Introduction to Xilinx System Generator
Introduction To Simulink And System Generator
Simulink
o Developed by MathworksInc. o A Model-based design environment integrated with Matlab o Provides various sets of block libraries to help you create , simulate, and test a complicated system,including control system, communication, image processing, etc.
System Generator
o Developed by Xilinx Corp. o Provides device-optimized DSP block library for Simulink o Let you model DSP algorithm, generate HDL code, and verify FPGA designs for Xilinx devices in a intuitive and efficient way.
System Generator
Type Simulink at the MATLAB command prompt or open the Simulink Library Browser by clicking the corresponding button in the MATLAB toolbar
Simulink Library Browser
Create a new model blank sheet by clicking the Create a new model button in the Simulink Library Browser
Example 1 To read and display an image
Simulink Model
Go to Video and Image processing blocksets Sources Image from file
Xilinx SysGen Model
Go to Video and Image processing blocksets Sources Image from file
Reshape is an operator that is used to convert the dimensions. The parameter one dimensionality is chosen as 1-D array since the XSG accepts 1-D as the inputs.
The number of bits is chosen as 8 since we are using a gray scale image. Quantization and overflow is kept as round and saturate by default.
Buffer size:
If an image of a X b is taken then the buffer size should always be a*b. Eg: 64X64 is the image size then buffer size must be 4096.
Reshape:
At the output the parameters of reshape operator must be: o Dimentionality: Customise o Dimension: it must be 2D since we are converting back to the original image. Hence specify the size of original image
SETTING THE STOP TIME
The run time/ stop time is always calculated as twice the dimension of the matrix. If the image is 64X64 then the stop time= 2*64*64.
Simulation
Stop time
Using Matrix Viewer
Workspace
Output using Workspace
After simulation the output is seen in the workspace.
Output obtained is 1-D array. It has to be converted to 2-D array.
Write the callback function in the editor window and link to the model
Output obtained
NOTE:In order to carry out processing on an image the modules have to be placed between gateway in and gateway out
Example 2 RGB to Gray Conversion
Reading RGB Image
Reading RGB Image
Conversion of RGB to Gray Image I=0.29*R+0.59*G+0.11*B Creating module for conversion
Creating module for conversion
Implementation of RGB Equation
RGB TO GRAY Conversion
RGB TO GRAY Conversion Output from Matrix Viewer
RGB TO GRAY Conversion Output from Workspace
RGB TO GRAY Conversion Output from Workspace
Example 3 Filtering
Filtering
Reading an image
Filtering module
5 x 5 Filter
Filtering
Set the Simulink System Period to 1/5 as 5*5 filter is used.
Filtering Output from Matrix Viewer
Filtering Output from Workspace
Filtering Output from Workspace
Example 3 Enhancement
Enhancement
Reading an image
Enhancement
Enhancement Output from Matrix Viewer
Enhancement Output from Workspace
Enhancement Output from Workspace
Example 4 Negation
Negation
Reading an image
Negation
Negation Output from Matrix Viewer
Negation Output from Workspace
Negation Output from Workspace
Example 4 Binarization
Binarization
Reading an image
Binarization
Binarization Output from Matrix Viewer
Binarization Output from Workspace
Binarization Output from Workspace
