Showing papers on "Intelligent word recognition published in 2012"

The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web]

Abstract: In this issue, “Best of the Web” presents the modified National Institute of Standards and Technology (MNIST) resources, consisting of a collection of handwritten digit images used extensively in optical character recognition and machine learning research.

Top-down and bottom-up cues for scene text recognition

Abstract: Scene text recognition has gained significant attention from the computer vision community in recent years. Recognizing such text is a challenging problem, even more so than the recognition of scanned documents. In this work, we focus on the problem of recognizing text extracted from street images. We present a framework that exploits both bottom-up and top-down cues. The bottom-up cues are derived from individual character detections from the image. We build a Conditional Random Field model on these detections to jointly model the strength of the detections and the interactions between them. We impose top-down cues obtained from a lexicon-based prior, i.e. language statistics, on the model. The optimal word represented by the text image is obtained by minimizing the energy function corresponding to the random field model. We show significant improvements in accuracies on two challenging public datasets, namely Street View Text (over 15%) and ICDAR 2003 (nearly 10%).

System and method of performing user-specific automatic speech recognition

Abstract: Speech recognition models are dynamically re-configurable based on user information, application information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. Word recognition lattices are generated for each data field of an application and dynamically concatenated into a single word recognition lattice. A language model is applied to the concatenated word recognition lattice to determine the relationships between the word recognition lattices and repeated until the generated word recognition lattices are acceptable or differ from a predetermined value only by a threshold amount. These techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.

The role of meaning in word recognition.

Abstract: The goal of the present discussion is to bring into focus a number of implicit assumptions in the area of word recognition research (see also Seidenberg, this volume). These assumptions revolve around what will here be referred to as the magic moment in word processing. The magic moment refers to that point in time where the subject has recognized the word but has yet to access meaning. Researchers have argued that they can both collect data and develop adequate models of this crucial point in word processing. The outline for this chapter is as follows: First, the empirical support for a magic moment is evaluated. The thrust of this discussion is that the major tasks used to provide data regarding the magic moment entail characteristics that question their utility as pure reflections of this crucial point in word processing. Second, an alternative framework is presented that emphasizes the functional utility of words in language processing, that is, to convey meaning. Third, empirical evidence is presented that suggests that meaning can contribute to components involved in early word processing. Finally, there is a brief discussion of how meaning might be incorporated into the current theoretical accounts of word processing.

KHATT: Arabic Offline Handwritten Text Database

Abstract: In this paper, we report our comprehensive Arabic offline Handwritten Text database (KHATT) after completion of the collection of 1000 handwritten forms written by 1000 writers from different countries. It is composed of an image database containing images of the written text at 200, 300, and 600 dpi resolutions, a manually verified ground truth database that contains meta-data describing the written text at the page, paragraph, and line levels. A formal verification procedure is implemented to align the handwritten text with its ground truth at the form, paragraph and line levels. Tools to extract paragraphs from pages and segment paragraphs into lines are developed. Preliminary experiments on Arabic handwritten text recognition are conducted using sample data from the database and the results are reported. The database will be made freely available to researchers world-wide for research in various handwritten-related problems such as text recognition, writer identification and verification, etc.

Offline recognition of handwritten Bangla characters: an efficient two-stage approach

Abstract: The present work deals with recognition of handwritten characters of Bangla, a major script of the Indian sub-continent The main contributions presented here are (a) generation of a database of handwritten basic characters of Bangla and (b) development of a handwritten character recognition scheme suitable for scripts like Bangla consisting of many similar shaped characters for the benchmark results The present database is a pioneering development in the context of recognition of off-line handwritten characters of this script It has 37,858 handwritten samples and accommodates a large spectrum of handwriting style by Bangla speaking population This database will be made available ( http://wwwisicalacin/~ujjwal/download/Banglabasiccharacterhtml ) free of cost to researchers for further studies Also, we identified two major factors affecting high recognition accuracies for the present character samples, namely, (a) erratic nature of the presence of headline (shapes of Bangla characters usually contain a horizontal line in its upper part) and (b) existence of several pairs of similar shaped characters The proposed recognition approach takes care of the above factors It identifies any confusion in the first stage classification between a pair of similar shaped character classes and resolves the same in the second stage classification by extracting a feature vector based on a non-uniform grid

Recognition of printed Devanagari text using BLSTM Neural Network

Abstract: In this paper, we propose a recognition scheme for the Indian script of Devanagari. Recognition accuracy of Devanagari script is not yet comparable to its Roman counterparts. This is mainly due to the complexity of the script, writing style etc. Our solution uses a Recurrent Neural Network known as Bidirectional LongShort Term Memory (BLSTM). Our approach does not require word to character segmentation, which is one of the most common reason for high word error rate. We report a reduction of more than 20% in word error rate and over 9% reduction in character error rate while comparing with the best available OCR system.

