Document Image Analysis System Research Articles

Pashto script and graphics detection in camera captured Pashto document images using deep learning model.

Layout analysis is the main component of a typical Document Image Analysis (DIA) system and plays an important role in pre-processing. However, regarding the Pashto language, the document images have not been explored so far. This research, for the first time, examines Pashto text along with graphics and proposes a deep learning-based classifier that can detect Pashto text and graphics per document. Another notable contribution of this research is the creation of a real dataset, which contains more than 1,000 images of the Pashto documents captured by a camera. For this dataset, we applied the convolution neural network (CNN) following a deep learning technique. Our intended method is based on the development of the advanced and classical variant of Faster R-CNN called Single-Shot Detector (SSD). The evaluation was performed by examining the 300 images from the test set. Through this way, we achieved a mean average precision (mAP) of 84.90%.

Anisotropic diffusion with fuzzy-based source for binarization of degraded document images

Document image binarization plays a vital role in the document image analysis system; however, it remains challenging due to various degradations. In this paper, we propose an anisotropic diffusion model involving fuzzy-based source for binarizing degraded document images, in which the diffusion term is response for edge-preserving smoothing and the source term is used to group intensity values of foreground and background pixels into two dominant modes separated by zero. Specifically, a fuzzy classification function (FCF) is first introduced for vaguely separating foreground from background, which is defined in a local neighborhood of each point rather than in the entire image domain. Then, the fuzzy-based source is constructed by FCF and a speed restrictor, involving no threshold. In numerical aspects, we develop a parallel-serial algorithm by combining finite differencing and parallel/serial splitting methods in the literature. This algorithm is tested on seven publicly available datasets (DIBCO 2009 to 2014 and 2016) and compared with six PDE-based models and two variational models in terms of degraded document binarization. Experimental results illustrate that our model is very effective for binarization of degrade document images, and is superior to the compared models subjectively and objectively.

PHTI: Pashto Handwritten Text Imagebase for Deep Learning Applications

Document Image Analysis (DIA) is one of the research areas of Artificial Intelligence (AI) that converts document images into machine-readable codes. In DIA systems, Optical Character Recognition (OCR) plays a key role in digitizing document images. The output of an OCR system is further used in many applications including, Natural Language Processing (NLP), Sentiment Analysis, Speech Recognition, and Translation Services. However, standard datasets are an essential requirement for the development, evaluation and comparison of different text recognition techniques. Pashto is one of such low resource languages that lacks availability regarding standard dataset of handwritten text. This paper therefore, addresses the unavailability of standard dataset for the Pashto handwritten text by developing a dataset named Pashto Handwritten Text Imagebase (PHTI). The PHTI is created by collecting handwritten samples from diverse genre of the Pashto language including poetry, religion, short stories, articles, novels, sports, culture and news. The dataset consists of 4,000 scanned images, written by 400 writers including 200 males and 200 females. These 4,000 images are further segmented into 36,082 text-line images. Each text-line image is annotated/ transcribed with UTF-8 codecs. The dataset can be used for many deep learning-based applications including, text recognition, skew detection, gender classification and age-groups classification.

An Efficient Cooperative Smearing Technique for Degraded Historical Document Image Segmentation

Segmentation is one of the critical steps in historical document image analysis systems that determines the quality of the search, understanding, recognition and interpretation processes. It allows isolating the objects to be considered and separating the regions of interest (paragraphs, lines, words and characters) from other entities (figures, graphs, tables, etc.). This stage follows the thresholding, which aims to improve the quality of the document and to extract its background from its foreground, also for detecting and correcting the skew that leads to redress the document. Here, a hybrid method is proposed in order to locate words and characters in both handwritten and printed documents. Numerical results prove the robustness and the high precision of our approach applied on old degraded document images over four common datasets, in which the pair (Recall, Precision) reaches approximately 97.7% and 97.9%.

