A5031191307- Handwritten Text Recognition Techniques
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Image Processing and 3D Reconstruction
- CCD and CMOS Imaging Sensors
- Image and Object Detection Techniques
- Analog and Mixed-Signal Circuit Design
- Hand Gesture Recognition Systems
- Natural Language Processing Techniques
- Advancements in PLL and VCO Technologies
- Vehicle License Plate Recognition
- Authorship Attribution and Profiling
- Human Motion and Animation
- Digital Image Processing Techniques
- Radiation Effects in Electronics
- Topic Modeling
- Lexicography and Language Studies
- Advanced Vision and Imaging
- Algorithms and Data Compression
- Image and Video Stabilization
- Embedded Systems Design Techniques
- Video Analysis and Summarization
- Translation Studies and Practices
- Interconnection Networks and Systems
- Advancements in Semiconductor Devices and Circuit Design
Institute of Informatics & Telecommunications
2008-2020
National Centre of Scientific Research "Demokritos"
2011-2020
Democritus University of Thrace
2002-2012
National and Kapodistrian University of Athens
2012
Institute of Informatics of the Slovak Academy of Sciences
2007
Johns Hopkins University Applied Physics Laboratory
2002-2005
This paper presents the results of Handwriting Segmentation Contest that was organized in context ICDAR2013. The general objective contest to use well established evaluation practices and procedures record recent advances off-line handwriting segmentation. Two benchmarking datasets, one for text line word segmentation, were created order test compare all submitted algorithms as some state-of-the-art methods handwritten document image segmentation realistic circumstances. Handwritten images...
Purpose An overview of the current use handwritten text recognition (HTR) on archival manuscript material, as provided by EU H2020 funded Transkribus platform. It explains HTR, demonstrates , gives examples cases, highlights affect HTR may have scholarship, and evidences this turning point advanced digitised heritage content. The paper aims to discuss these issues. Design/methodology/approach This adopts a case study approach, using development delivery one openly available platform for...
The Handwriting Segmentation Contest was organized in the context of ICDAR2009 conference order to record recent advances off-line handwriting segmentation. This paper describes contest details including dataset, ground truth and evaluation criteria presents results 12 participating methods. includes handwritten document images produced by many writers several languages (English, French, German Greek). These are manually annotated produce which corresponds correct text line word segmentation...
An 11-bit time-to-digital converter (TDC) with high time resolution implemented in CMOS VLSI is presented. The TDC operates a wide and clock-adjustable of LSB = 50 ps to 1 ns, good power supply, temperature, environmental effects compensation. dead the measurement as low 0.5 /spl mu/s event rate can be MEvents/s. dissipation function clock frequency; dissipates <10 mW at an 100 kEvents/s LSB=100 ps. was incorporated complete time-of-flight (TOF) system on chip that addition included...
In this paper a complete OCR methodology for recognizing historical documents, either printed or handwritten without any knowledge of the font, is presented. This consists three steps: The first two steps refer to creating database training using set while third one refers recognition new document images. First, pre-processing step that includes image binarization and enhancement takes place. At second top-down segmentation approach used in order detect text lines, words characters. A...
Document digitization with either flatbed scanners or camera-based systems results in document images which often suffer from warping and perspective distortions that deteriorate the performance of current OCR approaches. In this paper, we present a goal-oriented rectification methodology to compensate for undesirable image aiming improve result. Our approach relies upon coarse-to-fine strategy. First, coarse is accomplished aid computationally low cost transformation addresses projection...
Writer identification is important for forensic analysis, helping experts to deliberate on the authenticity of documents. The ICDAR2013 Competition Identification part a competition series (see also ICDAR2011 and ICFHR2012 Contests) which dedicated record recent advances in field writer Latin scripts using established evaluation performance measures. benchmarking dataset was created with help 250 writers that were asked copy four parts text two based languages (English Greek). This paper...
ICDAR 2011 Writer Identification Contest is the first contest which dedicated to record recent advances in field of writer identification using established evaluation performance measures. The benchmarking dataset was created with help 26 writers that were asked copy eight pages contain text several languages (English, French, German and Greek). This paper describes details including measures used as well 8 submitted methods along a short description each method.
H-KWS 2014 is the Handwritten Keyword Spotting Competition organized in conjunction with ICFHR conference. The main objective of competition to record current advances keyword spotting algorithms using established performance evaluation measures frequently encountered information retrieval literature. comprises two distinct tracks, namely, a segmentation-based and segmentation-free track. Five (5) research groups have participated three (3) methods for track four (4) benchmarking datasets...
