- Official Masters eCampus
- eCampus Masters, Postgraduates and Courses
- Timetables and calendars
- Future Students
- Contact information
- About the TecnoCampus Park
- TecnoCampus Facilities
- HUB4T TecnoCampus
- Alumni
- TecnoCampus University Center
- Teaching staff
- Service for Quality and Teaching Innovation
- Library -CRAI
- Host your company
- Communication and press room
- TecnoCampus corporate image
- Events
- Transparency portal
- Equality Unit
- Supplier's Office
- Work with us
- TecnoCampus quality
What are you looking for?
Advanced online writing preprocessing techniques for device interoperability
TAPEO-ID
The aim of this project is to research, develop and apply techniques to improve the interoperability between different devices for acquiring handwritten signals online. Online acquisition refers to simultaneous digitization while the user is typing. This being the modern alternative to the classic systems where the user writes and then it is digitized.
Handwriting is probably one of the most complex tasks that human beings can perform. It can reveal the identity of the author who wrote a specific text, as well as pathologies such as Parkinson's disease, Alzheimer's, fatigue, dysgraphia, etc. Performing graphoscript tests is a simple and non-invasive decision support method. Especially in the case of diseases in which there is no biological marker. It can also be applied as an objective test to measure improvement after a specific treatment, such as oxygen therapy in patients affected by chronic obstructive pulmonary disease (COPD), or taking a certain medication.
Interoperability between different acquisition devices (each with its own characteristics) is desirable if we want to jump from laboratory conditions to real-world operations in e-safety and e-health based on handwriting tasks. Especially in the current context of explosion of new commercial devices and obsolescence of previous models. To be able to continue working with signals and models acquired in the past, a normalization of values provided by the digitizing tablets, elimination of saturations, etc. is necessary. For example, in the case of pressure, older devices provide pressure values between [0,1023] while current ones easily reach [0,4095] ranges. On the other hand, some provide information of the trajectory followed in the air when lifting the pen to make a new stroke, while others do not collect this signal.
Today there is an explosion of online handwriting acquisition devices (smartphones, tablets, PCs, etc.) each with their own specific features that are directly provided by the analog to digital converter and vary from one brand to another. The scientific community has focused on machine learning tools to improve recognition accuracy (especially in online signatures) as well as possible applications in e-security and e-health. However, the signal preprocessing step has not been addressed to its full potential and has usually been limited to dealing with training (or modeling) signals and test signals always acquired with the same device, has ignored pressure saturation etc. This proposal focuses on this line of research that has been fundamental in applications based on other signals (voice, fingerprint, face, etc.).
Additionally, advanced preprocessing techniques for device interoperability will allow an improvement of the state of the art in biometric recognition (e-health and e-security) and a reduction in the obsolescence of acquired devices and databases in the past with old systems.
Project team
Jiri Mekyska
Researcher / Work Team
Funding sources
Project funded by the Ministry of Science, Innovation and Universities MICIU/AEI /10.13039/501100011033 through the "2020 Call for R&D Projects, within the framework of the State Programs for the Generation of Knowledge and Scientific and Technological Strengthening of the R&D System D+i oriented to the Challenges of Society", File number PID2020-113242RB-I00.
Granted amount
Amount of the project
Contact the project team
Institution
Study
Become an entrepreneur
TecnoCampus Park Mataró-Maresme
Tel. +34 93 169 65 00
info@tecnocampus.cat
Av. Ernest Lluch, 32 | 08302 Mataró
© 2024 TecnoCampus. All rights reserved.
Web project developed by ACTIUM Digital