Hardware software co simulation of edge detection for image processing system using delay block in xsg

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________________
Volume: 03 Issue: 05 | May-2014, Available @ http://www.ijret.org 549
HARDWARE SOFTWARE CO-SIMULATION OF EDGE DETECTION
FOR IMAGE PROCESSING SYSTEM USING DELAY BLOCK IN XSG
Aniket A. Ingle1
, Vrushali G. Raut2
1
Dept. of Electronics and Telecommunications, Sinhgad College of Engineering, Pune, India
2
Dept. of Electronics and Telecommunications, Sinhgad College of Engineering, Pune, India
Abstract
This paper implement an image processing algorithm applicable to Edge Detection for still image in a Xilinx FPGA using System
Generator. We prefer sobel algorithm which is most reliable and gives us an efficient output. If we prefer to write HDL code for
such algorithm in Xilinx FPGA then it’s too bulky and time consuming. We design this system with use of Xilinx System Generator
blocks. Its tool with a high- level graphical interface under Matlab environment Its Simulink based blocks which makes it very
easy to handle with respect to other software for hardware description.
Keywords— Matlab, Xilinx System Generator, FPGA, Edge detection algorithm.
--------------------------------------------------------------------***------------------------------------------------------------------
1. INTRODUCTION
For human interpretation image processing improve quality
of image. There are lots of examples where image
processing helps to analyze, infer and make decisions.
Image processing work pixel by pixel
In image processing before feature extraction and object
segmentation we have to detect edges from frame. This edge
detection process detects outlines of an object. Object and
background in image is isolated by edge detection feature. A
sharp discontinuity in image is located by edge detection.
The boundary of object in any image is characterized by
discontinuity which gives instant change in pixel intensity.
This process compress image without losing any important
feature of that image. Edge detection is initial stage of image
processing but it corrupted with white noise always. These
edge detection methods are classified into two categories,
gradient and Laplacian. Matrix area gradient operation is
basic Edge detection operator. By putting mask on detected
pixel calculate value of detected point using edge detector
operator. If the value is greater than threshold value then it
considered as edge point. Robert, Prewitt and Sobel are
example of gradient based edge detection algorithm .Kernel
operator play a vital role in gradient based edge detection
algorithm which calculates the slope in directions that are
perpendicular to each other.
The goal of this paper is to implement an image processing
algorithm applicable to Edge Detection system in a Xilinx
FPGA using System Generator with a focus on achieving
low cost and short development time. Edge detection
algorithms are implemented on software. But now days
hardware approach has become an alternative where
complex computation reduce and with help of FPGA or
Reconfigurable device we can use parallelism and pipelining
concept which easily increase speed. By Implementing
image processing on reconfigurable hardware like FPGA
which reduce the time of production cost, enables rapid
ASIC prototyping of complex algorithm .It also simplifies
debugging and verification.
2. LITERATURE REVIEW
A lot of work done on edge detection algorithm to detect
edge of an object .On the basis of edge detection algorithm
we can improve quality of image for human interpretation.
Image processing used in various field now days such as in
medical application, for digital aerial image detection from
satellite, for vehicle detection etc.
There are broadly three methods to detect edges
1) First order derivative (Gradient Method) Method.
Example: A) Robert Operator
B) Prewitt Operator
C) Sobel Operator
2) Second order derivative Method.
Example: A) Laplacian
B) Laplace of Gaussian
C) Difference of Gaussian
3) Optimal edge detection method.
A) Canny edge detection.
The derivative operators are used for image enhancement or
to enhance the details present in the image and these
derivatives operations can be used for detection of edges
present in the image.
In paper [1] represent how to implement an image
processing algorithm applicable to Edge Detection system in
a Xilinx FPGA using System Generator for still image, with
a focus on achieving overall high performance, low value
and short development time to come in market. The design
of edge detection is demonstrate with help of Xilinx system
generator block in Simulink environment. Spartan 3A board
use to implement this design.