In order to achieve accurate text recognition performance for historical handwritten document images, robust and efficient page segmentation is necessary. this paper, we propose a zone detection followed by line method suitable documents. Our aim handle several challenging cases such as horizontal vertical rule lines overlapping with the text, two column documents characters of different touching vertically. For detection, analyze lines, connected components well white runs while...
Dewarping of camera document images has attracted a lot interest over the last few years since warping not only reduces readability but also affects accuracy an OCR application. In this paper, two-step approach for efficient dewarping is presented. At first step, coarse accomplished with help transformation model which maps projection curved surface to 2D rectangular area. The delimited by two lines fit top and bottom text along straight left right boundaries. second fine achieved based on...
The general objective of the ICFHR 2010 Handwriting Segmentation Contest organized in context conference was to use well established evaluation practices and procedures order record recent advances off-line handwriting segmentation. Two new benchmarking datasets, one for text line word segmentation, were created test compare algorithms handwritten document segmentation realistic circumstances. Handwritten images produced by many writers several languages (English, French, German Greek)....
Recognition of old Greek document images containing polytonic (multi accent) characters is a challenging task due to the large number existing character classes (more than 270) which cannot be handled sufficiently by current OCR technologies. Taking into account that system was used from late antiquity until recently, amount scanned documents still remains without full test search capabilities. In order assist progress relevant research, this paper introduces first publicly available...
Writer identification is important for forensic analysis, helping experts to deliberate on the authenticity of documents. The general objective ICFHR 2012 Identification Contest record recent advances in field writer using established evaluation performance measures. Challenge 1 contest deals specifically with Latin scripts. benchmarking dataset challenge was created help 100 writers that were asked copy four parts text two languages (English and Greek). This paper describes details this...
In this article, a method for segmentation-based learning-free Query by Example (QbE) keyword spotting on handwritten documents is proposed. The consists of three steps, namely preprocessing, feature extraction and matching, which address critical variations text images (e.g., skew, translation, different writing styles). During the step, sequence descriptors generated using combination zoning scheme novel appearance descriptor, referred as modified Projections Oriented Gradients....
In this paper, we present a novel approach for segmentation-based handwritten keyword spotting. The proposed relies upon the extraction of simple yet efficient descriptor which is based on projections oriented gradients. To end, global and local word image descriptors, together with their combination, are proposed. Retrieval performed using to euclidean distance between descriptors query segmented images. methods have been evaluated dataset ICFHR 2014 Competition Experimental results prove...
A CMOS time-of-flight (TOF) system-on-a-chip (SoC) for precise time-interval measurement with low power and high rate has been developed. Microchannel plate electron multipliers or photomultiplier tubes (PMTs) typically produce the start-stop delta-Ts of radiation events to be processed. The TOF chip includes two constant fraction discriminators (CFDs) a time-to-digital converter (TDC). CFDs interface start stop anodes through simple preamplifiers perform signal processing time walk...
The detection and correction of document skew is one the most important image analysis steps. ICDAR2013 Document Image Skew Estimation Contest (DISEC'13) first contest which dedicated to record recent advances in field estimation using well established evaluation performance measures on a variety printed images. benchmarking dataset that used contains 1550 images were obtained from various sources such as newspapers, scientific books dictionaries. contain figures, tables, diagrams,...
This paper presents the results of handwriting segmentation contest that was organized in context ICDAR2007. The aim this to use well established evaluation practices and procedures order record recent advances off-line segmentation. Two benchmarking datasets (one for text line one word segmentation) were used a common platform test compare all submitted algorithms handwritten document realistic circumstances. five by participants as two state-of-the-art are presented. performance method is...
In this paper, we present an off-line methodology for isolated Greek handwritten character recognition based on efficient feature extraction followed by a suitable vector dimensionality reduction scheme. Extracted features are (i) horizontal and vertical zones, (ii) the projections of profiles, (Hi) distances from boundaries (iv) profiles edges. The combination these types leads to 325- dimensional vector. At next step, technique is applied, according which dimension space lowered down...
One of the major issues in document image processing is efficient creation ground truth order to be used for training and evaluation purposes. Since a large number tools have trained evaluated realistic circumstances, we need quick low cost way create corresponding truth. Moreover, specific having correct text correlated with area line word level makes process difficult, tedious costly task. In this paper, introduce an transcript mapping technique ease construction segmentation that includes...
Scanning two book pages at the same time helps to accelerate scanning process but on other hand introduces several difficulties if user needs have one page per image. A major difficulty is appearance of noisy black borders around text areas as well stripes between pages. In this paper, we propose a novel algorithm for detecting frames double document images. Our aim split image into remove borders. First apply pre-processing which includes binarization, noise removal and smoothing. Then,...