Evaluation of facial detection using adaboost and morphology / Maisarah Aripin
Face detection is a very important process and become part of face recognition. People are slightly aware of its significant and sometimes camiot really determine a face in an image before recognize it. This situation usually happened to the beginner. In recent years, there are a lot of researche...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Faculty of Computer and Mathematical Sciences
2007
|
| Subjects: | |
| Online Access: | http://ir.uitm.edu.my/id/eprint/1539/1/TD_MAISARAH%20ARIPIN%20CS%2007_5%20P01.pdf http://ir.uitm.edu.my/id/eprint/1539/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Face detection is a very important process and become part of face recognition. People
are slightly aware of its significant and sometimes camiot really determine a face in an
image before recognize it. This situation usually happened to the beginner. In recent
years, there are a lot of researches on face detection method. Thus, a lot of techniques
have been discovered in order to get a better technique and improvement in this area of
study. Each of those techniques has its own effectiveness in face detection, For example
knowledge based method that able to work well for face localization in uncluttered
background. Beside that, there is also another method that has the capability to run in
real time such as Adaboost. However, this research is only focusing on two techniques,
which are Adaboost and Morphology. Both of these algorithms are capable in face
detection but serve up in different way. Thus, the objective of this research is to evaluate
face detection tool using Adaboost and Morphology and to find out the effectiveness of
both algorithms. The system that has been used to test the images is Kihwan's Face
Detector. Analysis shows that this system is good in detecting large size images using
Adaboost After sizes of the images have been standardizing to become 400 x 300
pixels, it shows that this system can performs better by using Morphology as the
technique. |
|---|