An automated vision system for container-code recognition

Abstract: Highlights? An automatic container-code recognition system is developed by using computer vision. ? The characteristics of characters are made full use of to locate container-code. ? A two-step method is proposed to segment characters for various imaging conditions. Automatic container-code recognition is of great importance to the modern container management system. Similar techniques have been proposed for vehicle license plate recognition in past decades. Compared with license plate recognition, automatic container-code recognition faces more challenges due to the severity of nonuniform illumination and invalidation of color information. In this paper, a computer vision based container-code recognition technique is proposed. The system consists of three function modules, namely location, isolation, and character recognition. In location module, we propose a text-line region location algorithm, which takes into account the characteristics of single character as well as the spatial relationship between successive characters. This module locates the text-line regions by using a horizontal high-pass filter and scanline analysis. To resolve nonuniform illumination, a two-step procedure is applied to segment container-code characters, and a projection process is adopted to isolate characters in the isolation module. In character recognition module, the character recognition is achieved by classifying the extracted features, which represent the character image, with trained support vector machines (SVMs). The experimental results demonstrate the efficiency and effectiveness of the proposed technique for practical usage.

A Survey of OCR Applications

Abstract: Optical Character Recognition or OCR is the electronic translation of handwritten, typewritten or printed text into machine translated images. It is widely used to recognize and search text from electronic documents or to publish the text on a website. The paper presents a survey of applications of OCR in different fields and further presents the experimentation for three important applications such as Captcha, Institutional Repository and Optical Music Character Recognition. We make use of an enhanced image segmentation algorithm based on histogram equalization using genetic algorithms for optical character recognition. The paper will act as a good literature survey for researchers starting to work in the field of optical character recognition.

Indian sign language recognition

Abstract: Understanding human motions can be posed as a pattern recognition problem. Applications of pattern recognition in information processing problems are diverse ranging from Speech, Handwritten character recognition to medical research and astronomy. Humans express time-varying motion patterns (gestures), such as a wave, in order to convey a message to a recipient. If a computer can detect and distinguish these human motion patterns, the desired message can be reconstructed, and the computer can respond appropriately. This paper represents a framework for a human computer interface capable of recognizing gestures from the Indian sign language. The complexity of Indian sign language recognition system increases due to the involvement of both the hands and also the overlapping of the hands. Alphabets and numbers have been recognized successfully. This system can be extended for words and sentences Recognition is done with PCA (Principal Component analysis). This paper also proposes recognition with neural networks. Further it is proposed that number of finger tips and the distance of fingertips from the centroid of the hand can be used along with PCA for robustness and efficient results.

HMM based Offline Handwritten Writer Independent English Character Recognition using Global and Local Feature Extraction

Abstract: rate of handwritten character is still limited around 90 percent due to the presence of large variation of shape, scale and format in hand written characters. A sophisticated hand written character recognition system demands a better feature extraction technique that would take care of such variation of hand writing. In this paper, we propose a recognition model based on multiple Hidden Markov Models (HMMs) followed by few novel feature extraction techniques for a single character to tackle its different writing formats. We also propose a post-processing block at the final stage to enhance the recognition rate further. We have created a data-base of 13000 samples collected from 100 writers written five times for each character. 2600 samples have been used to train HMM and the rest are used to test recognition model. Using our proposed recognition system we have achieved a good average recognition rate of 98.26 percent.

Optical character recognition using artificial neural network

Abstract: Optical character recognition refers to the process of translat ing images of hand-written, typewritten, or printed text into a format understood by machines for the purpose of editing, indexing/searching, and a reduction in storage size. Optical character recognition is the mechanical or electronic translation of images of handwritten, typewritten or printed text into machine-editable text. Artificial neural networks are commonly used to perform character recognition due to their high noise tolerance. In this paper, an Optical character recognition based on Artificial Neural Networks (ANNs). The ANN is trained using the Back Propagation algorithm.

Moment-Based Image Normalization for Handwritten Text Recognition

Abstract: In this paper, we extend the concept of moment-based normalization of images from digit recognition to the recognition of handwritten text. Image moments provide robust estimates for text characteristics such as size and position of words within an image. For handwriting recognition the normalization procedure is applied to image slices independently. Additionally, a novel moment-based algorithm for line-thickness normalization is presented. The proposed normalization methods are evaluated on the RIMES database of French handwriting and the IAM database of English handwriting. For RIMES we achieve an improvement from 16.7% word error rate to 13.4% and for IAM from 46.6% to 40.4%.

