New mathematical and algorithmic schemes for pattern classification with application to the identification of writers of important ancient documents

Abstract

In this paper, a novel approach is introduced for classifying curves into proper families, according to their similarity. First, a mathematical quantity we call plane curvature is introduced and a number of propositions are stated and proved. Proper similarity measures of two curves are introduced and a subsequent statistical analysis is applied. First, the efficiency of the curve fitting process has been tested on 2 shapes datasets of reference. Next, the methodology has been applied to the very important problem of classifying 23 Byzantine codices and 46 Ancient inscriptions to their writers, thus achieving correct dating of their content. The inscriptions have been attributed to ten individual hands and the Byzantine codices to four writers.

Image-based historical manuscript dating using contour and stroke fragments

“…Therefore, an undated document can be dated by identifying the writer. This is reasonable because if we know the writer and his active period, the date of the document can be directly obtained [3,4]. We conduct experiments on writer identification on the MPS data set as well as historical document dating by handwriting style identification.…”

“…Increasing numbers of such documents are currently being digitized and stored in the computer, as in the Monk system [1], which contains more than 100K scanned page images. Thanks to this development, pattern recognition techniques can now be applied to solve historical document problems, which has already been attempted at length in the case of writer identification [2][3][4][5], word spotting [6,7] and character recognition [8,9]. These methods aim to provide efficient tools for scholars in the humanities to discover informative patterns in large digital collections.…”

Improving Arabic writer identification using score‐level fusion of textural descriptors

“…Analysis of handwriting has remained an interesting research area for many decades attracting forensic document examiners (FDEs), palaeographers, psychologist, and neurologists. In addition to the classical problem of handwriting recognition [1,2], handwriting has been studied for problems like characterisation of writer [3][4][5][6] and writers' demographics [7,8], signature verification [9][10][11], classification of writing styles [12][13][14], and manuscript dating [15]. Likewise, handwriting has also been investigated to predict personal [16,17] and behavioural [18,19] attributes.…”

Gradient‐based approach to offline text‐independent Persian writer identification

“…Saad used fuzzy logic in combination with a genetic algorithm to improve the identification rate. In [9], a combination of plane curvature variation as a feature with two affine transformations was used for matching character shapes. By using these transformations, a character shape fits best to another and the error of this fitting was used for classification.…”