Comparison of different edge detection methods for real time uses on FPGA.
Edge detection is very simple tool used in Image processing to find boundaries of objects within images. Robert, Prewitt, and Sobel edge detection are gradient based edge detection methods used to find edge pixels in an image. Field Programmable gate arrays (FPGAs) are often used as implementation platforms for real time Image processing applications. This article presents comparison of Robert, Prewitt and Sobel edge detection algorithms which are written with the help of hardware descriptive language VHDL.
Why are FPGAs suitable for real time image processing Applications?
FPGAs have shown very high performance in image processing applications because of large number of memory banks. This memory banks can be accessed in parallel manner allowing FPGA to execute functions in few clock cycles. Thus, parallelism and pipelining techniques make FPGA suitable for fast processing of images in real time.
Edge detection hardware system architecture.
The edge detection hardware architecture is divided into two modules. One module is for accumulation of input data and the second is for performing edge detection operations. The components used in both the modules are shown in the figure.
For Robert operator 1 FIFO and 1 shift register was used for sufficient data accumulation while Prewitt and Sobel operators used 2 FIFO AND 3 shift registers. Therefore, FPGA hardware utilization is less in Robert operator-based edge detection as compared to Prewitt and Sobel operator.
After simulation it was observed that the system took 12.8ms, meeting the real time requirements to complete the edge detection process on grey scale image of size 640x480.The results given by Sobel operator were more efficient as compared to the Robert and Prewitt operator.
REFERENCES:
G. N. Chaple, R. D. Daruwala and M. S. Gofane, “Comparisions of Robert, Prewitt, Sobel operator based edge detection methods for real time uses on FPGA,” 2015 International Conference on Technologies for Sustainable Development (ICTSD), Mumbai, India, 2015, pp. 1–4, doi: 10.1109/ICTSD.2015.7095920.
