Experimental investigation on the execution time of the C++ program
The purpose of this project is to conduct an experimental investigation on the execution time of the C++ programs that perform Gaussian Filtering and Edge Detection on a digital image. The report first gives an introduction about the project. The theory concepts of sine and cosine calculation, Gaus...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/70423 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | The purpose of this project is to conduct an experimental investigation on the execution time of the C++ programs that perform Gaussian Filtering and Edge Detection on a digital image.
The report first gives an introduction about the project. The theory concepts of sine and cosine calculation, Gaussian Filtering and Edge Detection are explained. The report then gives a brief introduction about the Canny operator, Zedboard and OpenCV library. The Canny operator is the method of Edge Detection used in this experiment. The OpenCV library algorithm is used for the C++ programs to implement the Gaussian Filtering and the Edge Detection function.
The second part of the report explains about the design documentation used in the project. The report will discuss about the C program which computes the value of the sine and cosine function and two C++ programs which perform Gaussian Filtering and Edge Detection on a digital image. The C++ programs will then be tested on the Zedboard and the execution time is recorded.
The third part of the report discusses about the results obtained in the project. The comparison between the original image and the output image of the C++ programs is done. The Kernal size used in the Gaussian filtering program and Edge detection program is experimented with different values. The C++ programs are run on both the desktop computer and the Zedboard and the execution time on both platforms are compared. The execution time of the Gaussian filtering program and Edge detection program run on Zedboard is 1.880s and 0.890s respectively. The execution time of the Gaussian filtering program and Edge detection program run on desktop computer is 1.867s and 0.614s respectively. Hence, the execution time of the C++ program run on the desktop computer is shorter than that of the Zedboard. This leads to the conclusion that the speed of the processor in the desktop computer is faster than that of the Zedboard. |
|---|