Doctoral Thesis
Refine
Document Type
- Doctoral Thesis (3) (remove)
Language
- English (3)
Has Fulltext
- yes (3)
Keywords
- Optische Zeichenerkennung (3) (remove)
Faculty / Organisational entity
Generic layout analysis--process of decomposing document image into homogeneous regions for a collection of diverse document images--has many important applications in document image analysis and understanding such as preprocessing of degraded warped, camera-captured document images, high performance layout analysis of document images containing complex cursive scripts, and word spotting in historical document images at page level. Many areas in this field like generic text line extraction method are considered as elusive goals so far, still beyond the reach of the state-of-the-art methods [NJ07, LSZT07, KB06]. This thesis addresses this problem in such a way that it presents generic, domain-independent, text line extraction and text and non-text segmentation methods, and then describes some important applications, that were developed based on these methods. An overview of the key contributions of this thesis is as follows.
The first part of this thesis presents a generic text line extraction method using a combination of matched filtering and ridge detection techniques, which are commonly used in computer vision. Unlike the state-of-the-art text line extraction methods in the literature, the generic text line extraction method can be equally and robustly applied to a large variety of document image classes including scanned and camera-captured documents, binary and grayscale documents, typed-text and handwritten documents, historical and contemporary documents, and documents containing different scripts. Different standard datasets are selected for performance evaluation that belong to different categories of document images such as the UW-III [GHHP97] dataset of scanned documents, the ICDAR 2007 [GAS07] and the UMD [LZDJ08] datasets of handwritten documents, the DFKI-I [SB07] dataset of camera-captured documents, Arabic/Urdu script documents dataset, and German calligraphic (Fraktur) script historical documents dataset. The generic text line extraction method achieves 86% (n = 23,763 text lines in 650 documents) text line detection accuracy which is better than the aggregate accuracy of 73% of the best performing domain-specific state-of-the-art methods. To the best of the author's knowledge, it is the first general-purpose text line extraction method that can be equally used for a diverse collection of documents.
This thesis also presents an active contour (snake) based curled text line extraction method for warped, camera-captured document images. The presented approach is applied to DFKI-I [SB07] dataset of camera-captured, Latin script document images for curled text line extraction. It achieves above 95% (n = 3,091 text lines in 102 documents) text line detection accuracy, which is significantly better than the competing state-of-the-art curled text line extraction methods. The presented text line extraction method can also be applied to document images containing different scripts like Chinese, Devanagari, and Arabic after small modifications.
The second part of this thesis presents an improved version of the state-of-the-art multiresolution morphology (Leptonica) based text and non-text segmentation method [Blo91], which is a domain-independent page segmentation approach and can be equally applied to a diverse collection of binarized document images. It is demonstrated that the presented improvements result in an increase in segmentation accuracy from 93% to 99% (n = 113 documents).
This thesis also introduces a discriminative learning based approach for page segmentation, where a self-tunable multi-layer perceptron (MLP) classifier [BS10] is trained for distinguishing between text and non-text connected components. Unlike other classification based page segmentation approaches in the literature, the connected components based discriminative learning based approach is faster than pixel based classification methods and does not require a block segmentation method beforehand. A segmentation accuracy of $96\%$ ($n = 113$ documents) is achieved in comparison to the state-of-the-art multiresolution morphology (Leptonica) based page segmentation method [Blo91] that achieves a segmentation accuracy of 93%. In addition to text and non-text segmentation of Latin script documents, the presented approach can also be adapted for document images containing other scripts as well as for other specialized layout analysis tasks such as digit and non-digit segmentation [HBSB12], orientation detection [RBSB09], and body-text and side-note segmentation [BAESB12].
Finally, this thesis presents important applications of the two generic layout analysis techniques, ridge-based text line extraction method and the multi-resolution morphology based text and non-text segmentation method, discussed above. First, a complete preprocessing pipeline is described for removing different types of degradations from grayscale warped, camera-captured document images that includes removal of grayscale degradations such as non-uniform shadows and blurring through binarization, noise cleanup applying page frame detection, and document rectification using monocular dewarping. Each of these preprocessing steps shows significant improvement in comparison to the analyzed state-of-the-art methods in the literature. Second, a high performance layout analysis method is described for complex Arabic script document images written in different languages such as Arabic, Urdu, and Persian and different styles for example Naskh and Nastaliq. The presented layout analysis system is robust against different types of document image degradations and shows better performance for text and non-text segmentation, text line extraction, and reading order determination on a variety of Arabic and Urdu document images as compared to the state-of-the-art methods. It can be used for large scale Arabic and Urdu documents' digitization processes. These applications demonstrate that the layout analysis methods, ridge-based text line extraction and the multi-resolution morphology based text and non-text segmentation, are generic and can be applied easily to a large collection of diverse document images.