_______________________________________________________________________________________________
In this paper [2] use System Generator tool in developing
vehicle image processing edge detection algorithms which is
developed by Xilinx based on MATLAB. Edge detection
algorithm model and design are finished in MATLAB
Simulink, preparation of top-level file in ISE 10.0
environment then achieve a System Generator functions and
other modules instantiated. Import the hardware design
which generate by System Generator into the paper, and
then the paper should be simulated, synthesis, finally
completed the hardware-based of the algorithm. And
display the processing image through VGA.
This paper prefers first order derivative method over second
order derivative method for edge detection. First derivation
can be computed by using gradient operators .The second
order derivative is very sensitive to noise present in the
image and that is the reason second derivative operators are
not usually used for edge detection operation but the second
derivative operators are gives some secondary information,
sign determine whether the point is lying on the darker side
of the image or a point is lying on the brighter side of the
image.
This paper prefer Sobel operator over Prewitt and Robert
Operator. Robert filters have shortest support and more
vulnerable to output noise. The Prewitt operator is based on
the idea of central difference and is much better than
Roberts’s operator. Prewitt’s operator has longer support
and is less vulnerable to noise. The Sobel operator is also a
central difference with more weights to the central pixels
where averaging as given by equation. It has improved noise
suppression than Prewitt’s operator.
In third section we describe about edge detection, fourth
section gives idea about proposed design in which we
describe flow of Xilinx system generator(XSG),sobel edge
detector operator and at last we discuss on implementation
and application which can be implemented and use in
practical world with help of our proposed design.
Canny edge detection is optimal edge detection method.
According result of several research papers canny gives us
finest edge detection as compare to First order derivative
method or second order derivative method.
XSG (Xilinx System Generator) is tool with high level
graphical interface under matlab environment with Simulink
based block. This is finest way for hardware approach. It
provides easier hardware verification & implementation
compare to HDL based approach. Our goal to achieve high
performance, low cost, short development time using Xilinx
system generator (XSG) it’s easily fulfils. It directly
generate .UCF file of VHDL or VERILOG code which we
can burn directly on FPGA board.
3. SOBEL EDGE DETECTION ALGORITHM
The Sobel operator is type of first order edge detection
operator. It computes the gradient of image intensity
function.
At every point in the image the resultant gradient at this
point is given by Sobel norms. There are only 0 and 90
degree convolution kernel used by sobel operator.
The magnitude of gradient at each point is find out by
combining these individual kernel. The gradient magnitude
is given by:
|𝐻| = Hx2 + Hy2 (1)
Hx (Convolution kernel Hy (Convolution kernel
in x direction) in y direction)
Fig.3.1: Convolution Kernel [1].
The magnitude of gradient at each point given by:
|H|=| Hx |+| Hy | (2)
This is much faster to compute. The sobel operator has
advantage of simplicity in calculation. Edge is detected with
help of two convolution kernel that’s reason it has low
accuracy. This convolution kernel shown in figure 3.1 [1].
4. PROPOSED WORK
The entire operation of edge detection proposed using
Simulink and Xilinx blocks goes through 3 phases,
A) Image pre-processing blocks.
B) Edge detection using XSG.
C) Image post-processing blocks
4.1 Image Preprocessing Blocksets
The model based design used for image pre-processing is
shown in figure 4.1[5]. The blocks utilized here are
discussed below. Input images which could be colour or
grayscale are provided as input to the File block. A colour
space conversion block converts three channel RGB image
into two channel grayscale image.
This data which is in 2D is to be converted to 1D for further
processing. Frame conversion block sets output signal to
frame based data of particular size and provided to unbuffer
block which converts this frame to scalar samples output at
a higher sampling rate.

_______________________________________________________________________________________________
Fig. 4.1: Image Pre-Processing.
4.2 Edge Detection using Xilinx Blocksets
The model based design using Xilinx blocksets for
processing the input image for edge detection is shown in
figure 4.2, Xilinx fixed point type conversion is made
possible by Gateway In block. To perform the edge
detection a convolution operation of the input image with
vertical and horizontal mask which is made up of delay
block, adder and substractor. This is followed by certain
arithmetic blocks to merge all the processed data’s.
Fig.4.2: XSG Model for edge detection
4.3 Image Post Processing Blocksets
The post-processing blocks which are used to convert the
image output back to floating point type we get 2d image at
output is shown in figure 4.3. It compare with threshold
value and finally we get fine edge detected image which we
can see in Simulink environment.
Fig.4.3: Image Post-Processing.
5. HARDWARE IMPLEMENTATION
For implementation of this design in a FPGA board the
entire module should be converted to FPGA synthesizable
format. For that purpose main module for edge detection is
converted to JTAG hardware co-simulation, this is done
with the help of System generator block specially its system
generator token. This block is configured according to the
target platform and a bit stream (*.bit) file is generated.
After the bit stream file is generated, hardware co-
simulation target is selected and in this work, Spartan 3E
starter kit (XC3S500E-FG320) is used for board level
implementation. The entire architecture with the hardware
and software co-simulation design is shown in figure 5.1.