Combining Multi-scale Character Recognition and Linguistic Knowledge for Natural Scene Text OCR

Abstract: Understanding text captured in real-world scenes is a challenging problem in the field of visual pattern recognition and continues to generate a significant interest in the OCR (Optical Character Recognition) community. This paper proposes a novel method to recognize scene texts avoiding the conventional character segmentation step. The idea is to scan the text image with multi-scale windows and apply a robust recognition model, relying on a neural classification approach, to every window in order to recognize valid characters and identify non valid ones. Recognition results are represented as a graph model in order to determine the best sequence of characters. Some linguistic knowledge is also incorporated to remove errors due to recognition confusions. The designed method is evaluated on the ICDAR 2003 database of scene text images and outperforms state-of-the-art approaches.

Handwritten English Word Recognition Based on Convolutional Neural Networks

Abstract: This paper presents a novel segmentation-based and lexicon-driven handwritten English recognition systems. For the segmentation, a modified online segmentation method based on rules are applied. Then, convolutional neural networks are introduced for offline character recognition. Experiments are evaluated on UNIPEN lowercase data sets, with the word recognition rate of 92.20%.

Power-law transformation for enhanced recognition of born-digital word images

Abstract: In this paper, we discuss the issues related to word recognition in born-digital word images. We introduce a novel method of power-law transformation on the word image for binarization. We show the improvement in image binarization and the consequent increase in the recognition performance of OCR engine on the word image. The optimal value of gamma for a word image is automatically chosen by our algorithm with fixed stroke width threshold. We have exhaustively experimented our algorithm by varying the gamma and stroke width threshold value. By varying the gamma value, we found that our algorithm performed better than the results reported in the literature. On the ICDAR Robust Reading Systems Challenge-1: Word Recognition Task on born digital dataset, as compared to the recognition rate of 61.5% achieved by TH-OCR after suitable pre-processing by Yang et. al. and 63.4% by ABBYY Fine Reader (used as baseline by the competition organizers without any preprocessing), we achieved 82.9% using Omnipage OCR applied on the images after being processed by our algorithm.

Offline handwritten Gurmukhi character recognition: study of different feature-classifier combinations

Abstract: Offline handwritten character recognition (OHCR) is the method of converting handwritten text into machine processable layout Since late sixties, efforts have been made for offline handwritten character recognition throughout the world Principal Component Analysis (PCA) has also been used for extracting representative features for character recognition In order to assess the prominence of features in offline handwritten Gurmukhi character recognition, we have recognized offline handwritten Gurmukhi characters with different combinations of features and classifiers The recognition system first sets up a skeleton of the character so that significant feature information about the character can be extracted For the purpose of classification, we have used k-NN, Linear-SVM, Polynomial-SVM and RBF-SVM based approaches In present work, we have collected 7,000 samples of isolated offline handwritten Gurmukhi characters from 200 different writers The set of basic 35 akhars of Gurmukhi has been considered here A partitioning policy for selecting the training and testing patterns has also been experimented in present work We have used zoning feature; diagonal feature; directional feature; intersection and open end points feature; transition feature; parabola curve fitting based feature and power curve fitting based feature extraction technique in order to find the feature set for a given character The proposed system achieves a recognition accuracy of 948% when PCA is not applied and a recognition accuracy of 977% when PCA is applied

HMM Based Online Handwritten Bangla Character Recognition Using Dirichlet Distributions

Abstract: A reasonably large database of online handwritten Bangla characters has been developed. Such a handwritten character sample is composed of one or more strokes. Seventy five such stroke classes have been identified on the basis of the varying handwriting styles present in the character database. Each character sample is a sequence of strokes emanating from these stroke classes. Another database of handwritten Bangla strokes has been developed from the character database. This is the first such database for Bangla script. Certain stroke level features are defined on the basis of certain extremum points which represent the stroke shape reasonably well. The proposed character classification method is a two-stage approach. First, a probability distribution is estimated for each stroke class using the stroke features and then an HMM based character classifier is designed using each stroke class as a state. The parameters of both the stroke class distributions and the character class HMMs are estimated on the basis of the training set having 29,951 character samples. The character level recognition accuracy obtained by the proposed method on the test set having 8,616 samples, is 91.85%.

