Segmentation of Page Images Using the Area Voronoi Diagram

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.

/pdf/document-structure-analysis-algorithms-a-literature-survey-4fkbwys4m0.pdf

Document structure analysis algorithms: a literature survey

Page segmentation is the process by which a scanned page is divided into columns and blocks which are then classified as halftones, graphics, or text. Past techniques have used the fact that such parts form right rectangles for most printed material. This property is not true when the page is tilted, and the heuristics based on it fail in such cases unless a rather expensive tilt angle estimation is performed. We describe a class of techniques based on smeared run length codes that divide a page into gray and nearly white parts. Segmentation is then performed by finding connected components either by the gray elements or of the white, the latter forming white streams that partition a page into blocks of printed material. Such techniques appear quite robust in the presence of severe tilt (even greater than 10 °) and are also quite fast (about a second a page on a SPARC station for gray element aggregation). Further classification into text or halftones is based mostly on properties of the across scanlines correlation. For text correlation of adjacent scanlines tends to be quite high, but then it drops rapidly. For halftones, the correlation of adjacent scanlines is usually well below that for text, but it does not change much with distance.

Page segmentation and classification

In this paper, we address the problem of the identification of text in noisy document images. We are especially focused on segmenting and identifying between handwriting and machine printed text because: 1) Handwriting in a document often indicates corrections, additions, or other supplemental information that should be treated differently from the main content and 2) the segmentation and recognition techniques requested for machine printed and handwritten text are significantly different. A novel aspect of our approach is that we treat noise as a separate class and model noise based on selected features. Trained Fisher classifiers are used to identify machine printed text and handwriting from noise and we further exploit context to refine the classification. A Markov Random Field-based (MRF) approach is used to model the geometrical structure of the printed text, handwriting, and noise to rectify misclassifications. Experimental results show that our approach is robust and can significantly improve page segmentation in noisy document collections.

/pdf/machine-printed-text-and-handwriting-identification-in-noisy-sd3e5xo6r7.pdf

Machine printed text and handwriting identification in noisy document images

A technique for image segmentation using shape-directed covers is described and applied to the fully automatic analysis of complex printed-page layouts. The structure of the background (white space) is analyzed, assisted by an enumeration of all maximal white rectangles. For this enumeration, the most computationally expensive step, an algorithm has been developed that, aside from a sort, achieves an expected runtime linear in the number of black connected components. The crucial engineering decision is the specification of a partial order on white rectangles to express domain-specific knowledge of preferred shapes and sizes. This order determines a sequence of partial covers of the background, and thus, a sequence of nested page segmentations. In experimental trials on Manhattan layouts, good segmentations often occur early in this sequence, using a simple and uniform shape-direction rule. This is a global-to-local strategy, which for some tasks is superior to strategies currently emphasized in the literature, including bottom-up and top-down. >

S.E. Jones

Papers

Image segmentation by shape-directed covers