Regresar

Study of the Pattern Recognition Algorithms for the Automation of a Smart Semaphore Through FPGAs

Abstract:

The following document presents the development of an intelligent traffic light, based on FPGA technology, which assessing the amount of traffic gives priority to the lane with the highest number of cars. The comparative study of pattern recognition algorithms was performed: color matching algorithm, cross correlation algorithm and optical character recognition algorithm (OCR). The methodology for the development of the system is based on a matrix of programmable logic gates in the field or FPGA, a camera that acquires images and sends them for digital processing programmed in LabView intelligently with pattern recognition algorithms for decision making in the area of vehicular traffic control. The density of vehicular traffic is determined and the card changes the duration of the green light given for each lane according to the number of existing vehicles. As a result, it was obtained that the most appropriate algorithm to implement an intelligent traffic light prototype using FPGAs was the color matching algorithm that has an accuracy of 100% and a response time of 3 ms.