Alireza Alaei

ICDAR 2009 Handwriting Segmentation Contest

This paper presents the results of the Handwriting Segmentation Contest that was organized in the context of the ICDAR2013. The general objective of the contest was to use well established evaluation practices and procedures to record recent advances in off-line handwriting segmentation. Two benchmarking datasets, one for text line and one for word segmentation, were created in order to test and compare all submitted algorithms as well as some state-of-the-art methods for handwritten document image segmentation in realistic circumstances. Handwritten document images were produced by many writers in two Latin based languages (English and Greek) and in one Indian language (Bangla, the second most popular language in India). These images were manually annotated in order to produce the ground truth which corresponds to the correct text line and word segmentation results. The datasets of previously organized contests (ICDAR2007, ICDAR2009 and ICFHR2010 Handwriting Segmentation Contests) along with a dataset of Bangla document images were used as training dataset. Eleven methods are submitted in this competition. A brief description of the submitted algorithms, the evaluation criteria and the segmentation results obtained from the submitted methods are also provided in this manuscript.

A new scheme for unconstrained handwritten text-line segmentation

Variations in inter-line gaps and skewed or curled text-lines are some of the challenging issues in segmentation of handwritten text-lines. Moreover, overlapping and touching text-lines that frequently appear in unconstrained handwritten text documents significantly increase segmentation complexities. In this paper, we propose a novel approach for unconstrained handwritten text-line segmentation. A new painting technique is employed to smear the foreground portion of the document image. The painting technique enhances the separability between the foreground and background portions enabling easy detection of text-lines. A dilation operation is employed on the foreground portion of the painted image to obtain a single component for each text-line. Thinning of the background portion of the dilated image and subsequently some trimming operations are performed to obtain a number of separating lines, called candidate line separators. By using the starting and ending points of the candidate line separators and analyzing the distances among them, related candidate line separators are connected to obtain segmented text-lines. Furthermore, the problems of overlapping and touching components are addressed using some novel techniques. We tested the proposed scheme on text-pages of English, French, German, Greek, Persian, Oriya and Bangla and remarkable results were obtained.

Using Modified Contour Features and SVM Based Classifier for the Recognition of Persian/Arabic Handwritten Numerals

In this paper, we propose a robust and efficient feature set based on modified contour chain code to achieve higher recognition accuracy of Persian/Arabic numerals. In classification part, we employ support vector machine (SVM) as classifier. Feature set consists of 196 dimensions, which are the chain-code direction frequencies in the contour pixels of input image. We evaluated our scheme on 80,000 handwritten samples of Persian numerals. Using 60,000 samples for training, we tested our scheme on other 20,000 samples and obtained 98.71% correct recognition rate. Further, we obtained 99.37% accuracy using five-fold cross validation technique on 80,000 dataset.

A Benchmark Kannada Handwritten Document Dataset and Its Segmentation

Research towards Indian handwritten document analysis achieved increasing attention in recent years. In pattern recognition and especially in handwritten document recognition, standard databases play vital roles for evaluating performances of algorithms and comparing results obtained by different groups of researchers. For Indian languages, there is a lack of standard database of handwritten texts to evaluate performance of different document recognition approaches and for comparison purpose. In this paper, an unconstrained Kannada handwritten text database (KHTD) is introduced. The KHTD contains 204 handwritten documents of four different categories written by 51 native speakers of Kannada. Total number of text-lines and words in the dataset are 4298 and 26115, respectively. In most of text-pages of the KHTD contains either an overlapping or a touching text-lines and the average number of text-lines in each document on the database is 21. Two types of ground truths based on pixels information and content information are generated for the database. Providing these two types of ground truths for the KHTD, it can be utilized in many areas of document image processing such as sentence recognition/understanding, text-line segmentation, word segmentation, word recognition, and character segmentation. To provide a framework for other researches, recent text-line segmentation results on this dataset are also reported. The KHTD is available for research purposes.

Fine Classification of Unconstrained Handwritten Persian/Arabic Numerals by Removing Confusion amongst Similar Classes

In this paper, we propose two types of feature sets based on modified chain-code direction frequencies in the contour pixels of input image and modified transition features (horizontally and vertically). A multi-level support vector machine (SVM) is proposed as classifier to recognize Persian isolated digits. In first level, we combine similar shaped numerals into a single group and as result; we obtain 7 classes instead of 10 classes. We compute 196-dimension chain-code direction frequencies as features to discriminate 7 classes. In the second level, classes containing more than one numeral because of high resemblance in their shapes are considered. We use modified transition features (horizontally and vertically) for discriminating between two overlapping classes (0 and 1). To separate another overlapping group containing three numerals 2, 3 and 4 we first eliminate common parts of these digits (tail) and then compute chain code features. We employ SVM classifier for the classification and evaluate our scheme on 80,000 handwritten samples of Persian numerals [10]. Using 60,000 samples for training, we tested our scheme on other 20,000 samples and obtained 99.02% accuracy.