Layout analysis--the division of page images into text blocks, lines, and determination of their reading order--is a major performance limiting step in large scale document digitization projects. This thesis addresses this problem in several ways: it presents new performance measures to identify important classes of layout errors, evaluates the performance of state-of-the-art layout analysis algorithms, presents a number of methods to reduce the error rate and catastrophic failures occurring during layout analysis, and develops a statistically motivated, trainable layout analysis system that addresses the needs of large-scale document analysis applications. An overview of the key contributions of this thesis is as follows. First, this thesis presents an efficient local adaptive thresholding algorithm that yields the same quality of binarization as that of state-of-the-art local binarization methods, but runs in time close to that of global thresholding methods, independent of the local window size. Tests on the UW-1 dataset demonstrate a 20-fold speedup compared to traditional local thresholding techniques. Then, this thesis presents a new perspective for document image cleanup. Instead of trying to explicitly detect and remove marginal noise, the approach focuses on locating the page frame, i.e. the actual page contents area. A geometric matching algorithm is presented to extract the page frame of a structured document. It is demonstrated that incorporating page frame detection step into document processing chain results in a reduction in OCR error rates from 4.3% to 1.7% (n=4,831,618 characters) on the UW-III dataset and layout-based retrieval error rates from 7.5% to 5.3% (n=815 documents) on the MARG dataset. The performance of six widely used page segmentation algorithms (x-y cut, smearing, whitespace analysis, constrained text-line finding, docstrum, and Voronoi) on the UW-III database is evaluated in this work using a state-of-the-art evaluation methodology. It is shown that current evaluation scores are insufficient for diagnosing specific errors in page segmentation and fail to identify some classes of serious segmentation errors altogether. Thus, a vectorial score is introduced that is sensitive to, and identifies, the most important classes of segmentation errors (over-, under-, and mis-segmentation) and what page components (lines, blocks, etc.) are affected. Unlike previous schemes, this evaluation method has a canonical representation of ground truth data and guarantees pixel-accurate evaluation results for arbitrary region shapes. Based on a detailed analysis of the errors made by different page segmentation algorithms, this thesis presents a novel combination of the line-based approach by Breuel with the area-based approach of Baird which solves the over-segmentation problem in area-based approaches. This new approach achieves a mean text-line extraction error rate of 4.4% (n=878 documents) on the UW-III dataset, which is the lowest among the analyzed algorithms. This thesis also describes a simple, fast, and accurate system for document image zone classification that results from a detailed comparative analysis of performance of widely used features in document analysis and content-based image retrieval. Using a novel combination of known algorithms, an error rate of 1.46% (n=13,811 zones) is achieved on the UW-III dataset in comparison to a state-of-the-art system that reports an error rate of 1.55% (n=24,177 zones) using more complicated techniques. In addition to layout analysis of Roman script documents, this work also presents the first high-performance layout analysis method for Urdu script. For that purpose a geometric text-line model for Urdu script is presented. It is shown that the method can accurately extract Urdu text-lines from documents of different layouts like prose books, poetry books, magazines, and newspapers. Finally, this thesis presents a novel algorithm for probabilistic layout analysis that specifically addresses the needs of large-scale digitization projects. The presented approach models known page layouts as a structural mixture model. A probabilistic matching algorithm is presented that gives multiple interpretations of input layout with associated probabilities. An algorithm based on A* search is presented for finding the most likely layout of a page, given its structural layout model. For training layout models, an EM-like algorithm is presented that is capable of learning the geometric variability of layout structures from data, without the need for a page segmentation ground-truth. Evaluation of the algorithm on documents from the MARG dataset shows an accuracy of above 95% for geometric layout analysis.
Sequence learning describes the process of understanding the spatio-temporal
relations in a sequence in order to classify it, label its elements or generate
new sequences. Due to the prevalence of structured sequences in nature
and everyday life, it has many practical applications including any language
related processing task. One particular such task that has seen recent success
using sequence learning techniques is the optical recognition of characters
(OCR).
State-of-the-art sequence learning solutions for OCR achieve high performance
through supervised training, which requires large amounts of transcribed
training data. On the other hand, few solutions have been proposed on how
to apply sequence learning in the absence of such data, which is especially
common for hard to transcribe historical documents. Rather than solving
the unsupervised training problem, research has focused on creating efficient
methods for collecting training data through smart annotation tools or generating
synthetic training data. These solutions come with various limitations
and do not solve all of the related problems.
In this work, first the use of erroneous transcriptions for supervised sequence
learning is introduced and it is described how this concept can be applied in
unsupervised training scenarios by collecting or generating such transcriptions.
The proposed OCR pipeline reduces the need of domain specific expertise
to apply OCR, with the goal of making it more accessible. Furthermore, an
approach for evaluating sequence learning OCR models in the absence of
reference transcriptions is presented and its different properties compared
to the standard method are discussed. In a second approach, unsupervised
OCR is treated as an alignment problem between the latent features of the
different language modalities. The outlined solution is to extract language
properties from both the text and image domain through adversarial training
and learn to align them by adding a cycle consistency constraint. The proposed
approach has some strict limitations on the input data, but the results
encourage future research into more widespread applications.