_______________________________________________________________________________________________
Fig.5.1: FPGA Hardware Implementation.
6. APPLICATION
There are various application in image processing which
needs edge detection and where we can easily apply our
proposed hardware software co-simulation technique such
application are given below:
1. It use in a Biometrics.
2. Edge detection of Digital aerial image.
3. Vehicles detection at check post.
4. For machine vision also we can use it.
7. RESULTS AND COMPARISONS
We use image preprocessing blocks for converting our RGB
image into grayscale then this 2D image is again converted
into 1D for hardware approach. Sobel Edge detection model
made with help of XSG blocks in Simulink environment.
Our edge detected image finally goes through image post
processing block where it again converted into 2D image.
Edge detected image obtain by this approach is given below
in figure 7.2.We use targeted board for implementation is
Xilinx Spartan 3E XC3S500e.
Fig.7.1: Original image Fig.7.2: Edge detected image
Table 1 Proposed work for XSG model of Edge Detection
Parameter Proposed Work
Slices 51 1%
Slice Flip Flop 97 1%
Slice LUTs 8 0 %
Block RAM 0 0 %
IOB’s 39 16%
Table 2 Comparisons of proposed with existing design
Parameter Proposed
Design
[1] [2]
Platforms Xilinx
Spartan 3E
XC3S500e
Xilinx
Spartan 3E
XC3S500e
Xilinx
Spartan 3E
XC3S500e
Slices 51 163 326
Slice Flip flop 96 116 232
LUT 8 130 260
IOB 39 49 38
Frequency
(MHz)
249.128 - 220.7
8. CONCLUSIONS
Edge detection using software is not tough job but when we
are going to implement it on hardware we have to face
challenges like total VHDL code or Verilog code actually
becomes very bulky it’s near about 5000 lines. To shrink it
we use Xilinx system generator. Simulation speed increase
by this hardware software co-simulation technique .We can
easily go for ASIC prototype by this approach. This design
is implemented in the Xilinx FPGA Development kit.
REFERENCES
[1] Yahia Said, Taoufik Saidani, Fethi Smach and
Mohamed Atri “Real Time Hardware Co-simulation
of Edge Detection for Video Processing System”,
2012 IEEE.
[2] Zhang Shanshan, Wang Xiaohong “Vehicle Image
Edge Detection Algorithm Hardware Implementation
on FPGA”, 2010 International Conference on
Computer Application and System Modeling
(ICCASM 2010).
[3] Kobzili EI Houari, Benbouchama Cherrad, “A
Software-Hardware Mixed Design for the FPGA
Implementation of the Real-Time Edge Detection”,
2010 IEEE
[4] Gonzalez, Rafael C, 2008. “Digital Image
Processing”, Pearson Education, Inc., publishing as
Prentice Hall.
[5] Mohamed Nasir Bin Mohamed Shukor, Lo Hai
Hiung, Patrick Sebastian3, 2007. “Implementation of
Real-time Simple Edge Detection on FPGA” pp.
1404-1405,IEEE
[6] Nick Kanopoulos, NageshVasanthanada, Robert
L.Baker, 1988. “Design of an Image Edge Detection
Filter Using the Sobel Operator” pp. 359, IEEE.
[7] J.F. Canny, “A computation approach to edge
detection,” IEEE Transactions on Pattern Analysis
and Machine Intelligence, vol. 8, no.6, pp. 769-798,
Nov 1986.
[8] D. Marr and E.Hildreth,”Theory of edge Detection,”
Proc. Royal Soc. of London, series B, vol. 207, pp.
187-217, 1980.
[9] D. Demigny, and T. Kamle,” A discrete expression
of Canny’s criteria for step edge detector
performances evaluation”, IEEE Transactions on

_______________________________________________________________________________________________
Pattern Analysis and Machine Intelligence, pp. 1199-
1211, 1997.
[10] H. Neoh, A. Hazanchuk, “Adaptive Edge Detection
for Real-Time Video processing using FPGAs”,
Global Signal Processing (2004)
[11] SHIGERU.A, “Consistent Gradient Operators”, IEEE
Transactions on Pattern Analysis and Machine
Intelligence, 22 (3), 2000.
[12] Dong, Q., Song, C., Ben, C., Quan, L., “A fast sub
pixel edge detection method using Sobel-Zernike
moments operator”, Image and Vision Computing,
Vol.23, pp.11-17, 2005.Minimal Resources”, Proc.
International Conference on Intelligent Systems and
Signal Processing, pp. 338-343 2013.

Hardware software co simulation of edge detection for image processing system using delay block in xsg

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to Hardware software co simulation of edge detection for image processing system using delay block in xsg (20)

More from eSAT Publishing House (20)

Recently uploaded (20)

Hardware software co simulation of edge detection for image processing system using delay block in xsg