Suchergebnisse

Document preprocessing is the first step of document image analysis systems and comprise a skew estimation of documents. The gradient information can be used to estimate the skew due to the characteristic of latin characters and specific fonts which are composed of horizontal and vertical strokes. This paper shows the accuracy of a gradient orientation measure on synthetic data and for printed text documents. Also the difference of the accuracy on a binarized and grayvalue dataset is presented. ; European Union's Horizon 2020

The final publication is available via https://doi.org/10.1109/ICFHR-2018.2018.00105 . ; MultiSpectral Imaging enhances the study of degraded historical documents. It allows for visualizing washed out or even invisible ink but also improves the automated analysis because of a denser spectral sampling. We present a new methodology for binarization of multispectral document images that groups spectral signatures of different sources by fitting two Gaussian Mixture Models (GMMs) with Expectation Maximization. Both GMMs assign cluster labels to the multispectral samples and the clustering results are combined for the identification of the handwriting regions. The method is evaluated on the ICDAR 2015 MS-TEx dataset. Results on this publicly available benchmarking set are encouraging. ; European Union's Horizon 2020

The final publication is available via https://doi.org/10.1145/3151509.3151526 . ; Digital copies of historical documents are needed for the Digital Humanities. Currently, cameras of standard mobile phones are able to capture documents with a resolution of about 330 dpi for document sizes up to DIN A4 (German standard, 297 x 210 mm), which allows a digitization of documents using a standard device. Thus, scholars are able to take images of documents in archives themselves without the need of book scanners or other devices. This paper presents a scanning app, which comprises a real time page detection, quality assessment (focus measure) and an automated detection of a page turn over if books are scanned. Additionally, a portable device - the ScanTent - to place the mobile phone during scanning is presented. The page detection is evaluated on the ICDAR2015 SmartDoc competition dataset and shows a reliable page detection with an average Jaccard index of 75%. ; European Union's Horizon 2020

The final publication is available via https://doi.org/10.1109/ICDAR.2017.222 . ; The cBAD competition aims at benchmarking state-of-the-art baseline detection algorithms. It is in line with previous competitions such as the ICDAR 2013 Handwriting Segmentation Contest. A new, challenging, dataset was created to test the behavior of state-of-the-art systems on real world data. Since traditional evaluation schemes are not applicable to the size and modality of this dataset, we present a new one that introduces baselines to measure performance. We received submissions from five different teams for both tracks. ; European Union's Horizon 2020

The final publication is available via https://doi.org/10.1109/ICFHR-2018.2018.00046 . ; In this paper we present a template-based table structure matching using association graphs for handwritten/printed historical documents. The recognition of the table structure consisting of column and header information is the prerequisite for the subsequent row detection and handwritten text recognition used for information extraction. The table matching is done by detecting the maximum clique in an association graph, which represents the matching of the line information of the template and a document of interest. This allows for variations of widths and heights of rows and columns. The presented methodology is evaluated on historical register books (death records) of the Archive of the Diocese of Passau. The method shows a reliable detection of the structure of handwritten/printed tables with a mean cell match of 88.28%. ; European Union's Horizon 2020

The final publication is available via https://doi.org/10.1109/ICDAR.2017.225 . ; The ICDAR 2017 Competition on Historical Document Writer Identification is dedicated to record the most recent advances made in the field of writer identification. The goal of the writer identification task is the retrieval of pages, which have been written by the same author. The test dataset used in this competition consists of 3600 handwritten pages originating from 13th to 20th century. It contains manuscripts from 720 different writers where each writer contributed five pages. This paper describes the dataset, as well as the details of the competition. Five different institutions submitted six methods which were ranked using identification and retrieval metrics. The paper describes the competition details including the dataset, the evaluation measures used as well as a short description of each submitted method. ; European Union's Horizon 2020