Text Lines and Snippets Extraction for 19th Century Handwriting Documents Layout Analysis

Abstract

In this paper we propose a new approach to improve electronic editions of human science corpus, providing an efficient estimation of manuscripts pages structure. In any handwriting documents analysis process, the text line segmentation is an important stage. The presence of variable inter-line spaces, of inconstant base-line skews, overlapping and occlusions in unconstrained ancient 19th handwritten documents complexifies the text lines segmentation task. In this paper, we only use as prior knowledge of script the fact that text lines skews can be random and irregular. In that context, we model text line detection as an image segmentation problem by enhancing text line structure using Hough transform and a clustering of connected components so as to make text line boundaries appear. The proposed approach of snippets decomposition for page layout analysis lies on a first step of content pages classification in five visual and genetic taxonomies, and a second step of text line extraction and snippets decomposition. Experiments show that the proposed method achieves high accuracy for detecting text lines in regular and semi-regular handwritten pages in the corpus of digitized Flaubert manuscripts ("Dossiers documentaires de Bouvard et Pécuchet", 1872-1880).

Arabic handwritten text line extraction using connected component analysis from a multi agent perspective

“…Other methods exist, as the k-means algorithm [11], the Hough transform [12,13] and active contour technique [14].…”

Adaptive Script-Independent Text Line Extraction

“…11 (c), the proposed method has correctly segmented the text lines. Figure 12 (a) [4] shows the result of the piece-wise projection method [4] 13 Segmentation results on a difficult Korean document image of the UMD dataset produced by (a) Mumford-Shah model [18], (b) PDF-based method [16], and (c) the proposed method. Shah model [18] and PDF-based [16] method use horizontal blurring filters.…”

Text area detection in handwritten documents scanned for further processing

“…The denoting of the text space is performed by a tracking script to create a curvilinear separation path between each pair of subsequent text lines, which in result leads to finding the separate text fragments. In [8] the text area is identified by first using image binarization and later separating the graph of connected components (CC) with segmentation methods. In a next step, Hough transform is applied to define each connected component and to calculate the distance vector for each graph component, resulting in designating the external edge blocks.…”