I Face Detection And Recognition System author: Mukund Agarwal supervisor: Professor Nishan Canagarajah Project Thesis submitted in support of the Degree of Bachelor of Engineering in Electronic and Communications Engineering


Table 2 Detection rate for various numbers of false positives on the MIT+CMU test set



Download 2,74 Mb.
Pdf ko'rish
bet12/35
Sana16.06.2022
Hajmi2,74 Mb.
#677052
1   ...   8   9   10   11   12   13   14   15   ...   35
Bog'liq
Thesis

Table 2 Detection rate for various numbers of false positives on the MIT+CMU test set 
containing 130 images and 507 faces. (Published in 
8

Table 2 shows the detection rate results of the different algorithms. From this we can clearly 
say that ‘Voila and Jones’ method also provides reasonably high detection rate. 
After analyzing each algorithm’s complexity, Voila & Jones seems a better choice. It offers a 
high detection rate combined with a very low processing time which is what the system needs. 
7.2
5.0
0.067
0
1
2
3
4
5
6
7
8
Neural network based 
face detection 
Image pyramid 
statistical method 
Voila and Jones method 
Tim

(s
)
Time taken by each method
Time taken


Mukund Agarwal 
Face Detection & Recognition System 
11 
4.1.3 Voila and Jones method theory 
This method uses ‘Haar’ wavelets for feature extraction from the images. These wavelets also 
allow feature evaluation.
A & B – Edge features 
C – Line features 
D – Four rectangle Features 
They are formed of one low interval and high interval or in other words are single wavelength 
square waves. A square wave is a pair of one light and one dark adjacent rectangles. The 
calculation of these wavelets is relatively easy as the white areas are just subtracted from the 
black ones. Figure 8 shows the four basic types of Haar wavelets in 2D.
Figure 9 Haar wavelets extracting features 
The feature extraction is made faster by integral image which is a special representation of the 
image. A machine learning method, called ‘AdaBoost’ enables classifier training and feature 
selection. All of the detected features are then combined efficiently by using a cascaded 
classifier. This is shown in figure 10.
 

Download 2,74 Mb.

Do'stlaringiz bilan baham:
1   ...   8   9   10   11   12   13   14   15   ...   35




Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©hozir.org 2024
ma'muriyatiga murojaat qiling

kiriting | ro'yxatdan o'tish
    Bosh sahifa
юртда тантана
Боғда битган
Бугун юртда
Эшитганлар жилманглар
Эшитмадим деманглар
битган бодомлар
Yangiariq tumani
qitish marakazi
Raqamli texnologiyalar
ilishida muhokamadan
tasdiqqa tavsiya
tavsiya etilgan
iqtisodiyot kafedrasi
steiermarkischen landesregierung
asarlaringizni yuboring
o'zingizning asarlaringizni
Iltimos faqat
faqat o'zingizning
steierm rkischen
landesregierung fachabteilung
rkischen landesregierung
hamshira loyihasi
loyihasi mavsum
faolyatining oqibatlari
asosiy adabiyotlar
fakulteti ahborot
ahborot havfsizligi
havfsizligi kafedrasi
fanidan bo’yicha
fakulteti iqtisodiyot
boshqaruv fakulteti
chiqarishda boshqaruv
ishlab chiqarishda
iqtisodiyot fakultet
multiservis tarmoqlari
fanidan asosiy
Uzbek fanidan
mavzulari potok
asosidagi multiservis
'aliyyil a'ziym
billahil 'aliyyil
illaa billahil
quvvata illaa
falah' deganida
Kompyuter savodxonligi
bo’yicha mustaqil
'alal falah'
Hayya 'alal
'alas soloh
Hayya 'alas
mavsum boyicha


yuklab olish