search query: @supervisor Hollmén, Jaakko / total: 10
reference: 2 / 10
Author: | Hinneri, Samuli |
Title: | Visual Attention Detection of Gaussian Profile 2D Structures in Medical Imaging |
Publication type: | Master's thesis |
Publication year: | 2005 |
Pages: | 64 Language: eng |
Department/School: | Tietotekniikan osasto |
Main subject: | Informaatiotekniikka (T-122) |
Supervisor: | Hollmén, Jaakko |
Instructor: | Iivarinen, Jukka |
OEVS: | Electronic archive copy is available via Aalto Thesis Database.
Instructions Reading digital theses in the closed network of the Aalto University Harald Herlin Learning CentreIn the closed network of Learning Centre you can read digital and digitized theses not available in the open network. The Learning Centre contact details and opening hours: https://learningcentre.aalto.fi/en/harald-herlin-learning-centre/ You can read theses on the Learning Centre customer computers, which are available on all floors.
Logging on to the customer computers
Opening a thesis
Reading the thesis
Printing the thesis
|
Location: | P1 Ark Aalto | Archive |
Keywords: | visual attention foveation region content-based compression JPEG2000 VI area of the brain blood vessel segmentation diabetes treatment visuaalinen huomio silmien fovea-alue sisältöpohjainen kuvanpakkaus JPEG2000 aivojen VI-alue verisuonten segmentointi diabeteksen hoito |
Abstract (eng): | A new approach to automatic visual attention detection is presented. It is based on detecting irregularities in gradient flow fields with a model which calculates 3 innovative measures for each pixel and using this information concludes which measure combinations are likely to be in salient areas. Studies with test images give good results, especially for tubular structure detection. The most relevant application is blood vessel segmentation, others include content-based image compression, brain research and visual attention modeling. The main line of thought is that the proposed model could be used as a response function in blood vessel detection, but to reach verifiably state-of-the-art would require further research. |
Abstract (fin): | Esitellään uusi lähestymistapa visuaalisten huomionkohteiden havaitsemiseksi. Se perustuu gradienttivirtauskenttien epäsäännöllisyyksien huomioimiseen mallilla, joka laskee 3 innovatiivista mitta-arvoa kullekin pikselille ja päättelee tästä tiedosta, mitkä arvokombinaatiot todennäköisimmin sijaitsevat kiintoisilla alueilla. Kokeet antoivat hyviä tuloksia, erityisesti tubulaaristen rakenteiden havainnointiin. Pätevin sovellusalue on verisuonten segmentointi; muihin kuuluu sisältöpohjainen kuvanpakkaus, aivotutkimus ja visuaalisen attention mallittaminen. Johtoajatuksena on, että esiteltyä menetelmää voisi käyttää vastefunktiona verisuonten segmentoinnissa, mutta alan huipulle nouseminen todistettavalla tavalla vaatisi jatkotutkimuksia. |
ED: | 2006-01-03 |
INSSI record number: 30449
+ add basket
INSSI