The design and functionality of a versatile search engine on handwritten documents is described. Documents are indexed using global image features, e.g., stroke width, slant, word gaps, as well local features that describe shapes of characters and words. Image indexing is done automatically using page analysis, page segmentation, line separation, word segmentation and recognition of characters and words. Several types of searches are

Document storage and retrieval capabilities of the CEDAR-FOX forensic handwritten document examination system are described. The system is designed for automated and semi-automated analysis of scanned handwritten documents. For library creation, the system provides functionalities for (i) entering document metadata, e.g., identification number, writer and other collateral information, (ii) creating a textual transcript of the image content at the word level, and (iii) including automatically extracted document level features, e.g, stroke width, slant, word gaps, as well as finer features that capture the structural characteristics of characters and words. For extracting these features the system performs page analysis, page segmentation, line separation, word segmentation and finally recognition of characters and words. The extracted features are used for writer identification by matching against a library built as a database. The system design is driven by questioned document examination with its emphasis on writer identification. Several query modalities are permitted for retrieval: (i) document level: the entire document image is the query; (ii) partial image: a region of interest (ROI) of a document; (ii) word image: which is also called word

The objective of this paper is to present a number of features that can be extracted from handwritten digits and used for author verification or identification of a person’s handwriting. The features under consideration are mainly computational features some of which cannot be easily evaluated by humans. On the other hand, these features can be extracted by computer algorithms with a high degree of accuracy. The eleven features used are described. All features were appropriately binarized so that binary feature vectors of constant lengths could be formed. These vectors were then used for author discrimination, using the Hamming distance measure. For this task a writer database consisting of 15 writers was created. Each writer was asked to write random strings of 0 to 9 at least 10 times. The results indicate that the combined features work well at discriminating writers and warrant further detailed investigation. Although the set of features was designed for dealing with handwritten digits (as may be written on cheques), it may also be used for isolated alphabetic characters. 1.

In some investigations of digital crime, the question of who was at the keyboard when incriminating documents were produced can be legitimately raised. Authorship attribution can then contribute to the investigation. Authorship methods which focus on linguistic characteristics currently have accuracy rates ranging from 72 % to 89%, within the computational paradigm. This article presents a computational, stylometric method which has obtained 95 % accuracy and has been successfully used in investigating and adjudicating several crimes involving digital evidence. The article concludes with a brief review of the current admissibility status of authorship identification techniques. Section 1: The Need and Available Methods In the investigation of certain crimes involving digital evidence, when a specific machine is identified as the source of documents, a legitimate question is, “Who was at the keyboard when the relevant documents were produced? ” For example, consider the following scenarios, drawn from actual cases. 1. A government employee wrote e-mails to his supervisor in which he disparaged

Since handwriting is influenced by physiology, training and other behavioral factors, a study of the handwriting of twins can shed light on the individuality of handwriting. The paper describes the methodology and results of such a study where handwriting samples of twins were compared by an automatic handwriting verification system. The results complement that of a previous study where a diverse population was used. The present study involves samples of 206 pairs of twins, where each sample consisted of a page of handwriting. The verification task was to determine whether two half-page documents (where the original samples were divided into upper and lower halves) were written by the same individual. For twins there were 1236 verification cases – including 824 tests where the textual content of writing was different and 412 tests where it was the same. An additional set of 1,648 test cases were obtained from handwriting samples of non-twins (general population). In order to make the handwriting comparison, the system computed macro features (overall pictorial attributes), micro features (characteristics of individual letters) , and style features (characteristics of whole-word shapes and letter pairs). Four testing scenarios were evaluated: twins and non-twins writing the same text and writing different texts. Results of the verification tests

In this paper a study of some structural features of handwritten letter `a' is presented. The features under consideration are structural, style and formation features document examiner's use when studying a questioned document to determine its authorship or authenticity.

Abstract — This paper describes two approaches to approximation of handwriting strokes for use in writer identification. One approach is based on a thinning method and produces raster skeleton whereas the other approximates handwriting strokes by cubic splines and produces a vector skeleton. The vector skeletonisation method is designed to preserve the individual features that can distinguish one writer from another. Extraction of structural character-level features of handwriting is performed using both skeletonisation methods and the results are compared. Use of the vector skeletonisation method resulted in lower error rate during the feature extraction stage. It also enabled to extract more structural features and improved the accuracy of writer identification from 78 % to 98 % in the experiment with 100 samples of grapheme “th ” collected from 20 writers. I.

einer Unterschrift bezogen auf ein Muster zu bestimmen. Einen guten Uberblick uber das Gebiet der automatischen Signaturverifikation findet man u.a. in [1]) und [2]. Generell k onnen Systeme zur Identifizierung von Handschriften in solche, die die Daten direkt von Eingabeger aten erhalten (OnLine) und jene, die die Identifizierung auf der Grundlage von im Graphikformat vorliegenden Daten durchf uhren (Off-Line), eingeteilt werden. Obwohl die Anwendungen und Methoden beider Systeme unterschiedlich sind, ist die generelle Taxonomie zur Identifikation von Signaturen ahnlich. Die Untersuchungen im vorliegenden Artikel konzentrieren sich auf eine OffLine -Trennung von Datum und Signum, d.h. die Festlegung von Datums- und Signumsbereichen f ur einen vorliegenden Beleg. Eine naheliegende Zeichenerkennung stellt sich als auerst schwierig dar, da einzelne Zeichen im Signum so gut wie nicht detektierbar sind. Einers

The analysis of handwritten documents from the viewpoint of determining their writership has great bearing on the criminal justice system. In many cases, only a limited amount of handwriting is available and sometimes it consists of only numerals. Using a large number of handwritten numeral images extracted from about 3000 samples written by 1000 writers, a study of the individuality of numerals for identification/verification purposes was conducted. The individuality of numerals was studied using cluster analysis. Numerals discriminability was measured for writer verification. The study shows that some numerals present a higher discriminatory power and that their performances for the verification/identification tasks are very different.

This paper discusses about various image processing techniques and tools which are available for identification of printing technologies. Printing technology identification and associated problems in document forensics have been projected as challenges in image processing application. Various image processing approaches based on textures, spatial variation, HSV color space, spatial correlation, and feature based on histogram and some of the pattern recognition methods, like gray level cooccurrence matrix, roughness of the text, perimeter of edge are highlighted. This paper devotes more on one of the recent contribution, namely, Gaussian Variogram Model (GVM) for printer classification. General Terms Image processing, Document forensics-printing technique classification, spatial statics