3. Introduction
Personalized and intelligent use of appliances for the
security purpose are necessities in our life today.
These appliances tend to be special-purpose, limited-
resource, network-connected devices, such as Cell
phones.
A low cost intelligent wireless security and
monitoring solution using moving object recognition
technology is presented.
5. Java Editions
Java 2 Platform
Java2 Java2 Java2
Standard Edition Enterprise Edition Micro Edition
(J2SE™) (J2EE™) (J2ME™)
Standard desktop & Heavy duty server Small & memory
workstation applications systems constrained devices
6. Java Editions
Each edition defines different sets of class
libraries.
J2ME provides a robust, flexible environment
for application running on a broad range of
other deceives
J2EE
J2SE
J2ME
7. J2ME Core Concepts
Configuration
Minimum platform
J2ME
required for a Profile
group of devices
Profile J2ME
Addresses specific Libraries
needs of a certain
Java Language
device family
Optional Packages Java Virtual Machine
Set of APIs in support of
additional, common Host Operating System
behaviors. E.g. Mobile
Media API.
9. Statistics
Mean
Center of gravity of the object
N N
1 1
x mean xi y mean yi
N is the number
N i 1
N i 1 of object pixels
Variance
The Variance measures the variations of the object-pixels’
positions around the center of gravity
N N
1 2 1 2
x var ( xi x mean ) y var ( yi y mean )
N i 1 N i 1
10. Statistics
Standard deviation: sigma ( ) x sigma x var
How to use it
”Automatic” thresholding based on statistics
Example: the color of the hand
Algorithm:
if: THmin < pixel < Thmax
then: hand pixel
else: non-hand pixel
Training
Average color of hand: mean
Variations in the color of the hand: variance =>
Use statistics: THmin = mean-2 and THmax =
mean+2
11. Segmentation in Video
Videos are Image Sequences over Time
x
• 10 Images
• An image is a function
t f ( x, y , t ) ft ( x , y )
y • At each time step two
have an image f ( x , y )
• Frame rate = the number
of images per second
12. Segmentation using Motion
Assuming that only the object is moving =>
motion can be used to find the object
Motion detection
We are using Background subtraction algorithm to
detect moving object.
14. Background Subtraction
Uses a reference background image for
comparison purposes.
Current image (containing target object) is
compared to reference image pixel by pixel.
Places where there are differences are detected
and classified as moving objects.
Motivation: simple difference of two images
shows moving objects
15. a. Original scene b. Same scene later
Subtraction of scene a from scene b Subtracted image with threshold of 100
16. Background Subtraction
Foreground is moving, background is stable
Algorithm
1. Capture image containing background
2. Capture N images and calculate the average
background image ( Background template)
3. Subtract image (difference = motion)
4. Threshold
5. Delete noise
18. ALERT RECEIVED
Alert sent to predefined number
System sends alert (SMS, MMS)
19. System Alert
When the system detect moving object, it sets
the alert on the Cell phone
The system create Message connection,
gathering the information required ( address ,
message text..)
Then the message is sent to notify the user
20. System architecture
If the difference
between real-time Real Time
frame and template Frame Capture
reaches predefined
threshold, moving Background
Subtraction Algorithm
object are considered
to appear
SMS Alert / MMS
Alert
22. Advantages
1. Low cost surveillance system
2. Little memory consumption
3. Easy to operate
4. Mobility is presented
5. Available to wide rang of mobile phones
23. conclusion
This approach is to develop a system which
will enable the user to apply security with
minimum cost and affords. The system should
be able to detect any theft action and alert the
user in minimum time.
24. References
IEEE DOI 10.1109/CISP.2008.235
The complete reference J2ME, McGRAW
HILL,2006
wifiplanet.org.
http://ieeexplore.ieee.org