Touching Characters Research Articles

The Trend of Segmentation for Arabic Handwritten Touching Characters

The paper is a comprehensive study of existing research trends in the sector of Arabic language, with a focus on state-of-the-art methods to illustrate the existing condition of various theory in that sector, with the goal of facilitating the adaptation and extension of prior ones into new systems and applications. In the Arabic alphabet, there are 28 letters. Depending on its place in the word, every Arabic letter has over one shape; a single character may have from one to four shapes. The Touching between character and the Overlapping occurred in the handwritten. Historical documents contained a massive knowledge and culture. There are many old books that need to be converted into readable format. Which would take a long time if humans converted it. However, the main problem is the lack of research in Arabic Handwritten especially for segmentation of touching characters. Thus, current trends of the segmentation techniques are investigated to identify the current state-of-the art of segmenting touching characters in other domains for constructing enhance techniques for Arabic touching characters. In this paper, it reviewed approaches for the segmentation of the touching characters. This paper presents the trend of approaches for the recognition process and segmentation of Arabic handwritten touching characters. In this paper, it highlighted the strength of each technique, the method used, and the drawback of the techniques. Based on the outcome, this will provide a good foundation for constructing a better technique for segmentation of Arabic touching characters, especially from the degraded documents.

Optical Character Recognition of Indian Language Manuscripts using Convolutional Neural Networks

India's constitution has 22 languages written in 17 different scripts. These materials have a limited lifespan, and as generations pass, these materials deteriorate, and the vital knowledge is lost. This work uses digital texts to convey information to future generations. Optical Character Recognition (OCR) helps extract information from scanned manuscripts (printed text). This paper proposes a simple and effective solution of optical character recognition (OCR) Sanskrit Character from text document images using long short-term memory (LSTM) and neural networks of Sanskrit Characters. Existing methods focuses only upon the single touching characters. But our main focus is to design a robust method using Bidirectional Long Short-Term Memory (BLSTM) architecture for overlapping lines, touching characters in middle and upper zone and half character which would increase the accuracy of the present OCR system for recognition of poorly maintained Sanskrit literature.

Tri-level handwritten text segmentation techniques for Gujarati language

Objectives: To improve the efficiency of tri-level segmentation tasks for handwritten Gujarati text. Methods: Using hybrid methods for tri-level segmentation, we have used line, word and character segmentation from the image. This study presents a segmentation paradigm that works with touching characters, slop of the line written on the page, character overlapping, etc. It evaluated on the dataset of 500+ images created by us on different writing sentences by different people. We have used the Horizontal projection technique for line segmentation, Scale-space technique for word segmentation and the Vertical projection technique for character segmentation. Findings: The experimental results show that the proposed method is more efficient for handwritten Gujarati text with diacritics. We have obtained the accuracy for character level segmentation is 82%, word-level is 90% and for the line-level segmentation is 87%. Novelty: We have designed a methodology to segment Gujarati handwritten text with diacritics at all three levels including characters, words and lines. Applications: We have proposed tri-level segmentation which is pre-processing task that can be used in any character recognition systems i.e. OCR. Keywords: Deep learning; trilevel segmentation; handwritten Gujarati text

Learning-Free Text Line Segmentation for Historical Handwritten Documents

We present a learning-free method for text line segmentation of historical handwritten document images. This method relies on automatic scale selection together with second derivative of anisotropic Gaussian filters to detect the blob lines that strike through the text lines. Detected blob lines guide an energy minimization procedure to extract the text lines. Historical handwritten documents contain noise, heterogeneous text line heights, skews and touching characters among text lines. Automatic scale selection allows for automatic adaption to the heterogeneous nature of handwritten text lines in case the character height range is correctly estimated. In the extraction phase, the method can accurately split the touching characters among the text lines. We provide results investigating various settings and compare the model with recent learning-free and learning-based methods on the cBAD competition dataset.

