Various methods are analyzed that have been proposed to realize the core of character recognition in an optical character recognition system. For the purpose of classification, we have used knn, linearsvm, polynomial svm and rbfsvm based approaches. Introduction optical character recognition ocr is a technique that allows convertingthe printed text into an editable format in computer. The proposed work depends on the handwriting word level, and it does not need for character segmentation stage. A novel hierarchical technique for offline handwritten.
The increasing need of a handwritten character recognition system in the indian offices such as banks, post offices and so forth, has made it an imperative field of research. Handwritten gurmukhi character recognition using statistical. Support vector machines svms have successfully been used in recognizing printed characters. Word segmentation and character segmentation is used as segmentation stages. Svm classifiers concepts and applications to character.
Structural features generally provide better results for the handwritten symbol recognition. Offline handwritten gurmukhi character recognition proceeding of. Online and offline character recognition and offline character recognition further divided. Dwt2 has also been considered with three different types, namely, haar wavelet, daubechies db 1. This paper represent handwritten gurmukhi feature recognition. Performance evaluation of classifiers for the recognition. Pdf offline handwritten gurmukhi character recognition using. Pdf offline handwritten character recognition techniques. Then feature extraction and recognition process is carried over the binary image. Mainly, character recognition machine will takes the raw data that for further implements. Offline handwritten gurmukhi character and numeral recognition.
For the purpose of classification, we have used knn, linearsvm, polynomialsvm and rbfsvm based approaches. An analytical study of handwritten character recognition. The recognition rate obtained in the case of online handwritten assamese numerals is higher than the recognition rate of 96. Handwritten digit recognition using support vector machine. A scheme for offline handwritten gurmukhi character recognition based on svms is presented in this paper. Performance analysis of zone based features for online. The system first prepares a skeleton of the character, so that feature information about the character is. An arabic handwriting dataset ahdb, dataset used for train and test the proposed system. Machine svm based classification method on khmer printed character set recognition pcr in bitmap document. In this paper, we deal with weka based classification methods for offline handwritten gurmukhi character recognition. Recognition of handwritten characters is a difficult task owing to various writing styles of individuals. Here two sets of features based on gradient and curvature of character image are computed. These techniques require good quality features as their input for the recognition process.
Handwritten character recognition is mainly of two types online and offline. Handwriting recognition is a technique that convert handwritten characters into machine processable formats. Moment invariant and affine moment invariant techniques are used as feature extraction. Handwriting word recognition based on svm classifier. A novel feature extraction technique for offline handwritten.
Khmer language has been identified as one of the most complex language with the total of 74 alphabets and the wording compound can has up to 5 vertical levels. Thus, this study provides a benchmark of online and offline handwritten chinese character recognition on the new standard datasets. Offline handwritten character recognition ohcr is the method of converting handwritten text into machine processable layout. Efficient feature extraction techniques for offline. Then the svm is used to estimate global correlations and classify the pattern. On the basis of data acquisition process, character recognition system can be classified into following categories. Pdf pca based offline handwritten gurmukhicharacter. This paper deals with the offline recognition of handwritten gurmukhi characters. Since late sixties, efforts have been made for offline handwritten character recognition throughout the world. Machine svm based classification method on khmer printed characterset recognition pcr in bitmap document. In our work we have considered 35 basic characters of gurmukhi script all assumed to be isolated and bearing header lines on. Indian character recognition and many recognition systems for.
This paper represent handwritten gurmukhi character recognition system using some statistical features like zone density, projection histograms, 8 directional zone density features in combination with some geometric features like area, perimeter, eccentricity, etc. We have also extended the work by applying the same methodology to recognize handwritten gurmukhi numerals. We only consider isolated handwritten chinese character recognition in this study since it is still an unsolved problem, while the handwritten text recognition will be considered indepth in other works. Offline character recognition is a more challenging and difficult task as there. The system first prepares a skeleton of the character. The character set of traditional handwritten gurmukhi script documents contains symbols that are not used in modern gurmukhi script. Handwritten character recognition has applications in postal code recognition, automatic data entry into large administrative systems, banking, digital libraries and invoice and receipt processing. Svm based offline handwritten gurmukhi character recognition munish kumar1, m. Optical character recognition, support vector machine, artificial neural network 1. In handwritten recognition, svm gives a better recognition result. In this paper, we have presented an offline handwritten gurmukhi character recognition system using various transformations techniques, namely, discrete wavelet transformations dwt2, discrete cosine transformations dct2, fast fourier transformations and fan beam transformations. Online and offline handwritten chinese character recognition. Handwritten gurmukhi numeral recognition using zone.
Principal component analysis pca has also been used for extracting representative features for character recognition. As shown in table 1, the maximum accuracy obtained in case of handwritten devanagari numerical recognition 22 is 95. Variations in handwriting are one prominent problem and achieving high degree of accuracy is a tedious task. Abed 8 presented an overview on handwritten character. Pdf offline handwritten gurmukhi character recognition. The aim of this paper is to develop an approach which improve the efficiency of handwritten recognition using artificial neural network. In this work, we have used support vector machine svm classifier for classification, as well as for linear kernel and polynomial kernel. Kumar et al pcabased offline handwritten character recognition system 348 in this phase, the graylevel character image is nor malized into a window sized 1 00. In case of offline character recognition, the typed handwritten characters are scanned and then converted into binary or gray scale image. Handwriting recognition is in research for over four decades and has attracted many researchers across the world. Optical character recognition, particle swarm optimization, handwriting recognition, gurmukhi characters, artificial neural network, handwritten character recognition. Jul 18, 2014 these techniques require good quality features as their input for the recognition process. In this paper, we have proposed two different feature extraction techniques, namely, parabola curve fitting based features and power curve fitting based features for. Handwritten character recognition, feature extraction, diagonal features, intersection and open end points features, svm.
