Publication: Car Plate Detection System Using RISC-V
Loading...
Date
2023-07
Authors
Teoh, Choon Hong
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Car plate detection is one of the image processing application that hugely consumed time and resources in computer. Many researchers proposed embedded
System-On-Chip that incorporating a Field Programmable Gate Arrays (FPGA) and a Hard Processor System (HPS) such as Advanced RISC Machine (ARM) processor as solution to obtain lowest execution time. However, the choice between soft core and hard core embedded SoCs influences the system's performance and flexibility. A Sapphire RISC-V SoC platform which provides a soft configurable RISC-V processor that can implement in Titanium FPGA is proposed in this study. The main purpose of this project is develop a car plate detection algorithm that is suitable for RISC-V soft core implementation and investigate its performance with hard processor in term of processing speed. In this project, total 11 image processing steps have developed and implemented in RISC-V soft core to detect the car plate. The car plate detection algorithm is capable to detect car plate from front view, back view, side view as well as car plate with yellowish character. The distance for maximum detection rate is between 1.5m to 2.0m. Besides, the performance of RISC-V solution that using soft core is identified and compared with the HPS solution that using hard core from previous study in term of processing time. As a result, the total execution time of
RISC-V and HPS are 66.827ms and 83.792ms respectively where RISC-V is
approximately 20.26% faster than HPS.