Development of multistage algorithm for text objects recognition in images

Abstract

The relevance of this study is stipulated by the necessity of designing algorithms allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Purpose: Development of algorithms and software system allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Experimental research to be carried out into the efficiency of developed algorithms and their comparison to their existing analogs. Findings: Viola-Jones object detection algorithm is suggested to detect the area of character location on images with complex background. The image normalization includes the following operations: boundary identification, Hough transformation, search for closed loops, image leveling. The structure of convolutional neural network (CNN) is suggested for characters recognition. The implementation of the developed algorithms is presented by characters recognition on images of vehicle number plates.

Original language	English
Title of host publication	2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781467383837
DOIs	https://doi.org/10.1109/SIBCON.2016.7491714
Publication status	Published - 14 Jun 2016
Event	2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Moscow, Russian Federation Duration: 12 May 2016 → 14 May 2016

Other

Other	2016 International Siberian Conference on Control and Communications, SIBCON 2016
Country	Russian Federation
City	Moscow
Period	12.5.16 → 14.5.16

Keywords

artificial neural networks
character recognition
Hough transform
Image processing
Viola-Jones algorithm

ASJC Scopus subject areas

Signal Processing
Control and Systems Engineering
Electrical and Electronic Engineering
Modelling and Simulation
Computer Networks and Communications

Access to Document

10.1109/SIBCON.2016.7491714

Fingerprint Dive into the research topics of 'Development of multistage algorithm for text objects recognition in images'. Together they form a unique fingerprint.

Cite this

Development of multistage algorithm for text objects recognition in images. / Cherneta, D. S.; Druki, Alexey Alexeevich ; Spitsyn, V. G.

2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. 7491714.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Cherneta, DS, Druki, AA & Spitsyn, VG 2016, Development of multistage algorithm for text objects recognition in images. in 2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Proceedings., 7491714, Institute of Electrical and Electronics Engineers Inc., 2016 International Siberian Conference on Control and Communications, SIBCON 2016, Moscow, Russian Federation, 12.5.16. https://doi.org/10.1109/SIBCON.2016.7491714

@inproceedings{6e4e5c7b6e3344669be9e49ef555a7db,

title = "Development of multistage algorithm for text objects recognition in images",

abstract = "The relevance of this study is stipulated by the necessity of designing algorithms allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Purpose: Development of algorithms and software system allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Experimental research to be carried out into the efficiency of developed algorithms and their comparison to their existing analogs. Findings: Viola-Jones object detection algorithm is suggested to detect the area of character location on images with complex background. The image normalization includes the following operations: boundary identification, Hough transformation, search for closed loops, image leveling. The structure of convolutional neural network (CNN) is suggested for characters recognition. The implementation of the developed algorithms is presented by characters recognition on images of vehicle number plates.",

keywords = "artificial neural networks, character recognition, Hough transform, Image processing, Viola-Jones algorithm",

author = "Cherneta, {D. S.} and Druki, {Alexey Alexeevich} and Spitsyn, {V. G.}",

year = "2016",

month = jun,

day = "14",

doi = "10.1109/SIBCON.2016.7491714",

language = "English",

booktitle = "2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2016 International Siberian Conference on Control and Communications, SIBCON 2016 ; Conference date: 12-05-2016 Through 14-05-2016",

}

TY - GEN

T1 - Development of multistage algorithm for text objects recognition in images

AU - Cherneta, D. S.

AU - Druki, Alexey Alexeevich

AU - Spitsyn, V. G.

PY - 2016/6/14

Y1 - 2016/6/14

N2 - The relevance of this study is stipulated by the necessity of designing algorithms allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Purpose: Development of algorithms and software system allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Experimental research to be carried out into the efficiency of developed algorithms and their comparison to their existing analogs. Findings: Viola-Jones object detection algorithm is suggested to detect the area of character location on images with complex background. The image normalization includes the following operations: boundary identification, Hough transformation, search for closed loops, image leveling. The structure of convolutional neural network (CNN) is suggested for characters recognition. The implementation of the developed algorithms is presented by characters recognition on images of vehicle number plates.

AB - The relevance of this study is stipulated by the necessity of designing algorithms allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Purpose: Development of algorithms and software system allowing to improve the efficiency of characters recognition on images with complex backgrounds subjected to noise, affine and projective transformations. Experimental research to be carried out into the efficiency of developed algorithms and their comparison to their existing analogs. Findings: Viola-Jones object detection algorithm is suggested to detect the area of character location on images with complex background. The image normalization includes the following operations: boundary identification, Hough transformation, search for closed loops, image leveling. The structure of convolutional neural network (CNN) is suggested for characters recognition. The implementation of the developed algorithms is presented by characters recognition on images of vehicle number plates.

KW - artificial neural networks

KW - character recognition

KW - Hough transform

KW - Image processing

KW - Viola-Jones algorithm

UR - http://www.scopus.com/inward/record.url?scp=84978060506&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978060506&partnerID=8YFLogxK

U2 - 10.1109/SIBCON.2016.7491714

DO - 10.1109/SIBCON.2016.7491714

M3 - Conference contribution

AN - SCOPUS:84978060506

BT - 2016 International Siberian Conference on Control and Communications, SIBCON 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2016 International Siberian Conference on Control and Communications, SIBCON 2016

Y2 - 12 May 2016 through 14 May 2016

ER -