Canhui Xu

Graphic composite segmentation for PDF documents with complex layouts

Converting the PDF books to re-flowable format has recently attracted various interests in the area of e-book reading. Robust graphic segmentation is highly desired for increasing the practicability of PDF converters. To cope with various layouts, a multi-layer concept is introduced to segment graphic composites including photographic images, drawings with text insets or surrounded with text elements. Both image based analysis and inherent digital born document advantages are exploited in this multi-layer based layout analysis method. By combining low-level page elements clustering applied on PDF documents and connected component analysis on synthetically generated PNG image document, graphic composites can be segmented for PDF documents with complex layouts. The experimental results on graphic composite segmentation of PDF document pages have shown satisfactory performance.

Character feature integration of Chinese calligraphy and font

A framework is proposed in this paper to effectively generate a new hybrid character type by means of integrating local contour feature of Chinese calligraphy with structural feature of font in computer system. To explore traditional art manifestation of calligraphy, multi-directional spatial filter is applied for local contour feature extraction. Then the contour of character image is divided into sub-images. The sub-images in the identical position from various characters are estimated by Gaussian distribution. According to its probability distribution, the dilation operator and erosion operator are designed to adjust the boundary of font image. And then new Chinese character images are generated which possess both contour feature of artistical calligraphy and elaborate structural feature of font. Experimental results demonstrate the new characters are visually acceptable, and the proposed framework is an effective and efficient strategy to automatically generate the new hybrid character of calligraphy and font.

Contextual modeling for logical labeling of PDF documents

The widely-used Portable Document Format (PDF) documents are known to be layout-oriented and not suitable for mobile applications. In this paper, a Conditional Random Fields (CRF) based model is proposed to learn latent semantics of PDF page content. Local and contextual observations constructed from PDF attributes are incorporated to facilitate the determination of semantic roles. The observations are carefully designed to work even in different styles of documents. A local classifier is first used to generate posterior probabilities. The local estimate is then fed to the CRF model for joint classification. The experimental results evidently approve the positive effects of contextual information in logical labeling. Our work has revealed the potential usability of existing born-digital fixed-layout documents for mobile applications.

Graph-based layout analysis for PDF documents

To increase the flexibility and enrich the reading experience of e-book on small portable screens, a graph based method is proposed to perform layout analysis on Portable Document Format (PDF) documents. Digital born document has its inherent advantages like representing texts and fractional images in explicit form, which can be straightforwardly exploited. To integrate traditional image-based document analysis and the inherent meta-data provided by PDF parser, the page primitives including text, image and path elements are processed to produce text and non text layer for respective analysis. Graph-based method is developed in superpixel representation level, and page text elements corresponding to vertices are used to construct an undirected graph. Euclidean distance between adjacent vertices is applied in a top-down manner to cut the graph tree formed by Kruskal’s algorithm. And edge orientation is then used in a bottom-up manner to extract text lines from each sub tree. On the other hand, non-textual objects are segmented by connected component analysis. For each segmented text and non-text composite, a 13-dimensional feature vector is extracted for labelling purpose. The experimental results on selected pages from PDF books are presented.

Automatic Generation of Chinese Character Based on Human Vision and Prior Knowledge of Calligraphy

Prior knowledge of Chinese calligraphy is modeled in this paper, and the hierarchical relationship of strokes and radicals is represented by a novel five layer framework. Calligraphist’s unique calligraphy skill is analyzed and his particular strokes, radicals and layout patterns provide raw element for the proposed five layers. The criteria of visual aesthetics based on Marr’s vision assumption are built for the proposed algorithm of automatic generation of Chinese character. The Bayesian statistics is introduced to characterize the character generation process as a Bayesian dynamic model, in which, parameters to translate, rotate and scale strokes, radicals are controlled by the state equation, as well as the proposed visual aesthetics is employed by the measurement equation. Experimental results show the automatically generated characters have almost the same visual acceptance compared to calligraphist’s artwork.

Ground-Truth and Performance Evaluation for Page Layout Analysis of Born-Digital Documents

In this paper, a new dataset is proposed for page layout analysis of born-digital documents. By extracting uniformly the document contents, an XML based data format is designed in terms of raw data and structure data. Utilizing a self-developed ground-truthing tool, a public dataset is constructed from diverse styles of document resources. With consideration of physical segmentation and logical labeling, automatic performance evaluation methods are adjusted to cope with different scenarios. The applications of the proposed dataset have shown that it is suitable for evaluating various layout analysis tasks.

Logical Labeling of Fixed Layout PDF Documents Using Multiple Contexts

The task of logical structure recovery is known to be of crucial importance, yet remains unsolved not only for image based document but also for born-digital document system. In this work, the modeling of contextual information based on 2D Conditional Random Fields is proposed to learn page structure for born-digital fixed-layout documents. Heuristic prior knowledge of Portable Document Format (PDF) content and layout are interpreted to construct neighborhood graphs and various pair wise clique templates for the modeling of multiple contexts. By integrating local and contextual observations obtained from PDF attributes, the ambiguities of semantic labels are better resolved. Experimental comparisons for six types of clique templates has demonstrated the benefits of contextual information in logical labeling of 16 finely defined categories.

Nonlinear and non-Gaussian Bayesian based handwriting beautification

A framework is proposed in this paper to effectively and efficiently beautify handwriting by means of a novel nonlinear and non-Gaussian Bayesian algorithm. In the proposed framework, format and size of handwriting image are firstly normalized, and then typeface in computer system is applied to optimize vision effect of handwriting. The Bayesian statistics is exploited to characterize the handwriting beautification process as a Bayesian dynamic model. The model parameters to translate, rotate and scale typeface in computer system are controlled by state equation, and the matching optimization between handwriting and transformed typeface is employed by measurement equation. Finally, the new typeface, which is transformed from the original one and gains the best nonlinear and non-Gaussian optimization, is the beautification result of handwriting. Experimental results demonstrate the proposed framework provides a creative handwriting beautification methodology to improve visual acceptance.

Document page structure learning for fixed-layout e-books using conditional random fields

In this paper, a model is proposed to learn logical structure of fixed-layout document pages by combining support vector machine (SVM) and conditional random fields (CRF). Features related to each logical label and their dependencies are extracted from various original Portable Document Format (PDF) attributes. Both local evidence and contextual dependencies are integrated in the proposed model so as to achieve better logical labeling performance. With the merits of SVM as local discriminative classifier and CRF modeling contextual correlations of adjacent fragments, it is capable of resolving the ambiguities of semantic labels. The experimental results show that CRF based models with both tree and chain graph structures outperform the SVM model with an increase of macro-averaged F1 by about 10%.

Integration of Text Information and Graphic Composite for PDF Document Analysis

The trend of large scale digitization has greatly motivated the research on the processing of the PDF documents with little structure information. Challenging problems like graphic segmentation integrating with texts remain unsolved for successful practical application of PDF layout analysis. To cope with PDF documents, a hybrid method incorporating text information and graphic composite is proposed to segment the pages that are difficult to handle by traditional methods. Specifically, the text information is derived accurately from born-digital documents embedded with low-level structure elements in explicit form. Then page text elements are clustered by applying graph based method according to proximity and feature similarity. Meanwhile, the graphic components are extracted by means of texture and morphological analysis. By integrating the clustered text elements with image based graphic components, the graphics are segmented for layout analysis. The experimental results on pages of PDF books have shown satisfactory performance.