A novel variational model for noise robust document image binarization

Abstract Binarization is a fundamental problem in document image analysis systems. Different from current thresholding techniques, a novel variational model is proposed for noise robust document image binarization in this work, which is inspired by two classical variational models that have been successfully applied in image segmentation and denoising. In our variational model, the energy functional consists of three terms: data fidelity term, binary classification term and regularization term. As a result, the proposed model is capable of performing image binarization and suppressing noise simultaneously. Concretely, we firstly design a variational model for the original document image, the minimizer of which is our desired binarization result. Secondly, the gradient descent flow equation of our model is derived by means of variational principle. Lastly, a simple finite difference scheme, time forward and space center difference, is employed to solve the gradient descent flow equation. Extensive experiments are conducted on three types of document image datasets to validate our model qualitatively and quantitatively. The experiment results not only demonstrate the effectiveness of our approach, but also verify its noise and illumination robustness.

AN APPROACH FOR PIXEL BASED BINARIZATION OF GRAY IMAGES

This paper presents a hybrid approach for binarization as well as a review on Binarization of images. Binarization is a process by which an image is converted into bits. It is an important step in most document image analysis systems. An image is a set of pixels. Most of the binarization techniques associate a certain intensity value for each and every pixel. A gray image or gray-scale image is just an image which has each pixel of same intensity. That means there is not much difference in color or value information of pixels. Usually, in common terms, a picture in black and white is considered as gray image in which black has least intensity and white have highest. Keywords— Binarization, Images, Segmentation, Restoration, etc ----------------------------------------------------------------------***--------------------------------------------------------------------

Text Normalization Framework for Handwritten Cursive Languages by Detection and Straightness the Writing Baseline

Abstract The process of baseline detection has an important role in optical recognition systems and document image analysis systems. It is widely used in many various preprocessing stages as a text normalization including skew, slant and slop corrections, writing lines straightness and characters segmentation, as well as in feature extraction process. In this work, a new framework for baseline detection and straightness for cursive handwritten texts is proposed based on analysis and extraction the directions features from the subwords of the text skeleton. Arabic script is chosen as a case study since it is cursive and widely adopted in many languages all around the world such as Arabic, Jawi, Urdu and Persian. The experiments results on a popular Arabic dataset showed the efficiency of the proposed framework.

DEBORA: Digital AccEss to BOoks of the RenAissance

EBORA (Digital AccEss to BOoks of the RenAissance) is a multidisciplinary European project aiming at digitizing and thus making rare sixteenth century books more accessible. End-users, librarians, historians, researchers in book history and computer scientists participated in the development of remote and collaborative access to digitized Renaissance books, necessary because of the reduced accessibility to digital libraries in image mode through the Internet. The size of files for the storage of images, the lack of a standard file format exchange suitable for progressive transmission, and limited querying possibilities currently limit remote access to digital libraries. To improve accessibility, historical documents must be digitized and retro-converted to extract a detailed description of the image contents suited to users' needs. Specialists of the Renaissance have described the metadata generally required by end-users and the ideal functionalities of the digital library. The retro-conversion of historical documents is a complex process that includes image capture, metadata extraction, image storage and indexing, automatic conversion in a reusable electronic form, publication on the Internet, and data compression for faster remote access. The steps of this process cannot be developed independently. DEBORA proposes a global approach to retro-conversion from the digitization to the final functionalities of the digital library centered on users' needs. The retro-conversion process is mainly based on a document image analysis system that simultaneously extracts the metadata and compresses the images. We also propose a file format to describe compressed books as heterogeneous data (images/text/links/ annotation/physical layout and logical structure) suitable for progressive transmission, editing, and annotation. DEBORA is an exploratory project that aims at demonstrating the feasibility of the concepts by developing prototypes tested by end-users.

