You are on page 1of 6

AUTOMATED INTELLIGENT ROBOT FOR SURVEILLANCE

AND SECURITY
K.SANTOSH, R.SRINIVAS, II YEAR, DEPT. OF EEE
SRI SAIRAM ENGINEERING COLLEGE.

Abstarct
Our paper aims at the continuous INTRODUCTION
monitoring of a particular area by AERIAL
PHOTOGRAPHY. The helicopter termed The basic process involving in this
as flying robot carries the CMOS camera system is the detection of an object by the
at a minimum height of 30m and is robot (the helicopter unit), sending the
airborne. The monitored images are images to a substation, where the image is
simultaneously transmitted to the ground processed, and analyzed. Then the detected
substation via wireless video object is given as a signal to one or many
communication. The image is then other stations.
processed in the substation for the Hence the process is divided into 3
presence of human being. This is achieved units-
by using the image processing techniques 1. MAINSTATION
like “GRAY SCALE EROSION FILTER 2. SUBSTATION
and EDGE DETECTION ALGORITHM”. 3. HELICOPTER
Once the human is detected, signals are
sent to the nearest defence camp (main UNIT- ROBOT.
station) in the form of text message to
mobile device using GSM communication, Working: HARDWARE
along with a signal to LCD to flash
appropriate message .The key features of The hardware components that are
this project is that it is low cost, light present in the main station are 10 keys, PIC
weight, high speed that it takes only 2 16F877A microcontroller, MAX 232, Liquid
seconds to complete the process, no noise Crystal Display and a mobile. The 10 keys
disturbance and an efficient autonomous present are nothing but push buttons that are
system. used for selecting the next command to be
executed. The commands include selecting
the substation or selecting the category that
TECHNOLOGY INVOLVED is to be detected. The micro controller used
HARDWARE is a 40 pin PDIP package PIC 16F877A.
This PIC is burnt with the process sequence
1. EMBEDDED SYSTEM program that has been written by us. The
2. RF COMMUNICATION program is given in the appendix. The MAX
3. GSM-WIRELESS
232 is used for the CMOS TTL logic
COMMUNICATION. conversion between the PIC and the mobile.
A 5x8 dots with cursor LCD is employed
SOFTWARE here to display the current status of the
1. IMAGE process like ‘Substation1 selected’, ‘human
PROCESSING. detect, etc. The mobile is used here for the
wireless full duplex communication between
the main station and the substations. It uses
the GSM protocol for this purpose.
OVERALL BLOCK DIAGRAM
is received from that substation via mobile
and the same is displayed in the LCD.
The hardware components present
Working: HARDWARE in the substation are a Personal computer, a
The process starts as preprogrammed MAX 232, two PIC 16F877A, an ULN
in the PIC. When the power supply is ON 2003, 5 relays, a RF transmitter (27 MHz) ,
the port A is set to a value of ‘FF’ and the a RF receiver (2.4 GHz), a USB port and a
port C is set to a value of ‘OF’. That is all mobile.
the pins connected to the keys are set in a The Personal computer is a regular use
HIGH state. The other two ports B and E are windows XP compatible system with
set to ‘0’. The 8 lines of the port D are installed dot net software. The MAX 232 is
connected to the data lines of LCD. Now the used for the CMOS TTL logic conversion
LCD starts displaying the current status of between the PIC, PC and the mobile. Here
the process. The displaying of messages in two PIC 16F877A are used because of
the LCD constitutes the main program of the memory restriction. One PIC is for the
PIC. The changes of state of the keys are communication between mobile and PC and
given as interrupts. The first 5 keys are another PIC is for driving the relay and thus
assigned to the selection of substation 1 to 5. helps in the control of the flying robot. The
The no. of substations to be connected to a ULN 2003 as we know is a current booster
main station can be our own choice but care and it helps in driving the relays by boosting
must be taken that it does not affect the up the signal from the PIC. 5 relays are used
speed and efficiency of the process. The 6th for the controlling of the flying robot. The 5
key is assigned for Human Detection, 7th is relays correspond to low, medium, high
for Vehicle detection, 8th is for emergency speeds along with left and right direction
stopping of the flying robot, 9th is for control. The 27 MHz RF transmitter is used
detection of combination of all and the 10th to transmit the control signals to the flying
is for saving the image. First, the substation robot from the substation. The 2.4 GHz
must be selected. For example, if key 1 is receiver is used to receive the Video images
pressed the PIN 2 becomes low and in the substation that are captured by the
becomes an interrupt to the main program. CMOS camera present in the flying robot.
Now, the PIC sends a data ‘1’ in the form of The USB port is nothing but a TV box that
text message using mobile to the substation is used for interfacing camera and the PC.
1. But PIC follows CMOS logic and mobile The mobile is used for the full duplex
