A5066336064- Handwritten Text Recognition Techniques
- Natural Language Processing Techniques
- Image Processing and 3D Reconstruction
- Text and Document Classification Technologies
- Topic Modeling
- Scheduling and Timetabling Solutions
- Image Retrieval and Classification Techniques
- Web Data Mining and Analysis
- Advanced Software Engineering Methodologies
- Scheduling and Optimization Algorithms
- Algorithms and Data Compression
- Infrastructure Maintenance and Monitoring
- Vehicle License Plate Recognition
- Usability and User Interface Design
- Distributed systems and fault tolerance
- Digital and Cyber Forensics
- Advanced Text Analysis Techniques
- Service-Oriented Architecture and Web Services
- Structural Health Monitoring Techniques
- Image and Object Detection Techniques
- Mathematics, Computing, and Information Processing
- Embedded Systems and FPGA Design
- Interactive and Immersive Displays
- Innovative Human-Technology Interaction
- Caching and Content Delivery
Budapest University of Technology and Economics
2013-2024
Andrássy University Budapest
2007
This article presents SVC-onGoing1, an on-going competition for on-line signature verification where researchers can easily benchmark their systems against the state of art in open common platform using large-scale public databases, such as DeepSignDB2 and SVC2021_EvalDB3, standard experimental protocols. SVC-onGoing is based on ICDAR 2021 Competition On-Line Signature Verification (SVC 2021), which has been extended to allow participants anytime. The goal evaluate limits popular scenarios...
Several breakthroughs have appeared in different applications due to deep learning which can be applied Bridge Weigh-In-Motion (BWIM) systems as well. Therefore, should examined BWIM systems. However, some learning-based solutions already been proposed the international literature, they cannot compared static methods because available datasets are not appropriate train and test artificial neural networks. In addition, numerous other aspects considered during tests. current paper, a...
This paper presents a novel approach for off-line signature verification. First the different aspects and problems of verification are discussed in conjunction with analysis methods. It is shown, that on-line methods perform usually better than because they can make use temporal information (and thereby get perception semantics signature). An assumption made, have difficulties improving their results high level abstraction use, which direct consequence missing semantics. Afterwards solution...
Abstract Online signature verification considers signatures as time sequences of different measurements the signing instrument. These signals are captured on digital devices and therefore consist a discrete number samples. To enrich or simplify this information, several verifiers employ resampling interpolation preprocessing step to improve their results; however, design decisions may be difficult generalize. This study investigates direct effect sampling rate input accuracy online systems...
Signature verification is an important emerging field.New results get published both in the online and offline areas.It's often hard to compare because researchers use different databases.Although there are some publicly available databases their quality size may vary.This paper aims deliver a comprehensive list of most significant field.Using our around world can chose best common database that fits needs.
Object-oriented design patterns offer a structural and behavioral solution for given problem in context using basic building blocks. They are widely used software engineering as they promote extensibility, testability, maintainability reusability. However, besides their use, detection also has great significance the industry. This paper first presents general, programming language neutral model architecture automated pattern object-oriented source code. To validate recognition-capability of...
Previous studies have shown that choosing a different sample rate or signature point count provides better accuracy in online verification systems. However, the sampling minimizes error may vary on database and signer levels. In this work, we studied effect of individual frequencies for each proposed system based signer-dependent frequency. The was tested five databases, using several features preprocessing methods. Results showed improvement 70% overall 500 tests 92% chosen where...
Abstract Amongst different approaches, dynamic time warping has shown promising results during the online signature verification competitions of previous years. To improve warping, preprocessing steps may be applied and dimensions samples compared. The choice comparing significantly influence results. Thus, to aid researchers with these decisions, a comparison made between algorithms as horizontal scaling, vertical scaling alignment using in their combinations on two datasets (SVC2004...
In this paper we present a number of available serialization libraries for the .NET platform, both integrated and 3rd party solutions analyze compare their performance. Through analysis comparison aim to deduct how process can be further optimized which parts act as bottleneck. The also gives insight into data exchange format itself affects whole process.
The automatic generation of schedules has been in focus researches for decades.Since small changes the input have exponential impact on tested state space, methods based heuristics and artificial intelligence are most successful.In our paper we analysed opportunities academic final exam scheduling, which is a special subtask schedule generation.We examined problem with genetic algorithm approach.We created point system, wherewith goodness generated measurable comparable.We effect achievable...
Online signature verification is a currently evolving field. Numerous approaches are proposed each year that claimed to be capable of delivering improved results in some aspects over others. These systems usually based on the experiences author(s), and often limited single databases. factors make it difficult compare or even reproduce results. The lack negative evaluation design choices creation new, online difficult. In this work, we addressed problem by conducting systematic most common...
One of the key tasks in computer vision is identification and matching coherent shapes different images. In signature verification these are represented by strokes signature. Strokes may fall apart, close up, change or simply disappear among signatures same signer. This paper proposes an algorithm to tackle stroke problem using area based method for finding linking corresponding dynamic time warping perform intra matching. Using above two dimensional image comparison can be reduced a one...
Mobile devices and mobile applications have a significant effect on the present future of software industry. The diversity platforms necessitates development same application for all major platforms, which requires considerable effort. developers are multiplatform developers, but they prioritize therefore, not equally important them. Appropriate methods, processes tools required to support in order achieve better productivity. main motivation our research activity is provide method,...
There are several classification algorithms used for signature verification purposes. The k-nearest neighbor (KNN) algorithm was previously in online verification, but this paper, we present an evaluation of the using JKNN classifier, which is a generalized case KNN classifier. An optimal classifier to evaluate algorithm's main parameters that provide most accurate results. each parameter presented and tested SVC2004 database. Our results show can by optimizing algorithm. Both false...
In this paper, we propose an enhanced jk-nearest neighbor (jk-NN) classifier for online signature verification. After studying the algorithm's main parameters, use four separate databases to present and evaluate each algorithm parameter. The results show that proposed method can increase verification accuracy by 0.73-10% compared a traditional one class k-NN classifier. has achieved reasonable different databases, 3.93% error rate when using SVC2004 database, 2.6% MCYT-100 1.75% SigComp'11...