Arbitrary oriented multilingual text detection and segmentation using level set and Gaussian mixture model

A pioneer conceptual combination of level set method and Gaussian Mixture Model (GMM) is presented in the described multilingual, arbitrary-oriented character segmentation. The method is a serial accomplishment processes of Gaussian low pass filter, single level 2 Dimensional Discrete Wavelet Transform (2D DWT) for better feature extraction and implemented level set method, k-means clustering algorithm with GMM to achieve a veracious character segmentation results by detecting great measure of true text region of an image. The proposed method segments a character chiefly by distinguishing the touching character constituents and also deals discontinuities presence in a character by using Laplacian of Gaussian filter and morphological bridge function in a intellectual way. The exhibited method was explored on Multi-script Robust Reading Competition dataset and on our privately collected graphical and handwritten multilingual, arbitrarily-oriented text images. The suggested method is compared with the well known multilingual and arbitrarily-oriented charter segmentation methods. The described method attains better segmentation outcomes when compared to the familiar functioning methods. Hence, the suggested method is highly suitable to consider as an improved, standard and procedural technique.

Comparative Analysis of Segmentation and Recognition Techniques for Offline Handwritten Words

A Pre-processing is the initial and vital phase in optical character recognition is the Pre-processing. Segmentation deals with the extraction of individual component from a document image. Number of techniques like projection profile, connected components, gaps between characters/components is reported in the literature for component extraction followed by feature extraction and recognition of the individual component. These techniques gives good results if components are isolated but fails if components are touched, shadowed or skewed. A novel technique is required to address such issues to enhance the recognition rate. The problem of segmentation for Roman script cursive handwriting is addressed by various authors but not enough addressed for Indian script especially Devanagari script. This paper is a review which is confined to offline handwritten script domain. It attempt to review various techniques for character segmentation considering touching characters for offline handwritten words in Devanagari script and scripts sharing similar characteristics (like Bangla, Gurumukhi), database used and their accuracy reported in the literature.

A Comparative analysis for identification and classification of text segmentation challenges in Takri Script

Takri is an Indian regional class of scripts, used in hilly areas of north-west India which include Jammu and Kashmir (J & K), Himachal Pradesh (H.P.), Punjab and Uttarakhand. This script has immense variations; almost 13 identified in the whole region of North-west India. It has been observed that no work for text identification and recognition of Takri script has been done so far. Therefore, our work focuses on identifying and classifying the various challenges in the script based on comparative analysis of existing text segmentation approaches, as correct segmentation of text leads to more accurate machine recognition. As there were no metal fonts available for the script, it is required to collect the machine-printed form of data for solving the text identification problem in Takri script. The paper surveys for different text segmentation approaches and based on the structural properties of the script, shows an implementation of these on Takri data in three steps- Gurmukhi segmentation technique, Connected Component segmentation approach, and Gurmukhi touching characters segmentation approach. Results are analyzed for Segmentation Accuracy and Challenges are identified along with their statistical analysis. Further, the challenges identified as half- forms, numerous types of touching characters, overlapping bounding boxes, are classified. The effectiveness of these challenges was evaluated using Naive-Bayesian machine learning algorithm. The results showed 80% accuracy in text identification and classification of Takri script.

Optimal Character Segmentation for Touching Characters in Tamil Language Palm Leaf Manuscripts using Horver Method

An optimality of an automatic character recognition for Tamil palm leaf manuscripts can be achieved only by an efficient segmentation of touching characters. In this article, the touching characters are segmented as a single character to achieve an optimum solution by the recognizer in Optical Character Recognition (OCR). The proposed method provides a novelty in touching character segmentation of Tamil palm leaf manuscripts. An initial process of separation of background image and foreground characters is applied on the palm leaf images by filtering and removing unwanted pieces of characters by noise removal methods. The thickening process overcomes the difficulty of small breakages in the characters. The aspect ratio of the character image can be used to categorize the character such as single or multi touching. Single touching is divided by yet another ways such as horizontal or vertical touching. Finally, the proposed algorithm for Horizontal and Vertical character segmentation named as HorVer method is applied on the horizontally and vertically touching characters to segment as independent character. Experimental result produces 91% of an accuracy on segmenting the touching characters in Tamil palm leaf manuscript images collected from various resources and Tamil Heritage Foundation (THF). A novelty method can be achieved in Tamil touching character segmentation by the proposed algorithm.