follows TTL logic. Hence to make the logic communication between the main station
conversion MAX 232 is employed here. and the substation. It uses the GSM protocol
Now, when the substation is selected for this purpose.
and the acknowledgement is received from The process in the substation starts only
the selected substation LCD displays the when it receives a message from the main
message ‘substation 1’. Simultaneously, the station. First, all the ports of the 1st PIC are
flying robot starts flying and starts set in LOW state. When the mobile receives
monitoring. Next command should be the the message, 25th PIN of this PIC goes high.
category to be detected. When the This now sends the acknowledgement
corresponding key is pressed, the detection message back to the main station through
starts. If elements pertaining to that category the same mobile. Again MAX 232 comes in
are detected, a successful detection message to play for logic conversion. Simultaneously
as the message is sent 1st PIC sets the 25th receiver (2.4 GHz) inbuilt in it that sends the
pin of the 2nd PIC to high. This initiates the captured video images to the substation as
processing in the 2nd PIC. This sets the 3rd 4th analog signals.
5th and 6th pin of ULN 2003 to logic 1 state.
This boosts the current to required level to SOFTWARE
drive the relay. IMAGE PROCESSING
When the category to be detected is The image processing is done to
sent by the main station through the mobile, detect and analyze the moving object in two
the PIC sends the command to the PC phases. The first phase is the detection of
through the 2nd pin of the RS232 serial port. any moving object by the use of gray scale
Thus image processing starts and if any and erosion filter along with the pixellate
element is detected PC sends signal back to function. Only if any moving object is
PIC via 3rd pin of the serial port. After the detected it moves to the next phase which is
logic conversion in the MAX 232 the output the detection of any particular category by
signal goes to the 1st PIC which sends the the use of edge detection algorithm.
message to the mobile in the main station.
FIRST PHASE
The video images caught by the
CMOS camera in the flying robot are The JPEG images which we receive
received in the substation with a frequency must be converted into a grayscale image.
of 2.4 GHz. The data received is an analog This will greatly increase the efficiency of
one. This is converted in to a digital image detection.
in the TV box or the USB port. The MPEG
images are converted to JPEG images and GRAY SCALE FILTER
are received at 96 frames per second thus
giving an illusion of moving object. The In photography and computing, a grayscale
JPEG images have 3 bytes per pixel. or greyscale digital image is an image in
which the value of each pixel is a single
The hardware components present in sample, that is, it carries only intensity
this flying robot unit are the RF receiver information. Images of this sort, also known
circuit (27MHz), RF transmitter circuit (2.4 as black-and-white, are composed
GHz inbuilt in the CMOS camera), RF exclusively of shades of gray, varying from
decoder, CMOS camera with the OV7950 black at the weakest intensity to white at the
sensor and a motor driver circuit. The RF strongest.
receiver, RF decoder and the motor driver
circuit are inbuilt in the helicopter that was Grayscale images are distinct from one-bit
designed by the company. A battery is black-and-white images, which in the
required to give power supply to the flying context of computer imaging are images
robot unit. RF receiver receives the control with only the two colors, black, and white
signal and forwards it to the decoder. The (also called bilevel or binary images).
decoded signal is then given to the motor Grayscale images have many shades of gray
driver circuit which drives the main motor in between. Grayscale images are also called
and the tail motor. The movement of the monochromatic, denoting the absence of any
helicopter is preprogrammed in the PIC in chromatic variation.
the substation. The CMOS camera is simply Grayscale images are often the result of
attached to the flying robot with the help of measuring the intensity of light at each pixel
an insulation tape and it has got a RF
in a single band of the electromagnetic want to be sure which is which. These three
spectrum (e.g. infrared, visible light, filters also allow specification of a
ultraviolet, etc.), and in such cases they are threshold. Any value below this threshold
monochromatic proper when only a given will be clamped to it. For this project testing
frequency is captured. But also they can be we have kept the threshold at 15.
synthesized from a full color image;
Thus using the edge detection method, the
EROSION OPERATION no. of edges is found. If the no. of edges
Once the image is converted into grayscale, equals the no. of edges for human (32
the background and the foreground of the edges), the message ‘human is detected’ is
image must be separated. This is done using sent. The image processing part ends here.
the erosion operation. The process in erosion
operation is given below MOTION DETECTION

