OCR error correction using correction patterns and self-organizing migrating algorithm

Optical character recognition (OCR) systems help to digitize paper-based historical achieves. However, poor quality of scanned documents and limitations of text recognition techniques result in different kinds of errors in OCR outputs. Post-processing is an essential step in improving the output quality of OCR systems by detecting and cleaning the errors. In this paper, we present an automatic model consisting of both error detection and error correction phases for OCR post-processing. We propose a novel approach of OCR post-processing error correction using correction pattern edits and evolutionary algorithm which has been mainly used for solving optimization problems. Our model adopts a variant of the self-organizing migrating algorithm along with a fitness function based on modifications of important linguistic features. We illustrate how to construct the table of correction pattern edits involving all types of edit operations and being directly learned from the training dataset. Through efficient settings of the algorithm parameters, our model can be performed with high-quality candidate generation and error correction. The experimental results show that our proposed approach outperforms various baseline approaches as evaluated on the benchmark dataset of ICDAR 2017 Post-OCR text correction competition.

     

一种stroke滤波器文字分割算法

为解决复杂背景中准确地进行文字分割的问题,提出了一种应用stroke滤波器进行文本分割的新方法。首先进行stroke滤波器的合理设计,并应用所设计的stroke滤波器来判别文本的彩色极性,得到初次分割的二值图。然后进行基于区域生长的文字分割。最后,应用OCR(optical character recognition)模块提高文本分割的整体性能。将提出的算法与其他算法进行了比较,结果表明,所提算法更为有效。     

An omnifont open-vocabulary OCR system for English and Arabic

We present an omnifont, unlimited-vocabulary OCR system for English and Arabic. The system is based on hidden Markov models (HMM), an approach that has proven to be very successful in the area of automatic speech recognition. We focus on two aspects of the OCR system. First, we address the issue of how to perform OCR on omnifont and multi-style data, such as plain and italic, without the need to have a separate model for each style. The amount of training data from each style, which is used to train a single model, becomes an important issue in the face of the conditional independence assumption inherent in the use of HMMs. We demonstrate mathematically and empirically how to allocate training data among the different styles to alleviate this problem. Second, we show how to use a word-based HMM system to perform character recognition with unlimited vocabulary. The method includes the use of a trigram language model on character sequences. Using all these techniques, we have achieved character error rates of 1.1 percent on data from the University of Washington English Document Image Database and 3.3 percent on data from the DARPA Arabic OCR Corpus     

Building a multi-modal Arabic corpus (MMAC)

Traditionally, a corpus is a large structured set of text, electronically stored and processed. Corpora have become very important in the study of languages. They have opened new areas of linguistic research, which were unknown until recently. Corpora are also key to the development of optical character recognition (OCR) applications. Access to a corpus of both language and images is essential during OCR development, particularly while training and testing a recognition application. Excellent corpora have been developed for Latin-based languages, but few relate to the Arabic language. This limits the penetration of both corpus linguistics and OCR in Arabic-speaking countries. This paper describes the construction and provides a comprehensive study and analysis of a multi-modal Arabic corpus (MMAC) that is suitable for use in both OCR development and linguistics. MMAC currently contains six million Arabic words and, unlike previous corpora, also includes connected segments or pieces of Arabic words (PAWs) as well as naked pieces of Arabic words (NPAWs) and naked words (NWords); PAWs and Words without diacritical marks. Multi-modal data is generated from both text, gathered from a wide variety of sources, and images of existing documents. Text-based data is complemented by a set of artificially generated images showing each of the Words, NWords, PAWs and NPAWs involved. Applications are provided to generate a natural-looking degradation to the generated images. A ground truth annotation is offered for each such image, while natural images showing small paragraphs and full pages are augmented with representations of the text they depict. A statistical analysis and verification of the dataset has been carried out and is presented. MMAC was also tested using commercial OCR software and is publicly and freely available.     