A neural network approach to online Devanagari handwritten character recognition

Abstract: This paper proposes a neural network based framework to classify online Devanagari characters into one of 46 characters in the alphabet set. The uniqueness of this work is three-fold: (1) The feature extraction is just the Discrete Cosine Transform of the temporal sequence of the character points (utilizing the nature of online data input). We show that if used right, a simple feature set yielded by the DCT can be very reliable for accurate recognition of handwriting, (2) The mode of character input is through a computer mouse, and (3) We have built the online handwritten database of Devanagari characters from scratch, and there are some unique features in the way we have built up the database. Lastly, the testing has been carried on 2760 characters, and recognition rates of up to 97.2% are achieved.

Stroke Segmentation and Recognition from Bangla Online Handwritten Text

Abstract: This paper deals with recognition of online handwritten Bangla (Bengali) text. Here, at first, we segment cursive words into strokes. A stroke may represent a character or a part of a character. We selected a set of Bangla words written by different groups of people such that they contain all basic characters, all vowel and consonant modifiers and almost all types of possible joining among them. For segmentation of text into strokes, we discovered some rules analyzing different joining patterns of Bangla characters. Combination of online and offline information was used for segmentation. We achieved correct segmentation rate of 97.89% on the dataset. We manually analyzed different strokes to create a ground truth set of distinct stroke classes for result verification and we obtained 85 stroke classes. Directional features were used in SVM for recognition and we achieved correct stroke recognition rate of 97.68%.

Scanning Neural Network for Text Line Recognition

Abstract: Optical character recognition (OCR) of machine printed Latin script documents is ubiquitously claimed as a solved problem. However, error free OCR of degraded or noisy text is still challenging for modern OCR systems. Most recent approaches perform segmentation based character recognition. This is tricky because segmentation of degraded text is itself problematic. This paper describes a segmentation free text line recognition approach using multi layer perceptron (MLP) and hidden markov models (HMMs). A line scanning neural network â€"trained with character level contextual information and a special garbage classâ€" is used to extract class probabilities at every pixel succession. The output of this scanning neural network is decoded by HMMs to provide character level recognition. In evaluations on a subset of UNLV-ISRI document collection, we achieve 98.4% character recognition accuracy that is statistically significantly better in comparison with character recognition accuracies obtained from state-of-the-art open source OCR systems.

Handwritten Character Recognition using Different Kernel based SVM Classifier and MLP Neural Network (A COMPARISON)

Abstract: Neural Networks and SVM are recently being used in various kind of pattern recognition. As humans, it is easy to recognize numbers, letters, voices, and objects, to name a few. However, making a machine solve these types of problems is a very difficult task .Character Recognition has been an active area of research in the field of image processing and pattern recognition and due to its diverse applicable environment, it continues to be a challenging research topic. It has numerous applications which include, reading aid for blind, bank cheques and conversion of any hand written document into structural text form. In this paper an attempt is made to recognized handwritten character using the multi layer feed forward back propagation neural network without feature extraction and SVM classifier. Character data is used for training the neural network and SVM. The trained network is used for classification and recognition. For the neural network, each character is resized into 70x50 pixels, which is directly subjected to training. That is, each resized character has 3500 pixels and these pixels are taken as features for training the neural network. For the SVM classifier recognition model is divided in two phases namely, training and testing phase. In the training phase 25 features are extracted from each character and these features are used to train the SVM. In the testing phase SVM classifier is used to recognize the characters. The results show that by applying the proposed system, we reached a high accuracy for the problem of handwritten character recognition.

Relative positioning of stroke-based clustering: a new approach to online handwritten devanagari character recognition

Abstract: In this paper, we propose a new scheme for Devanagari natural handwritten character recognition. It is primarily based on spatial similarity-based stroke clustering. A feature of a stroke consists of a string of pen-tip positions and directions at every pen-tip position along the trajectory. It uses the dynamic time warping algorithm to align handwritten strokes with stored stroke templates and determine their similarity. Experiments are carried out with the help of 25 native writers and a recognition rate of approximately 95% is achieved. Our recognizer is robust to a large range of writing style and handles variation in the number of strokes, their order, shapes and sizes and similarities among classes.

