Melanie Gau

Enhancement of Multispectral Images of Degraded Documents by Employing Spatial Information

This work aims at enhancing ancient and degraded writings, which are captured by MultiSpectral Imaging systems. The manuscripts captured, contain faded out characters and are partly corrupted by mold and hardly legible. Several works have shown that such writings can be enhanced by applying unsupervised dimension reduction tools - like Principal Component Analysis (PCA) or Independent Component Analysis (ICA). In this work the Fisher Linear Discriminate Analysis (LDA) is applied in order to reduce the dimension of the multispectral scan and to enhance the degraded writings. Since Fisher LDA is a supervised dimension reduction tool, it is necessary to label a subset of multispectral data. For this purpose, a semi-automated label generation step is conducted, which is based on an automated detection of text lines. Thus, the approach is not only based on spectral information - like PCA and ICA - but also on spatial information. The method has been tested on two Slavonic manuscripts. A qualitative analysis shows, that the LDA based dimension reduction gains better performance, compared to unsupervised techniques.

Ancient document analysis based on text line extraction

In order to preserve our cultural heritage and for automated document processing libraries and national archives have started digitizing historical documents. In the case of degraded manuscripts (e.g. by mold, humidity, bad storage conditions) the text or parts of it can disappear. The remaining parts of the text can be segmented and the ruling can be extrapolated with the a priori knowledge. Since the ruling defines the position of the text within a page, it can be used for layout analysis and as a basis for the enhancement of the readability. Furthermore, information about the scribe (hand) of the manuscript, its spatiotemporal origin can be gained by analyzing the ruling. This paper presents an algorithm for ruling estimation of Glagolitic texts based on text line extraction and is suitable for degraded manuscripts by extrapolating the baselines with the a priori knowledge of the ruling. The algorithm was tested on 30 pages of the Missale Sinaiticum and the evaluation was based on visual criteria.

Multispectral image acquisition of ancient manuscripts

This paper presents image acquisition and readability enhancement techniques based on multispectral imaging. In an interdisciplinary manuscript and palimpsest research project an imaging system using a combination of LED illumination and spectral filtering was developed. On basis of the resulting multispectral image information the readability of the texts is enhanced and palimpsest texts are made visible by applying two different methods of Blind Source Separation, namely Principal Component Analysis and Independent Component Analysis.

Writer identification on historical Glagolitic documents

This work aims at automatically identifying scribes of historical Slavonic manuscripts. The quality of the ancient documents is partially degraded by faded-out ink or varying background. The writer identification method used is based on image features, which are described with Scale Invariant Feature Transform (SIFT) features. A visual vocabulary is used for the description of handwriting characteristics, whereby the features are clustered using a Gaussian Mixture Model and employing the Fisher kernel. The writer identification approach is originally designed for grayscale images of modern handwritings. But contrary to modern documents, the historical manuscripts are partially corrupted by background clutter and water stains. As a result, SIFT features are also found on the background. Since the method shows also good results on binarized images of modern handwritings, the approach was additionally applied on binarized images of the ancient writings. Experiments show that this preprocessing step leads to a significant performance increase: The identification rate on binarized images is 98.9%, compared to an identification rate of 87.6% gained on grayscale images.

A Clinical Tool for Automated Flow Cytometry Based on Machine Learning Methods

Clinical researchers working in flow cytometry (FCM) nowadays experience increasing demands to perform experiments that involve high throughput, rare event analysis and detailed immunophenotyping. Beckman Coulter and Becton Dickinson offer multi-use flow cytometry sorters that can analyze up to 70K EPS (events per seconds) with more than nine parameters enabled. While this multi-parametric feature provides a great power for hypothesis testing, it also generates a vast amount of data, which is analyzed manually through a processing called gating. For large experiments, this manual gating turns out to be time consuming and requires intensive operator training and experience. The lack of required expertise leads to wrong interpretation of data, thus a wrong therapy course for the case of patients with acute lymphoblastic leukemia (ALL) is followed. This paper aims to present a pipeline-software, as a ready-to-use machine learning based automated FCM assessment tool for the daily clinical practice for patients with ALL. The new system increases accuracy in assessment of FCM based on minimal residual disease (MRD) method in samples analyzed by conventional operator-based gating since computer-aided analysis potentially has a higher power due to the use of the whole multi-parametric FCM-data space at once instead of using methods restricted to two-dimensional decision rules. The tool is implemented as a telemedical network for analysis, clinical follow-up, treatment monitoring of leukemia and allows dissemination of automated FCM-MRD analysis to medical centres in the world.

Image Acquisition & Processing Routines for Damaged Manuscripts

This paper presents an overview of data acquisition and processing procedures of an interdisciplinary project of philologists and image processing experts aiming at the decipherment and reconstruction of damaged manuscripts. The digital raw image data was acquired via multi-spectral imaging. As a preparatory step we developed a method of foreground-background separation (binarisation) especially designed for multi-spectral images of degraded documents. On the basis of the binarised images further applications were developed: an automatic character decomposition and primitive extraction dissects the scriptural elements into analysable pieces that are necessary for palaeographic and graphemic analyses, writing tool recognition, text restoration, and optical character recognition. The results of the relevant procedures can be stored and interrogated in a database application. Furthermore, a semi-automatic page layout analysis provides codicological information on latent page contents (script, ruling, decorations).