SECOND PHASE Motion detection is the action of sensing


physical movement in a given area. Motion
If any moving object is detected it moves to can be detected by measuring change in
the next phase which is finding out the speed or vector of an object or objects in the
detection of any particular category. This is field of view. This can be achieved either by
done using the edge detection algorithm. mechanical devices that physically interact
with the field or by electronic devices that
EDGE DETECTION ALGORITHM quantifies and measures changes in the
given environment.
Edge detection is a terminology in image Thus by using the filters- Grey Scale
processing and computer vision, particularly Erosion filter. Motion detection algorithm,
in the areas of feature detection and feature the moving object is differentiated into man
extraction, to refer to algorithms which aim and animal.
at identifying points in a digital image at
which the image brightness changes sharply ADVANTAGES
or more formally has discontinuities.
• High processing time (2ms)
Edge detection filters work essentially by
looking for contrast in an image. This can • Low cost
be done a number of different ways; the • Less weight
convolution filters do it by applying a • Exclusively human can be detected using
negative weight on one edge, and a positive motion detection algorithm.
on the other. This has the net effect of • No external noise disturbance.
trending towards zero if the values are the • Autonomous system.
same, and trending upwards as contrast • Optional manual available
exists. This is precisely how our emboss
filter worked, and using an offset of 127
would again make these filters look similar
to our previous embossing filter. The CONCLUSION
following examples follow the different
filter types in the same order as the filters
above. The images have a tool tip if you
With the positive results obtained, our 6. Russ, J.C., The Image Processing
proposed autonomous surveillance system Handbook. Second ed. 1995, Boca Raton,
satisfies the design and implementation of a Florida: CRC Press.
recognizable real-time surveillance system 7. Image processing in MATLAB 6.1
for detecting man sized object. The by national instruments.
grayscale erosion and edge detection 8. Through the Eyes of the Condor: An
algorithm is fast, robust and computationally Aerial Vision of Latin America by Robert B.
inexpensive. The helicopter keeps the Haas, Marie Arana
camera roughly perpendicular to the ground. 9. Manual of Aerial Photography by
The whole system process is completed in 2 Ronald Graham, Roger Read
seconds, and therefore it is highly efficient.
If the power to the helicopter is increased,
the fly time can be increased. Safety
equipment can be installed but it will lead to
higher cost. The light weight of the flying
unit can be changed by using heavier
material for the head wings. Another
solution for this issue is the use of motor to
obstruct the opposite wind. But these will
increase the cost of the system. In future we
plan to achieve these disadvantages with low
cost in mind.

REFERENCE

1. IEEE conference paper 2007 titled”


Arial photography” in department of
aviation.
2. Dudgeon, D.E. and R.M. Mersereau,
Multidimensional Digital Signal Processing.
1984, Englewood Cliffs, New Jersey:
Prentice-Hall.
3. Castleman, K.R., Digital Image
Processing. Second ed. 1996, Englewood
Cliffs, New Jersey: Prentice-Hall.
4. Oppenheim, A.V., A.S. Willsky, and
I.T. Young, Systems and Signals. 1983,
Englewood Cliffs, New Jersey: Prentice-
Hall.
5. Papoulis, A., Systems and
Transforms with Applications in Optics.
1968, New York: McGraw-Hill.

You might also like