scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

A CRF Based Scheme for Overlapping Multi-colored Text Graphics Separation

TL;DR: A novel framework for segmentation of documents with complex layouts performed by combination of clustering and conditional random fields (CRF) based modeling and has been extensively tested on multi-colored document images with text overlapping graphics/image.
Abstract: In this paper, we propose a novel framework for segmentation of documents with complex layouts. The document segmentation is performed by combination of clustering and conditional random fields (CRF) based modeling. The bottom-up approach for segmentation assigns each pixel to a cluster plane based on color intensity. A CRF based discriminative model is learned to extract the local neighborhood information in different cluster/color planes. The final category assignment is done by a top-level CRF based on the semantic correlation learned across clusters. The proposed framework has been extensively tested on multi-colored document images with text overlapping graphics/image.
Citations
More filters
Journal ArticleDOI
TL;DR: This survey highlights the variety of the approaches that have been proposed for document image segmentation since 2008 and provides a clear typology of documents and of document images segmentation algorithms.
Abstract: In document image analysis, segmentation is the task that identifies the regions of a document. The increasing number of applications of document analysis requires a good knowledge of the available technologies. This survey highlights the variety of the approaches that have been proposed for document image segmentation since 2008. It provides a clear typology of documents and of document image segmentation algorithms. We also discuss the technical limitations of these algorithms, the way they are evaluated and the general trends of the community.

84 citations


Cites background from "A CRF Based Scheme for Overlapping ..."

  • ...[110] separate text and graphics in challenging magazine covers....

    [...]

  • ...Algorithm Input layout Multilayered Color depth Labels Training Type of output Text orientation Text alignment Data set test size Nb of languages Nb of doc types [72] Any Yes Color Yes No Text lines Horizontal Straight 448 1 2 [104] Any Yes Color Yes Yes Regions Any Curved 87 1 2 [110] Any Yes Color Yes Yes Regions Any Curved 16 2 1 [51] Any Yes Color No No Regions Any Curved 2000 2 2 [74] Any Yes Color No No Text lines Any Curved 50 1 1 [84] Any Yes Color No Yes Text lines Any Straight 21 2 1 [54] Any Yes Gray Yes No Regions Horizontal Straight 65 2 1 [112] Any Yes BW Yes Yes Regions Horizontal Straight 1000 3 6 [89] Any No Color Yes Yes Regions Any Curved 100 1 1 Continued on next page...

    [...]

Proceedings ArticleDOI
09 Sep 2013
TL;DR: This paper tries to find a new way which can utilize existing methods to detect and extract text from born-digital image.
Abstract: The text appears in the images is important for fully understanding the images. The number of digital images and digital videos has increased tremendously. Although there are many methods have been proposed over the past years for the text extraction from natural scene images, the text detection and extraction from born-digital images are still a challenge. In this paper, we describe existing methods key ideas and try to summarize their advantages and disadvantages. We try to find a new way which can Comprehensive utilize existing methods to detect and extract text from born-digital image.

12 citations


Cites methods from "A CRF Based Scheme for Overlapping ..."

  • ...1) Method based on CC [2-15] The method consists of two steps....

    [...]

Book ChapterDOI
10 Dec 2013
TL;DR: An approach to separate the non-texts from texts of a manuscript, mainly in the form of doodles and drawings of some exceptional thinkers and writers, and a computational approach to recover the struck-out texts to reduce human effort.
Abstract: In this paper we propose an approach to separate the non-texts from texts of a manuscript. The non-texts are mainly in the form of doodles and drawings of some exceptional thinkers and writers. These have enormous historical values due to study on those writers’ subconscious as well as productive mind. We also propose a computational approach to recover the struck-out texts to reduce human effort. The proposed technique has a preprocessing stage, which removes noise using median filter and segments object region using fuzzy c-means clustering. Now connected component analysis finds the major portions of non-texts, and window examination eliminates the partially attached texts. The struck-out texts are extracted by eliminating straight lines, measuring degree of continuity, using some morphological operations.

10 citations


Cites methods from "A CRF Based Scheme for Overlapping ..."

  • ...The existing methods [2-10] deal with different logos, diagrams, maps, engineering drawings and photographic images....

    [...]

Journal ArticleDOI
TL;DR: A consensus-based clustering approach for document image segmentation that is used iteratively with a classifier to label each primitive block and shows that the dependency of classification performance on the training data is significantly reduced.
Abstract: Segmentation of a document image plays an important role in automatic document processing. In this paper, we propose a consensus-based clustering approach for document image segmentation. In this method, the foreground regions of a document image are grouped into a set of primitive blocks, and a set of features is extracted from them. Similarities among the blocks are computed on each feature using a hypothesis test-based similarity measure. Based on the consensus of these similarities, clustering is performed on the primitive blocks. This clustering approach is used iteratively with a classifier to label each primitive block. Experimental results show the effectiveness of the proposed method. It is further shown in the experimental results that the dependency of classification performance on the training data is significantly reduced.

9 citations


Cites methods from "A CRF Based Scheme for Overlapping ..."

  • ...A bottom-up approach using conditional random field (CRF) for text–graphics separation has been reported in [21]....

    [...]