Segmentation and Recognition for Historical Tibetan Document Images

As a shining pearl in traditional Tibetan culture, historical Tibetan documents have received extensive attention from historians, linguists and Buddhist scholars. These documents are converted into digital form using Tibetan document segmentation and recognition methods. The document digitization is of great significance for the research, protection and inheritance of Tibetan history. This paper proposes an overall segmentation and recognition framework for historical Tibetan document images. Firstly, the historical Tibetan document image is preprocessed to correct imbalanced illumination, tilt and noises, and is further transformed into the binarized image. Secondly, we propose a layout segmentation method based on block projection to segment Tibetan document images into texts, lines and frames. Thirdly, in order to solve the problems of touching strokes between text-lines and curvilinear text-lines, we present a text-line segmentation method based on graph model for historical Tibetan text-line segmentation. Lastly, we present a touching segmentation method to segment touching Tibetan character string, and then recognize Tibetan characters. Experimental results show our proposed methods on layout segmentation, text-line segmentation and touching character string segmentation, achieve the satisfactory performance. The proposed methods can also be applied to other fonts in Tibetan font family.

Drop flow method: an iterative algorithm for complete segmentation of Devanagari ancient manuscripts

One of the major challenges of ancient manuscripts recognition is character segmentation. Because of many distinct features of ancient documents (thick characters, overlapping and touching characters), character segmentation is a very difficult task. Devanagari ancient manuscripts consist of vowels, consonants, modifiers, conjuncts and compound characters. Using existing techniques, segmentation of overlapping and touching characters is problematic. In this paper, an iterative character segmentation algorithm is presented for ancient documents in Devanagari script. At the beginning, the lines are extracted from the ancient documents by dividing the document image into vertical stripes and then using piecewise horizontal projection profiles. After that, these lines are segmented into words using vertical projection profiles and finally, words are segmented in characters using an iterative algorithm. In each iteration, character segmentation is refined. In the present work, we have proposed a new algorithm with the name ‘Drop Flow Method’ to find the segmentation path between touching components. The proposed algorithm can segment touching characters and 96.0% accuracy has been achieved for complete segmentation of Devanagari ancient manuscripts.

Segmentation of Touching Arabic Characters in Handwritten Documents by Overlapping Set Theory and Contour Tracing

Segmentation of handwritten words into characters is one of the challenging problem in the field of OCR. In presence of touching characters, make this problem more difficult and challenging. There are many obstacles/challenges in segmentation of touching Arabic handwritten text. Although researches are busy in solving the problem of segmentation of these touching characters but still there exist unsolved problems of segmentation of touching offline Arabic handwritten characters. This is due to large variety of characters and their shapes. So in this research, a new method for segmentation of touching Arabic Handwritten character has been developed. The main idea of the proposed method is to segment the touching characters by identifying the touching point by overlapping set theory and ending points of the Arabic word by applying some standard morphology operation methods. After identifying all the points, segmentation method is applied to trace the boundaries of characters to separate these touching characters. Experiments were conducted on touching characters taken from different data sets. The results show the accuracy of the proposed method.

Junction Point Detection and Identification of Broken Character in Touching Arabic Handwritten Text using Overlapping Set Theory

Touching characters are formed when two or more characters share the same space with each other. Therefore, segmentation of these touching character is very challenging research topic especially for handwritten Arabic degraded documents. This is one of the key issue in recognition of the handwritten Arabic text. In order to make the recognition system more effective segmentation of these touching handwritten Arabic characters is considered to be very important research area. In this research, a new method is proposed, which is used to identify the junction or common point of Arabic touching word image by applying overlapping or intersection set theory operation, which will help to trace the correct boundary of the touching characters, identify the broken characters and also segmented these touching handwritten text in an efficient way. The proposed method has been evaluated on Arabic touching handwritten characters taken from handwritten datasets. The results show the efficiency of the proposed method. The proposed method is applicable to both degraded handwritten documents and printed documents.

