Hardware Implementation of Hough Transform for Straight Line Detection

Abstract

The use of line detection algorithms is becoming increasingly important in the field of image processing, particularly in areas that require automation, such as lane detection in self-driving cars and the localization of robots in the field of robotics. The Hough Transform is considered a common algorithm for line detection, but its computationally intensive nature can lead to poor performance when implemented on a general-purpose processor. The proposed work looks at implementing the Hough Transform on a Field Programmable Gate Array (FPGA) instead, as FPGAs hold parallel data processing capabilities, which reduces system latency. The work involves identifying the necessary parameters for the application of Hough Transform and designing the system by utilizing RAM and ROM modules on the Intel Quartus Prime while taking into account the arithmetic operations required to produce the desired output. The execution time of the Hough Transform on an FPGA was compared to its execution time in the MATLAB software environment, with the result showing that the FPGA’s implementation is 94 times faster than the execution of Hough Transform in MATLAB.