Journal ArticleDOI
Jun Chen1, Hong Zhao1, Jufeng Yang1, Jian Zhang1, Tao Li1, Kai Wang1 
01 Feb 2017
TL;DR: STRHOG, an extended version of HOG that is helpful for filtering spam images on cloud and a fair comparison with other methods, nearest neighbor classifier is used for the intelligent character recognition.
Abstract: Cloud storage has become an important way for data sharing in recent years. Data protection for data owner and harmful data filtering for data recipients are two non-negligible problems in cloud storage. Illegal or unsuitable messages on cloud have a negative impact on minors and they are easily converted into images to avoid text-based filtering. To detect the spam image with the embedded harmful messages on cloud, soft computing methods are required for intelligent character recognition. HOG, proposed by Dalal and Triggs, has been demonstrated so far to be one of the best features for intelligent character recognition. A pre-defined sliding window is always used for the generation of candidate character images when HOG is applied to recognize the whole word. However, due to the difference in character sizes, the pre-defined window cannot exactly match with each character. Variations on scale and translation usually occur in the character image to be recognized, which have a great influence on the performance of intelligent character recognition. Aiming to solve this problem, STRHOG, an extended version of HOG, is proposed in this paper. Experiments on two public datasets and one our dataset have shown encouraging results for our work. The improved intelligent character recognition is helpful for filtering spam images on cloud. To make a fair comparison with other methods, nearest neighbor classifier is used for the intelligent character recognition. It is expected that the performance should be further improved by using better classifiers such as fuzzy neural network.

7 citations

References
More filters
Proceedings ArticleDOI
27 Sep 1999
TL;DR: It is shown that, in general, the discrimination effectiveness of the features increases with the amount of post-Gabor processing.
Abstract: The performance of a number of texture feature operators is evaluated. The features are all based on the local spectrum which is obtained by a bank of Gabor filters. The comparison is made using a quantitative method which is based on Fisher's criterion. It is shown that, in general, the discrimination effectiveness of the features increases with the amount of post-Gabor processing.

340 citations


"A CRF Based Scheme for Overlapping ..." refers methods in this paper

  • ...We extract the following local features in each of these color planes: a. Gabor Features: Texture features are based on the local power spectrum that are computed using 2D gabor filter....

    [...]

Proceedings ArticleDOI
13 Jan 2003
TL;DR: This paper provides a detailed survey of past work on document structure analysis algorithms and summarize the limitations of past approaches.
Abstract: Document structure analysis can be regarded as a syntactic analysis problem. The order and containment relations among the physical or logical components of a document page can be described by an ordered tree structure and can be modeled by a tree grammar which describes the page at the component level in terms of regions or blocks. This paper provides a detailed survey of past work on document structure analysis algorithms and summarize the limitations of past approaches. In particular, we survey past work on document physical layout representations and algorithms, document logical structure representations and algorithms, and performance evaluation of document structure analysis algorithms. In the last section, we summarize this work and point out its limitations.

278 citations


"A CRF Based Scheme for Overlapping ..." refers background in this paper

  • ...Over the years several document layout analysis algorithms have been proposed in [1], [2]....

    [...]

  • ...…+ γ2y′2 2δ2 ) cos(2π x′ λ +ϕ) (1) Here x′ = xcosθ + ysinθ, y′ = −xsinθ + ycosθ and λ is the wavelength of the cosine factor of the gabor filter kernel, θ is the orientation, ϕ is the phase offset with ϕ = [0 π/2], δ is the standard deviation of Gaussian function and γ is the aspect ratio (γ = 0.5)....

    [...]

Book
11 Jul 2003
TL;DR: The origins of modern color science can be traced back to the early 1970s, when the color vision was described by Mollon et al. as discussed by the authors, who used light, retinal image, and photoreceptors.
Abstract: Preface. Dedication. Contributors. 1. The origins of modern color science (J. Mollon). 2. Light, the retinal image, and photoreceptors (O. Packer, D.R. Williams). 3. Color matching and color discrimination (V.C. Smith, J. Pokorny). 4. Color appearance (S.K. Shevell). 5. Color appearance and color difference specification (D.H. Brainard). 6. The physiology of color vision (P. Lennie). 7. The physics and chemistry of color: The fifteen mechanisms (K. Nassau). 8. Digital color reproduction (B. Wandell, L.D. Silverstein). Index.

169 citations

Journal ArticleDOI
TL;DR: A clustering-based technique has been devised for estimating globally matched wavelet filters using a collection of groundtruth images and a text extraction scheme for the segmentation of document images into text, background, and picture components is extended.
Abstract: In this paper, we have proposed a novel scheme for the extraction of textual areas of an image using globally matched wavelet filters. A clustering-based technique has been devised for estimating globally matched wavelet filters using a collection of groundtruth images. We have extended our text extraction scheme for the segmentation of document images into text, background, and picture components (which include graphics and continuous tone images). Multiple, two-class Fisher classifiers have been used for this purpose. We also exploit contextual information by using a Markov random field formulation-based pixel labeling scheme for refinement of the segmentation results. Experimental results have established effectiveness of our approach.

159 citations


"A CRF Based Scheme for Overlapping ..." refers methods in this paper

  • ...The algorithm uses connected component analysis based on color similarity in the RGB color space....

    [...]

01 Jan 2008
TL;DR: The main purpose of the present report is to describe the current status of DIU with particular attention to two subprocesses: document skew angle estimation and page decomposition.
Abstract: Document Image Understanding (DIU) is an interesting research area with a large variety of challenging applications. Researchers have worked from decades on this topic, as witnessed by the scientific literature. The main purpose of the present report is to describe the current status of DIU with particular attention to two subprocesses: document skew angle estimation and page decomposition. Several algorithms proposed in the literature are synthetically described. They are included in a novel classification scheme. Some methods proposed for the evaluation of page decomposition algorithms are described. Critical discussions are reported about the current status of the field and about the open problems. Some considerations about the logical layout analysis are also reported.

128 citations


"A CRF Based Scheme for Overlapping ..." refers background in this paper

  • ...Over the years several document layout analysis algorithms have been proposed in [1], [2]....

    [...]