Vibro-Perception of Optical Bio-Inspired Fiber-Skin.

In this research, based on the principle of optical interferometry, the Mach-Zehnder and Optical Phase-locked Loop (OPLL) vibro-perception systems of bio-inspired fiber-skin are designed to mimic the tactile perception of human skin. The fiber-skin is made of the optical fiber embedded in the silicone elastomer. The optical fiber is an instinctive and alternative sensor for tactile perception with high sensitivity and reliability, also low cost and susceptibility to the magnetic interference. The silicone elastomer serves as a substrate with high flexibility and biocompatibility, and the optical fiber core serves as the vibro-perception sensor to detect physical motions like tapping and sliding. According to the experimental results, the designed optical fiber-skin demonstrates the ability to detect the physical motions like tapping and sliding in both the Mach-Zehnder and OPLL vibro-perception systems. For direct contact condition, the OPLL vibro-perception system shows better performance compared with the Mach-Zehnder vibro-perception system. However, the Mach-Zehnder vibro-perception system is preferable to the OPLL system in the indirect contact experiment. In summary, the fiber-skin is validated to have light touch character and excellent repeatability, which is highly-suitable for skin-mimic sensing.

“Zaïre”, une anti-“Polyeucte”: Voltaire contre Corneille

Although the relation of Voltaire’s Zaire to Corneille’s Polyeucte is never clearly declared by the Enlightenment philosopher, it is real nonetheless. A comparison between the two tragedies focusing on the issue of religion and the relationship between Christianity and tragedy, shows to what extent Voltaire thought of Corneille while writing his tragedy and to what extent he wanted to write an anti-Polyeucte. Both tragedies are based on the religion shock but while Christianity is glorified by Corneille, it is put in doubt by Voltaire. Considering how the latter twists some essential aspects of Christianity and in particular, through the touching character of Zaire, his impressive criticism of a religion which causes misfortune to humanity, one could even say that Zaire is an anti-Christian piece.

Text line segmentation using a fully convolutional network in handwritten document images

Line detection in handwritten documents is an important problem for processing of scanned documents. While existing approaches mainly use hand-designed features or heuristic rules to estimate the location of text lines, the authors present a novel approach that trains a fully convolutional network (FCN) to predict text line structure in document images. A rough estimation of text line, or a line map, is obtained by using FCN, from which text strings that pass through characters in each text line are constructed. Finally, the touching characters should be separated and assigned to different text lines to complete the segmentation, for which line adjacency graph is used. Experimental results on ICDAR2013 Handwritten Segmentation Contest data set show high performance together with the robustness of the system with different types of languages and multi-skewed text lines.

A fuzzy approach to segment touching characters

Abstract Despite many years of research, correct and reliable segmentation of touching characters is still a hard task to solve. In the recent years, many methods and algorithms have been proposed; nevertheless the problem is still open. In this paper, we propose a novel method based on fuzzy logic that combines three different techniques to segment touching characters. These techniques have already been used in other studies but they have never been used all together. We propose a 3–input/1–output fuzzy inference system with fuzzy rules that are specifically optimized to segment touching Latin characters. The method is applicable to both printed and handwritten characters. We discuss the performances of our method by comparing it with state of the art. Results show that our method provide a better accuracy to segment characters even with noisy touching characters.

A Review on Character Segmentation of Touching and Half Character in Handwritten Hindi Text

Segmentation is one of the challenging phases of OCR in case of handwritten text. It has been observed that different languages have different nature of writing the text that causes a number of issues in segmentation aspect. Hindi is one of the official languages of India having the complex structural features. The major problem in handwritten character segmentation is to deal with the touching characters such as conjunct, half-characters, overlapping characters, highly skewed, uneven header lines found in writing pattern of well known Hindi language[]. Segmentation is done on the basis of observed structural properties examined from the writing style of different individuals. This paper reviews different techniques available to dissect the simple, touching, overlapping Hindi text. Also, it discusses the existing techniques with their merits and demerits. It is observed on the basis of related work, few techniques are available for touching and half characters. Hybrid approach including projection profile combined along with the structural property such as width, height, pixel intensity etc is giving quite promising results but still, there exist some challenges due to the constraints that are taken in stated algorithms which need concerned efforts and immense research to improve the segmentation level. Keywords: character segmentation, touching character, devnagari script, header line modifier, half character