Neural Networks Pipeline for Offline Machine Printed Arabic OCR

In the context of Arabic optical characters recognition, Arabic poses more challenges because of its cursive nature. We purpose a system for recognizing a document containing Arabic text, using a pipeline of three neural networks. The first network model predicts the font size of an Arabic word, then the word is normalized to an 18pt font size that will be used to train the next two models. The second model is used to segment a word into characters. The problem of words segmentation in the Arabic language, as in many similar cursive languages, presents a challenge to the OCR systems. This paper presents a multichannel neural network to solve the offline segmentation of machine-printed Arabic documents. The segmented characters are then fed as an input to a convolutional neural network for Arabic characters recognition. The font size prediction model produced a test accuracy of 99.1%. The accuracy of the segmentation model using one font is 98.9%, while four-font model showed 95.5% accuracy. The whole pipeline showed an accuracy of 94.38% on Arabic Transparent font of size 18pt from APTI data set.     

Recognition of Handwritten Arabic Characters using Histograms of Oriented Gradient (HOG)

Optical Character Recognition (OCR) is the process of recognizing printed or handwritten text on paper documents. This paper proposes an OCR system for Arabic characters. In addition to the preprocessing phase, the proposed recognition system consists mainly of three phases. In the first phase, we employ word segmentation to extract characters. In the second phase, Histograms of Oriented Gradient (HOG) are used for feature extraction. The final phase employs Support Vector Machine (SVM) for classifying characters. We have applied the proposed method for the recognition of Jordanian city, town, and village names as a case study, in addition to many other words that offers the characters shapes that are not covered with Jordan cites. The set has carefully been selected to include every Arabic character in its all four forms. To this end, we have built our own dataset consisting of more than 43.000 handwritten Arabic words (30000 used in the training stage and 13000 used in the testing stage). Experimental results showed a great success of our recognition method compared to the state of the art techniques, where we could achieve very high recognition rates exceeding 99%.     

Robust named entity detection from optical character recognition output

In this paper, we focus on information extraction from optical character recognition (OCR) output. Since the content from OCR inherently has many errors, we present robust algorithms for information extraction from OCR lattices instead of merely looking them up in the top-choice (1-best) OCR output. Specifically, we address the challenge of named entity detection in noisy OCR output and show that searching for named entities in the recognition lattice significantly improves detection accuracy over 1-best search. While lattice-based named entity (NE) detection improves NE recall from OCR output, there are two problems with this approach: (1) the number of false alarms can be prohibitive for certain applications and (2) lattice-based search is computationally more expensive than 1-best NE lookup. To mitigate the above challenges, we present techniques for reducing false alarms using confidence measures and for reducing the amount of computation involved in performing the NE search. Furthermore, to demonstrate that our techniques are applicable across multiple domains and languages, we experiment with optical character recognition systems for videotext in English and scanned handwritten text in Arabic.     

End-to-end scene text recognition using tree-structured models

Detecting and recognizing text in natural images are quite challenging and have received much attention from the computer vision community in recent years. In this paper, we propose a robust end-to-end scene text recognition method, which utilizes tree-structured character models and normalized pictorial structured word models. For each category of characters, we build a part-based tree-structured model (TSM) so as to make use of the character-specific structure information as well as the local appearance information. The TSM could detect each part of the character and recognize the unique structure as well, seamlessly combining character detection and recognition together. As the TSMs could accurately detect characters from complex background, for text localization, we apply TSMs for all the characters on the coarse text detection regions to eliminate the false positives and search the possible missing characters as well. While for word recognition, we propose a normalized pictorial structure (PS) framework to deal with the bias caused by words of different lengths. Experimental results on a range of challenging public datasets (ICDAR 2003, ICDAR 2011, SVT) demonstrate that the proposed method outperforms state-of-the-art methods both for text localization and word recognition.     

A hierarchical and scalable model for contemporary document image segmentation

In this paper, we introduce a novel color segmentation approach robust against digitization noise and adapted to contemporary document images. This system is scalable, hierarchical, versatile and completely automated, i.e. user independent. It proposes an adaptive binarization/quantization without any penalizing information loss. This model may be used for many purposes. For instance, we rely on it to carry out the first steps leading to advertisement recognition in document images. Furthermore, the color segmentation output is used to localize text areas and enhance optical character recognition (OCR) performances. We held tests on a variety of magazine images to point up our contribution to the well-known OCR product Abby FinerReader. We also get promising results with our ad detection system on a large set of complex layout testing images.     