Piece-wise painting technique for line segmentation of unconstrained handwritten text: a specific study with Persian text documents

The most important and difficult task in text document analysis is to achieve line segmentation accurately, particularly when the document is composed of unconstrained handwritten text. To accomplish this objective a painting scheme is proposed in this research work. Being motivated by the fact that the handwritten Persian texts offer the most critical challenges in the process of text-line segmentation, the new method has been devised by studying the cursive Persian text scripts extensively; yet, in general the proposed line segmentation algorithm is applicable to handwritten text in any language/script. The text block is vertically decomposed into parallel pipe structures called as strip. Each row in each strip is painted by a gray intensity, which is the average intensity value of gray values of all pixels present in that row-strip. Subsequently, the painted pipes are converted into two-tone painting and it is smoothed. The white/black spaces in each pipe of the smoothed image are analyzed to get a short line of separation, phrased as Piece-wise Potential Separating Line (PPSL), between two consecutive black spaces. The PPSLs are concatenated to produce the segmentation of text lines. Some additional procedures are built to handle certain anomalies, which may occur. The scheme is validated by extensive experimentation. We tested the proposed algorithm with 52 pages of Persian text documents containing totally 823 lines and correct line segmentation of 92.35% is achieved. Moreover, the proposed algorithm was also tested with two different datasets of 152 and 200 handwritten text-pages of different languages. Efficiency and script independency of the proposed algorithm were proved when compared with various approaches presented in recent literature.

Word-Wise Handwritten Persian and Roman Script Identification

Most of the countries use bi-script documents. This is because every country uses its own national language and English as second/foreign language. Therefore, bi-lingual document with one language being the English and other being the national language is very common. Postal documents are a very good example of such bi-lingual/script document. This paper deals with word-wise handwritten script identification from bi-script documents written in Persian and Roman. In the proposed scheme, simple but fast computable set of 12 features based on fractal dimension, position of small component, topology etc. are used and a set of classifiers are employed for script identification experiments. We tested our scheme on a dataset of 5000 handwritten Persian and English words and 99.20% of correct script identification is obtained.

A New Two-Stage Scheme for the Recognition of Persian Handwritten Characters

In this paper, a two-stage scheme for the recognition of Persian handwritten isolated characters is proposed. In the first stage, similar shaped characters are categorized into groups and as a result, 8 groups are obtained from 32 Persian basic characters. In the second stage, the groups containing more than one similar shape characters are considered further for the final recognition. Feature extraction is based on under sampled bitmaps technique and modified chain-code direction frequencies. For the first stage features, we compute 49-dimension features based on under sampled bitmaps from 49 non-overlapping 7×7 window-maps. 196-dimension chain-code direction frequencies from 49 overlapping 9×9 window-maps are computed and used as features for the second stage of the proposed scheme. Classifiers are one-against-other support vector machines (SVM). We evaluated our scheme on a standard dataset of Persian handwritten characters. Using 36682 samples for training, we tested our scheme on other 15338 samples and obtained 98.10% and 96.68% correct recognition rates when considered 8-class and 32-class problems, respectively.

DATASET AND GROUND TRUTH FOR HANDWRITTEN TEXT IN FOUR DIFFERENT SCRIPTS

In document image analysis (DIA) especially in handwritten document recognition, standard databases play significant roles for evaluating performances of algorithms and comparing results obtained by different groups of researchers. The field of DIA regard to Indo-Persian documents is still at its infancy compared to Latin script-based documents; as such standard datasets are not still available in literature. This paper is an effort towards alleviating this gap. In this paper, an unconstrained handwritten dataset containing documents of Persian, Bangla, Oriya and Kannada (PBOK) is introduced. The PBOK contains 707 text-pages written in four different languages (Persian, Bangla, Oriya and Kannada) by 436 individuals. Total number of text-lines, words/subwords and characters are 12,565, 104,541 and 423,980, respectively. In most documents of PBOK dataset contain either an overlapping or a touching text-lines. The average number of text-lines in text-pages of the PBOK dataset is 18. Two types of ground truths, based on pixels information and content information, are generated for the dataset. Because of such ground truths, the PBOK dataset can be utilized in many areas of document image processing e.g. text-line segmentation, word segmentation and word recognition. To provide an insight for other researches, recent text-line segmentation results on this dataset are also reported.

A Painting Based Technique for Skew Estimation of Scanned Documents

In this paper, we propose an efficient skew estimation technique based on Piece-wise Painting Algorithm (PPA) for scanned documents. Here we, at first, employ the PPA on the document image horizontally and vertically. Applying the PPA on both the directions, two painted images (one for horizontally painted and other for vertically painted) are obtained. Next, based on statistical analysis some regions with specific height (width) from horizontally (vertically) painted images are selected and top (left), middle (middle) and bottom (right) points of such selected regions are categorized in 6 separate lists. Utilizing linear regression, a few lines are drawn using the lists of points. A new majority voting approach is also proposed to find the best-fit line amongst all the lines. The skew angle of the document image is estimated from the slope of the best-fit line. The proposed technique was tested extensively on a dataset containing various categories of documents. Experimental results showed that the proposed technique achieved more accurate results than the state-of-the-art methodologies.