Handwritten and Machine Printed Text Separation in Document Images Using the Bag of Visual Words Paradigm

Abstract: In a number of types of documents, ranging from forms to archive documents and books with annotations, machine printed and handwritten text may be present in the same document image, giving rise to significant issues within a digitisation and recognition pipeline. It is therefore necessary to separate the two types of text before applying different recognition methodologies to each. In this paper, a new approach is proposed which strives towards identifying and separating handwritten from machine printed text using the Bag of Visual Words paradigm (BoVW). Initially, blocks of interest are detected in the document image. For each block, a descriptor is calculated based on the BoVW. The final characterization of the blocks as Handwritten, Machine Printed or Noise is made by a Support Vector Machine classifier. The promising performance of the proposed approach is shown by using a consistent evaluation methodology which couples meaningful measures along with a new dataset.

Off-line Nepali handwritten character recognition using Multilayer Perceptron and Radial Basis Function neural networks

Abstract: An off-line Nepali handwritten character recognition, based on the neural networks, is described in this paper. A good set of spatial features are extracted from character images. Accuracy and efficiency of Multilayer Perceptron (MLP) and Radial Basis Function (RBF) classifiers are analyzed. Recognition systems are tested with three datasets for Nepali handwritten numerals, vowels and consonants. The strength of this research is the efficient feature extraction and the comprehensive recognition techniques, due to which, the recognition accuracy of 94.44% is obtained for numeral dataset, 86.04% is obtained for vowel dataset and 80.25% is obtained for consonant dataset. In all cases, RBF based recognition system outperforms MLP based recognition system but RBF based recognition system takes little more time while training.

Region selection in handwritten character recognition using Artificial Bee Colony Optimization

Abstract: Detection of local regions with optimal discriminating information from a sample of handwritten character image is one of the most challenging tasks to the pattern recognition community. In order to identify such regions, the idea of Artificial Bee Colony Optimization has been utilized in the present work. The technique is evaluated to pin point the set of local regions offering optimal discriminating feature set for handwritten numeral and character recognition. Initially, 8 directional gradient features are extracted from every region of different levels of partitions created using a CG based Quad Tree partitioning approach. Then, using the present approach, at each level, sampling process is done based on support Vector Machine (SVM) in every single region. Applying the technique we have gained 33%, 14%, 9%, 19%interms of region reduction and 0.2%, 0.4%, 0%, 1.6% in terms of recognition for Arabic, Hindi, Telugu numerals and Bangla Basic character datasets respectively. Though the success rate has not improved significantly for all the datasets, sizable amount of reduction in regions has occurred for every dataset using the present technique. Thus the cost and time of feature extraction is reduced significantly without dropping the general recognition rate.

A Kinect-based handwritten digit recognition for TV remote controller

Abstract: Human-machine interaction is a popular research filed recently. Microsoft Kinect can provide us an economical way for tracking the skeleton of the human body. In this study, we present an efficient Kinect-based method for a potential application of TV remote controller. The proposed method consists of two parts: four direction (up/down/left/right) recognition and handwritten digit recognition. The developed recognitions include five main steps: initialization, tracking skeleton, judging the start condition, recording the path, and recognition. We use an intuitive way to start the event and achieve high accurate recognition for the users. We apply the least square method for the four direction recognition and a support vector machine library for the handwritten digit recognition. Experimental results show that the proposed system has great potential on future human-machine development.

Recognition Techniques for Online Arabic Handwriting Recognition Systems

Abstract: Online recognition of Arabic handwritten text has been an on-going research problem for many years. Generally, online text recognition field has been gaining more interest lately due to the increasing popularity of hand-held computers, digital notebooks and advanced cellular phones. However, different techniques have been used to build several online handwritten recognition systems for Arabic text, such as Neural Networks, Hidden Markov Model, Template Matching and others. Most of the researches on online text recognition have divided the recognition system into these three main phases which are preprocessing phase, feature extraction phase and recognition phase which considers as the most important phase and the heart of the whole system. This paper presents and compares techniques that have been used to recognize the Arabic handwriting scripts in online recognition systems. Those techniques attempt to recognize Arabic handwritten words, characters, digits or strokes. The structure and strategy of those reviewed techniques are explained in this article. The strengths and weaknesses of using these techniques will also be discussed.

Freeman code based online handwritten character recognition for malayalam using backpropagation neural networks

Abstract: Handwritten character recognition is conversion of handwritten text to machine readable and editable form. Online character recognition deals with live conversion of characters. Malayalam is a language spoken by millions of people in the state of Kerala and the union territories of Lakshadweep and Pondicherry in India. It is written mostly in clockwise direction and consists of loops and curves. The method aims at training a simple neural network with three layers using backpropagation algorithm. Freeman codes are used to represent each character as feature vector. These feature vectors act as inputs to the network during the training and testing phases of the neural network. The output is the character expressed in the Unicode format.