Video OCR: indexing digital news libraries by recognition of superimposed captions

The automatic extraction and recognition of news captions and annotations can be of great help locating topics of interest in digital news video libraries. To achieve this goal, we present a technique, called Video OCR (Optical Character Reader), which detects, extracts, and reads text areas in digital video data. In this paper, we address problems, describe the method by which Video OCR operates, and suggest applications for its use in digital news archives. To solve two problems of character recognition for videos, low-resolution characters and extremely complex backgrounds, we apply an interpolation filter, multi-frame integration and character extraction filters. Character segmentation is performed by a recognition-based segmentation method, and intermediate character recognition results are used to improve the segmentation. We also include a method for locating text areas using text-like properties and the use of a language-based postprocessing technique to increase word recognition rates. The overall recognition results are satisfactory for use in news indexing. Performing Video OCR on news video and combining its results with other video understanding techniques will improve the overall understanding of the news video content.     

Off-line Arabic character recognition

A personal computer-based Arabic character recognition system that performs three preprocessing stages sequentially, thinning, stroke segmentation, and sampling, is described. The eight-direction code used for stroke representation and classification, the character classification done at primary and secondary levels, and the contextual postprocessor used for error detection and correction are described. Experimental results obtained using samples of handwritten and typewritten Arabic words are presented     

Document cleanup using page frame detection

When a page of a book is scanned or photocopied, textual noise (extraneous symbols from the neighboring page) and/or non-textual noise (black borders, speckles, ...) appear along the border of the document. Existing document analysis methods can handle non-textual noise reasonably well, whereas textual noise still presents a major issue for document analysis systems. Textual noise may result in undesired text in optical character recognition (OCR) output that needs to be removed afterwards. Existing document cleanup methods try to explicitly detect and remove marginal noise. This paper presents a new perspective for document image cleanup by detecting the page frame of the document. The goal of page frame detection is to find the actual page contents area, ignoring marginal noise along the page border. We use a geometric matching algorithm to find the optimal page frame of structured documents (journal articles, books, magazines) by exploiting their text alignment property. We evaluate the algorithm on the UW-III database. The results show that the error rates are below 4% each of the performance measures used. Further tests were run on a dataset of magazine pages and on a set of camera captured document images. To demonstrate the benefits of using page frame detection in practical applications, we choose OCR and layout-based document image retrieval as sample applications. Experiments using a commercial OCR system show that by removing characters outside the computed page frame, the OCR error rate is reduced from 4.3 to 1.7% on the UW-III dataset. The use of page frame detection in layout-based document image retrieval application decreases the retrieval error rates by 30%.     

Machine translation from Arabic to English and French

As the cognitive processes of natural language understanding and generation are better understood, it is becoming easier, nowadays, to perform machine translation. In this paper we present our work on machine translation from Arabic to English and French, and illustrate it with a fully operational system, which runs on PC compatibles with Arabic/Latin interface. This system is an extension of an earlier system, whose task was the analysis of the natural language Arabic. Thanks to the regularity of its phrase structures and word patterns, Arabic lends itself quite naturally to a Fillmore-like analysis. The meaning of a phrase is stored in a star-like data structure, where the verb occupies the center of the star and the various noun sentences occupy specific peripheral nodes of the star. The data structure is then translated into an internal representation in the target language, which is then mapped into the target text.     

Automatic performance evaluation of printed Chinese character recognition systems

Performance evaluation is crucial for improving the performance of OCR systems. However, this is trivial and sophisticated work to do by hand. Therefore, we have developed an automatic performance evaluation system for a printed Chinese character recognition (PCCR) system. Our system is characterized by using real-world data as test data and automatically obtaining the performance of the PCCR system by comparing the correct text and the recognition result of the document image. In addition, our performance evaluation system also provides some evaluation of performance for the segmentation module, the classification module, and the post-processing module of the PCCR system. For this purpose, a segmentation error-tolerant character-string matching algorithm is proposed to obtain the correspondence between the correct text and the recognition result. The experiments show that our performance evaluation system is an accurate and powerful tool for studying deficiencies in the PCCR system. Although our approach is aimed at the PCCR system, the idea also can be applied to other OCR systems.