Skew detection in handwritten scripts

This paper is devoted to the study of supervised learning methods as part of pattern recognition and especially the Amazigh Characters Recognition. The goal is to compare a partial list of the popular automatic classification methods, and test the performance of the proposed features set extracted from isolated characters using statistical methods with these different classifiers. In Experimental evaluation, several runs have been conducted for the different algorithms and the best accuracy observed is for the multilayer perceptron with a recognition rate about 96,47% which is very satisfactory.

show abstract

“…Several methods are available in [14]. The two most popular are the Hough transform and histograms projection.…”

Section: B Skew Correctionmentioning

confidence: 99%

A comparison of supervised classification methods for a statistical set of features: Application: Amazigh OCR

Aharrane¹,

Moutaouakil²,

Satori³

2015

2015 Intelligent Systems and Computer Vision (ISCV)

View full text Add to dashboard Cite

show abstract

“…The few proposed methods for handwritten or mixed (i.e., handwritten and printed) documents concern specific applications [12]. Chin [13] analyses this problem and emphasizes that the methods which handle printed pages successfully do not successfully manage handwritten pages very well or even if they do the computational cost is great. Our approach concerns general-purpose applications and can handle handwritten, printed or mixed documents.…”

Section: System Presentationmentioning

confidence: 99%

An unconstrained handwriting recognition system

Kavallieratou¹,

Fakotakis²,

Kokkinakis

2002

International Journal on Document Analysis and Recognition

View full text Add to dashboard Cite

Abstract. In this paper, an integrated offline recognition system for unconstrained handwriting is presented. The proposed system consists of seven main modules: skew angle estimation and correction, printedhandwritten text discrimination, line segmentation, slant removing, word segmentation, and character segmentation and recognition, stemming from the implementation of already existing algorithms as well as novel algorithms. This system has been tested on the NIST, IAM-DB, and GRUHD databases and has achieved accuracy that varies from 65.6% to 100% depending on the database and the experiment.

show abstract

“…In general, projection profile methods are limited to estimate skew angle within ±10 degrees [1]. Many methods are based on Hough Transform [3] [4]. Hough Transform is applied to find the best direction of the straight lines in documents.…”

Section: Introductionmentioning

confidence: 99%

A Novel Approach of Skew Normalization for Handwritten Text Lines and Words

Nain

2012

2012 Eighth International Conference on Signal Image Technology and Internet Based Systems

View full text Add to dashboard Cite

Handwritten text recognition is an active research area for many years. Handwritten text recognition needs to perform some preprocessing steps for better recognition. Initially, we find binary image of given handwritten text document and then after performing the line segmentation task on handwritten text document, we have to normalize to the segmented lines. There are various normalization task we have to perform as skew normalization, slant normalization and size normalization. This paper, focuses on the skew normalization of handwritten text lines and we propose a new skew normalization approach which is based on orthogonal projection of the segmented line with respect to x-axis. The algorithm detects the exact skew angle, and corrects it efficiently. The method has been experimented on various text document images and achieves more than 98% accuracy. A comparative study has been reported to provide a detailed analysis of the proposed method together with some other existing methods in the literature.

show abstract

Skew detection in handwritten scripts

Cited by 20 publications

References 5 publications

A comparison of supervised classification methods for a statistical set of features: Application: Amazigh OCR

A comparison of supervised classification methods for a statistical set of features: Application: Amazigh OCR

An unconstrained handwriting recognition system

A Novel Approach of Skew Normalization for Handwritten Text Lines and Words

Contact Info

Product

Resources

About