Segmentation of Degraded Text using Dynamic Profile Projection in Handwritten Gurmukhi Script

Character segmentation is a vital area of research for optical character recognition. The process of OCR involves several steps to recognize the character. After converting the scanner's output as bitmap image, set its threshold value. The recogni tion phase's output is totally depends upon the segmentation phase. Touching and broken character affects the accuracy rate of the recognition phase badly. Existing methods concentrate only upon the single touching characters problem with fixed size and mostly works upon th e middle zone. Segmentation of the broken character is an uphill task. In this paper, we used proposed technique for identification and segmentation of multiple touching characters in handwritten Gurmukhi words. In proposed technique firstly scanner's output is saved as bitmap image and then set its threshold value 200. The bitmap image is segmented into individual characters, called segments after applying the dynamically projection technique. The name of developed method i.e. Dynamic profile projection technique means t hat it works for variable size. Thus, the new technique works upon the segmentation of broken, multiple touching characters of variable size including the three zones of the handwritten gurmukhi script and increase the accuracy for touching characters. Thus, after the implementation of this concept got encouraging results than the existing systems.

System recognizing Bahamian license plate with touching characters

Various methods are proposed for license plate recognition, but none of them are universal. Some common methods for license plate localization, character extraction, and recognition are analyzed. Then a system is proposed to recognize the Bahamian license plate with touching characters. A vertical edge-based method with a modified sliding window technique is used to locate the license plate, and a machine learning process is used to trim the region. The located license plate is rectified by using the minimum enclosing box and the stroke width value. Then the vertical projection and pairs of extreme points are combined to segment the characters. Finally, a deep learning method is used to recognize the characters. 2996 images are experimented on and the total recognition accuracy achieves 83.29%. Typical methods of each stage are implemented to compare with the proposed methods. In addition, the proposed system is experimented on a public dataset to show the generalization ability of the system. The experimental results show that the proposed system performs better than the other methods and is able to be used in a real-time application.

Preparation of a Dataset and Issues Related with Recognition of Optical Character in Assamese Script

According to the website ‘ethnologue.com’, which does a lot of survey and statistical analysis on languages, has mentioned that currently 7102 living languages are available on earth. Recent trend is that the number of living languages is always going down, which is becoming an alarming matter. An article published by UNESCO in 2009, says that most of the endangered languages belong to India. In this digital era, we can keep a language alive, if it can be highly used in computers; software applications with interface in regional language. In this context, researchers from this region are working for developing an Optical Character Recognition system that can digitize the optical image written in major North-East Indian language. As the characteristics of scripts vary from one another so are the challenges. Keeping in mind the need of the researcher, we have developed a novel offline dataset of Assamese Historical and Machine Printed as well as handwritten documents, which could be used for experimentation of various techniques for Assamese character recognition task. The dataset comprise of a variety of modern and old Assamese texts that are collected from a variety of sources, which can be broadly divided into Machine printed and Handwritten documents. Both good quality and degraded documents are available in the dataset. Many researchers are working for the development of an OCR system for Assamese script; however there are a lot of challenges that need to be addressed. Discussion of various issues related with degraded text, historical documents, handwritten Assamese text and machine printed texts with reference to the data sample available in the dataset are mentioned here. Problems related with segmentation of characters in touching characters, difficulty in determining compound character and touching character. Skewed document and how its variation makes line segmentation difficult. Heavily printed documents make feature extraction a complicated task. In the dataset we have pages with backside text visible, making the document a noisy one. Besides, all these inherent issues of character recognition, issues related with recognition of old Assamese script is also discussed in detail. This dataset will be of ample use and the issues we have discussed will certainly increase attraction of researchers working in this field. More research and innovation with digitization of Assamese documents, books and historical documents will definitely help sustainability of the language and the script as well.