Abstract

A document image is composed of a variety of physical entities or regions such as text blocks, lines, words, figures, tables, and background. We could also assign functional or logical labels such as sentences, titles, captions, author names, and addresses to some of these regions. The process of document structure and layout analysis tries to decompose a given document image into its component regions and understand their functional roles and relationships. The processing is carried out in multiple steps, such as preprocessing, page decomposition, structure understanding, etc. We will look into each of these steps in detail in the following sections. Document images are often generated from ph

Abstract

One of the prime challenges in designing a classifier for large-class problems such as Indian language OCRs is the presence of a large similar looking character set. The nature of the character set introduces problems with accuracy and efficiency of the classifier. Hierarchical classifiers such as Binary Hierarchical Decision Trees (BHDTs) using SVMs as component classifiers have been effectively used to tackle such large-class classification problems. The accuracy and efficiency of a BHDT classifier will depend on: i) the accuracy of the component classifiers, ii) the separability of the clusters at each node in a hierarchical classifier, and iii) the balance of the BHDT. We propose methods to tackle each of the above problems in the case of binary character images. We present a new distance measure, which is intuitively suitable when Support Vector Machines are used as component classifiers. We also propose a novel method for balancing the BHDT to improve its efficiency, while maintaining the accuracy. Finally we propose a method to generate overlapping partitions to improve the accuracy of BHDTs. Comparison of the method with other forms of classifier combination techniques such as 1vs1, 1vsRest and Decision Directed Acyclic Graphs shows that the proposed approach is highly efficient, while being comparable with the more expensive techniques in terms of accuracy. The experiments are focused on the problem of Indian language OCR, while the framework is usable for other problems as well

Abstract

Accurate registration of images is essential for many computer vision algorithms for medical image analysis, super-resolution, and image mosaicing. Performance of traditional correspondence-based approaches is restricted by the reliability of the feature detector. Popular frequency domain approaches use the magnitude of global frequencies for registration, and are limited in the class of transformations that can be estimated. We propose the use of local phase information for accurate image registration as it is robust to noise and illumination conditions and the estimates are obtained at sub-pixel accuracy without any correspondence computation. We form an overdetermined system of equations from the phase differences to estimate the parameters of image registration. We demonstrate the effectiveness of the approach for affine transformation under Gaussian white noise and varying illumination conditions.

Abstract

Forensic document verification presents a different and interesting set of challenges as opposed to traditional writer identification and verification tasks using natural handwriting. The handwritten data presented to a forensic examiner is often deliberately altered, in addition to being limited in quantity. Specifically, the alterations can be either forged, where one imitates another person’s handwriting; or repudiated, where one deliberately distorts his handwriting in order to avoid identification. In this paper, we present a framework to detect repudiation in forensic documents, where we only have one pair of documents to arrive at a decision. The approach generates a statistically significant confidence score from matching two documents, which can be used to screen the documents that are passed on to an expert examiner. The approach can be extended for detection of forgeries as well.

Abstract

Post-processors are critical to the performance of language recognizers like OCRs, speech recognizers, etc. Dictionary-based post-processing commonly employ either an algorithmic approach or a statistical approach. Other linguistic features are not exploited for this purpose. The language analysis is also largely limited to the prose form.This paper proposes a framework to use the rich metric and formal structure of classical poetic forms in Indian languages for post-processing a recognizer like an OCR engine. We show that the structure present in the form of the vrtta and prasa ¯ can be efficiently used to disambiguate some cases that may be difficult for an OCR. The approach is efficient, and complementary to other post-processing approaches and can be used in conjunction with them.

Abstract

Most of the state-of-the-art segmentation algorithms are de-signed to handle complex document layouts and backgrounds, while assuming a simple script structure such as in Roman script. They perform poorly when used with Indian languages, where the components are not strictly collinear. In this paper, we propose a document segmentation algorithm that can handle the complexity of Indian scripts in large document image collections. Segmentation is posed as a graph cut problem that incorporates the a prior information from script structure in the objective function of the cut. We show that this information can be learned automatically and be adapted within a collection of documents (a book)and across collections to achieve accurate segmentation. We show the results on Indian language documents in Telugu script. The approach is also applicable to other languages with complex scripts such as Bangla,Kannada, Malayalam, and Urdu

Abstract

Annotated datasets of handwriting are a prerequisite to attempt a variety of problems such as building recognizers, developing writer identification algorithms, etc. However, the annotation of large datasets is a tedious and expensive process, especially at the character or stroke level. In this paper we propose a novel, automated method for annotation at the character level, given a parallel corpus of online handwritten data and the corresponding text. The method employs a model-based handwriting synthesis unit to map the two corpora to the same space and the annotation is propagated to the word level and then to the individual characters using elastic matching. The initial results of annotation are used to improve the handwriting synthesis model for the user under consideration, which in turn refines the annotation. The method can take care of errors in the handwriting such as spurious and missing strokes or characters. The output is stored in the UPXInkML format

Abstract

Automatic identification of the author of a document has a variety of applications for both online and offline handwritten data such as facilitating the use of write dependent recognizers, verification of claimed identity for security, enabling personalized HCI and countering repudiations for legal purposes. Most of the existing writer identification techniques require the data to be from a specific text or a recognizer be available, which is not always feasible. Text-independent approaches often require large amount of data to be confident of good results. In this work, we propose a text-independent writer identification framework that uses a specified set of primitives of online handwritten data to ascertain the identity of the writer. The framework allows us to learn the properties of the script and the writers simultaneously and hence can be used with multiple languages or scripts. We demonstrate the applicability of our framework by choosing shapes of curves as primitives and show results on five different scripts and on different data sets.

Abstract

We present a document understanding system in which the arrangement of lines of text and block separators within a document are modeled by stochastic context free gram-mars. A grammar corresponds to a document genre; our system may be adapted to a new genre simply by replacing the input grammar. The system incorporates an optical character recognition system that outputs characters,their positions and font sizes. These features are combined to form a document representation of lines of text and separators. Lines of text are labeled as tokens using regular expression matching. The maximum likelihood parse of this stream of tokens and separators yields a functional labeling of the document lines. We describe business card and business letter applications.

Abstract

A hierarchical framework for document segmentation is proposed as an optimization problem. The model incorporates the dependencies between various levels of the hierarchy unlike traditional document segmentation algorithms.This framework is applied to learn the parameters of the document segmentation algorithm using optimization methods like gradient descent and Q-learning.The novelty of our approach lies in learning the segmentation parameters in the absence of ground truth.