Using colour information to understand censorship cards of film archives

Many European film archives are involved in the digitization of 20th century historical paper documents. In the context of the IST project COLLATE three of them were interested in the semi-automatic annotation of censorship cards and their subsequent retrieval on the basis of both annotations and content. Processing censorship cards, which is the main subject of this paper, leads to a number of challenges for many document image analysis (DIA) systems. Problems arise due to the low layout quality and standard of such material, which introduces a considerable amount of noise in its description. The layout quality is often negatively affected by the presence of stamps, signatures, ink specks, manual annotations and so on that overlap those layout components involved in the understanding or annotation processes. In order to effectively reduce the presence and the effect of noise, we propose an improved version of the knowledge-based DIA system WISDOM++ allowing it to take full advantage of the use of colour information in all processing steps: namely, image segmentation, layout analysis, document image classification and understanding. Experiments have been conducted on a corpus of multi-format documents concerning rare historic film censorships provided by the three film archives involved in the COLLATE project.

Stochastic language models for style-directed layout analysis of document images

Image segmentation is an important component of any document image analysis system. While many segmentation algorithms exist in the literature, very few i) allow users to specify the physical style, and ii) incorporate user-specified style information into the algorithm's objective function that is to be minimized. We describe a segmentation algorithm that models a document's physical structure as a hierarchical structure where each node describes a region of the document using a stochastic regular grammar. The exact form of the hierarchy and the stochastic language is specified by the user, while the probabilities associated with the transitions are estimated from groundtruth data. We demonstrate the segmentation algorithm on images of bilingual dictionaries.

Extraction of newspaper headlines from microfilm for automatic indexing

This paper proposes a document image analysis system that extracts newspaper headlines from microfilm images with a view to providing automatic indexing for news articles in microfilm. A major challenge in achieving this is the poor image quality of microfilm as most images are usually inadequately illuminated and considerably dirty. To overcome the problem we propose a new effective method for separating characters from noisy background since conventional threshold selection techniques are inadequate to deal with this kind of image. A run length smoothing algorithm is then applied to the headline extraction. Experimental results confirm the validity of the approach.

Automated performance evaluation of document image analysis systems: Issues and practice

The performance of document image analysis systems is affected by a variety of variables that alter the quality of documents. Objective evaluation and characterization of systems usually require large quantities of test data, and it is important to automate evaluation processes. In this article, issues in designing tools for automated evaluation of document image analysis techniques and systems are discussed, and some examples are presented. © 1996 John Wiley & Sons, Inc.

Goal-directed evaluation of binarization methods

This paper presents a methodology for evaluation of low-level image analysis methods, using binarization (two-level thresholding) as an example. Binarization of scanned gray scale images is the first step in most document image analysis systems. Selection of an appropriate binarization method for an input image domain is a difficult problem. Typically, a human expert evaluates the binarized images according to his/her visual criteria. However, to conduct an objective evaluation, one needs to investigate how well the subsequent image analysis steps will perform on the binarized image. We call this approach goal-directed evaluation, and it can be used to evaluate other low-level image processing methods as well. Our evaluation of binarization methods is in the context of digit recognition, so we define the performance of the character recognition module as the objective measure. Eleven different locally adaptive binarization methods were evaluated, and Niblack's method gave the best performance.

A prototype document image analysis system for technical journals

Gobbledoc, a system providing remote access to stored documents, which is based on syntactic document analysis and optical character recognition (OCR), is discussed. In Gobbledoc, image processing, document analysis, and OCR operations take place in batch mode when the documents are acquired. The document image acquisition process and the knowledge base that must be entered into the system to process a family of page images are described. The process by which the X-Y tree data structure converts a 2-D page-segmentation problem into a series of 1-D string-parsing problems that can be tackled using conventional compiler tools is also described. Syntactic analysis is used in Gobbledoc to divide each page into labeled rectangular blocks. Blocks labeled text are converted by OCR to obtain a secondary (ASCII) document representation. Since such symbolic files are better suited for computerized search than for human access to the document content and because too many visual layout clues are lost in the OCR process (including some special characters), Gobbledoc preserves the original block images for human browsing. Storage, networking, and display issues specific to document images are also discussed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>