Offline handwritten gurmukhi character recognition. Nov 18, 2014 the increasing need of a handwritten character recognition system in the indian offices such as banks, post offices and so forth, has made it an imperative field of research. Isolated curved gurmukhi character recognition using. There are lots of touching characters in a single word.
Pcabased offline handwritten character recognition system. In online handwriting recognition, data is captured during the writing process with the help of a special pen on electronic surface. Recognition of isolated handwritten characters in gurmukhi. Offline handwritten gurmukhi character recognition using particle swarm optimized neural network. Devanagari and gurmukhi script recognition in the context. The variability of writing styles, both between different. Table 1 indicated the summarized results obtained so far in handwritten devanagari character recognition. Based on the learning adaptability and capability to solve complex computations, classifiers are always the best suited for the pattern recognition problems. Gurmukhi printed character recognition using hierarchical.
In this paper, we have proposed two different feature extraction techniques, namely, parabola curve fitting based features and power curve fitting based features for offline handwritten gurmukhi character recognition. The offline handwritten character recognition is the frontier. Handwritten character recognition has been broadly classified in to two types. Pdf k nearest neighbour based offline handwritten gurmukhi. Recognition of handwritten characters is a difficult. Support vector machine svm based classifier for khmer. Handwritten gurmukhi numeral recognition using zonebased. Pdf svm based offline handwritten gurmukhi character. This paper may give the significance of an offline handwritten character recognition system in various applications, and may help to give different.
Offline handwritten presegmented character recognition of. The knearest neighbor, support vector machine and probalistic neural network are used as classifier. Svm based offline handwritten gurmukhi character recognition. This paper proposes one new method, svm for khmer character classification. In that work, they performed recognition without using pca and used only an svm classifier for classification purpose. A little more detailed survey on gurmukhi recognition is presented in 6 and 19. Character recognition is a process of conversion of an image of a handwritten or printed text in to a computer editable format. The system first prepares a skeleton of the character, so that feature information about the character is extracted.
As a result of advances in optical character recognition research, several techniques for handwritten character recognition have surfaced. In this paper, we deal with wekabased classification methods for offline handwritten gurmukhi character recognition. For the purpose of training and testing data set, we have collected around 10,500 samples of isolated offline handwritten gurmukhi characters. Gurmukhi is the script of punjabi language which is widely spokenacross the globe. A dataset of online handwritten assamese characters. Recognition of isolated handwritten characters is the process. Role of offline handwritten character recognition system. This paper presents a comparative study of various classifiers and the results achieved for offline handwritten. A novel feature extraction technique is presented in this paper for an offline handwritten gurmukhi character recognition system. Offline handwritten gurmukhi character recognition using. Handwritten character recognition is a complex task because of various writing styles of different individuals. Combination of different feature sets and svm classifier. A robust feature set of 105 feature elements is proposed under this work for. Pdf online handwritten gurmukhi character recognition.
The main goal of this thesis is to develop an online handwritten gurmukhi character recognition system. In the present work, we have used this classification technique to recognize handwritten characters. In offline handwriting recognition, prewritten data generally written on a sheet of paper is scanned. Support vector machine svm is an alternative to nn. Sometimes more than two characters touch each other, making the algorithm process more complicated. The languages on which massive work has been done are biographical notes. Sharma3 1assistant professor, computer science department, ggs college for women, chandigarh, india 2associate professor, department of computer science and applications, panjab university regional centre, muktsar, india.
Benchmark datasets for offline handwritten gurmukhi script. The system first prepares a skeleton of the character, so that feature information about. In this thesis work we have proposed offline recognition of isolated handwritten characters of gurmukhi script. This paper described seven applications based on offline handwritten characters recognition system. For offline handwritten gurmukhi character recognition two approaches are reported. Pdf handwritten digit recognition using support vector. International journal of information technology and computer science 6 2, 2014. Offline handwritten gurmukhi word recognition using deep. The extracted features were then fused together to. In case of offline character recognition, the typedhandwritten characters are scanned and then converted into binary or gray scale image. A offline handwritten gurmukhi character recognition based on k recognition.
The recognition of handwriting can, however, still is considered an open research problem due to its substantial variation in appearance. The preprocessing stage reduces noise and distortion, removes skewness and performs skeletonization of the image. A brief outline of each chapter is given in the following paragraphs. Pdf weka based offline handwritten gurmukhi character. A scheme for offline handwritten gurmukhi character recognition based on svms is presented by munish, jindal and sharma 2011. Anoop rekha 4 has presented a complete survey on different feature sets and classifiers used in offline handwritten gurumukhi character and numeral recognition. In present paper, authors have presented a novel hierarchical technique for isolated offline handwritten gurmukhi character recognition. Comparison between neural network and support vector.
1253 1487 1044 476 476 424 392 664 257 890 1060 1363 1399 116 949 1247 40 1302 764 92 510 1482 325 1100 1374 198 1170 900 492 251 814 1155 605 643 536 1427 1041 530 1453 928 472