Face Extraction from Image with Weak Cascade Classifier

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61988987%3A17310%2F14%3AA1501B2H" target="_blank" >RIV/61988987:17310/14:A1501B2H - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Face Extraction from Image with Weak Cascade Classifier
Original language description
The aim of this paper is to propose an artificial vision-based face detection approach, which could be primarily used in robotics. Three main problems arise from this expectation. The first one is the computation time of the whole process. The second oneis the quality of the input information due to a camera with low resolution. The third one is the robustness of the involved techniques regarding the implementation. The paper discusses all three problems in the first part and introduces the Haar Cascade theory. The second part of the paper proposes a new noise reduction approach to improve detection result mostly in eyes and mouth area. Next part of the paper shows experimental results and finds the best threshold parameter to minimize overlapping areas. The last part explains advantages of the proposed technique.
Czech name
—
Czech description
—

Publication year
2014
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Modern Trends and Techniques in Computer Science, AISC 285
ISBN
978-3-319-06739-1
ISSN
2194-5357
e-ISSN
—
Number of pages
11
Pages from-to
495-505
Publisher name
Springer Verlag
Place of publication
Switzerland
Event location
UTB Zlín
Event date
Apr 28, 2014
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)