism is used to weigh the distance measure in applying KNN for segmenting/recognizing connected characters (numerals and Chinese characters) in the left-to-right scanning direction. In recognizing connected characters, a high quality segmentation technique is essential. Conventional approaches attempt to separate the string into individual characters without recognition and apply a recognition algorithm onto each isolated character, resulting improper segmentation and poor recognition results in many situations. Our proposed algorithm simulates the human beings's process in recognizing connected character strings where segmentation and recognition is mingled with each other. Experimental results on 1959 character strings from the USPS database of postal envelopes show that the algorithm works robustly and efficiently.

Original language	English
Title of host publication	11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010
Pages	1611-1615
Number of pages	5
DOIs	https://doi.org/10.1109/ICARCV.2010.5707382
Publication status	Published - 1 Dec 2010
Event	11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010 - Singapore, Singapore Duration: 7 Dec 2010 → 10 Dec 2010

Conference

Conference	11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010
Country/Territory	Singapore
City	Singapore
Period	7/12/10 → 10/12/10

Keywords

Character recognition
Connected handwritten character
Segmentation-recognition
String segmentation

ASJC Scopus subject areas

Artificial Intelligence
Control and Systems Engineering

More information

10.1109/ICARCV.2010.5707382

Cite this

@inproceedings{c8c9f910858448dd9b2a02faa9f1a6c4,

title = "An improved algorithm for segmenting and recognizing connected handwritten characters",

abstract = "In this paper, an improved algorithm is proposed for the segmentation and recognition of handwritten character strings. In the method, a gradient descent mechanism is used to weigh the distance measure in applying KNN for segmenting/recognizing connected characters (numerals and Chinese characters) in the left-to-right scanning direction. In recognizing connected characters, a high quality segmentation technique is essential. Conventional approaches attempt to separate the string into individual characters without recognition and apply a recognition algorithm onto each isolated character, resulting improper segmentation and poor recognition results in many situations. Our proposed algorithm simulates the human beings's process in recognizing connected character strings where segmentation and recognition is mingled with each other. Experimental results on 1959 character strings from the USPS database of postal envelopes show that the algorithm works robustly and efficiently.",

keywords = "Character recognition, Connected handwritten character, Segmentation-recognition, String segmentation",

author = "Xiaoyu Zhao and Zheru Chi and Dagan Feng",

year = "2010",

month = dec,

day = "1",

doi = "10.1109/ICARCV.2010.5707382",

language = "English",

isbn = "9781424478132",

pages = "1611--1615",

booktitle = "11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010",

note = "11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010 ; Conference date: 07-12-2010 Through 10-12-2010",

}

Zhao, X, Chi, Z & Feng, D 2010, An improved algorithm for segmenting and recognizing connected handwritten characters. in 11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010., 5707382, pp. 1611-1615, 11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010, Singapore, Singapore, 7/12/10. https://doi.org/10.1109/ICARCV.2010.5707382

An improved algorithm for segmenting and recognizing connected handwritten characters. / Zhao, Xiaoyu; Chi, Zheru; Feng, Dagan.
11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010. 2010. p. 1611-1615 5707382.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - An improved algorithm for segmenting and recognizing connected handwritten characters

AU - Zhao, Xiaoyu

AU - Chi, Zheru

AU - Feng, Dagan

PY - 2010/12/1

Y1 - 2010/12/1

N2 - In this paper, an improved algorithm is proposed for the segmentation and recognition of handwritten character strings. In the method, a gradient descent mechanism is used to weigh the distance measure in applying KNN for segmenting/recognizing connected characters (numerals and Chinese characters) in the left-to-right scanning direction. In recognizing connected characters, a high quality segmentation technique is essential. Conventional approaches attempt to separate the string into individual characters without recognition and apply a recognition algorithm onto each isolated character, resulting improper segmentation and poor recognition results in many situations. Our proposed algorithm simulates the human beings's process in recognizing connected character strings where segmentation and recognition is mingled with each other. Experimental results on 1959 character strings from the USPS database of postal envelopes show that the algorithm works robustly and efficiently.

AB - In this paper, an improved algorithm is proposed for the segmentation and recognition of handwritten character strings. In the method, a gradient descent mechanism is used to weigh the distance measure in applying KNN for segmenting/recognizing connected characters (numerals and Chinese characters) in the left-to-right scanning direction. In recognizing connected characters, a high quality segmentation technique is essential. Conventional approaches attempt to separate the string into individual characters without recognition and apply a recognition algorithm onto each isolated character, resulting improper segmentation and poor recognition results in many situations. Our proposed algorithm simulates the human beings's process in recognizing connected character strings where segmentation and recognition is mingled with each other. Experimental results on 1959 character strings from the USPS database of postal envelopes show that the algorithm works robustly and efficiently.

KW - Character recognition

KW - Connected handwritten character

KW - Segmentation-recognition

KW - String segmentation

UR - http://www.scopus.com/inward/record.url?scp=79952391704&partnerID=8YFLogxK

U2 - 10.1109/ICARCV.2010.5707382

DO - 10.1109/ICARCV.2010.5707382

M3 - Conference article published in proceeding or book

SN - 9781424478132

SP - 1611

EP - 1615

BT - 11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010

T2 - 11th International Conference on Control, Automation, Robotics and Vision, ICARCV 2010

Y2 - 7 December 2010 through 10 December 2010

ER -

An improved algorithm for segmenting and recognizing connected handwritten characters

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this