Implementasi Convolution Neural Network dalam Melakukan Pembayaran Parkir pada Kendaraan Roda Empat di Area Publik

Abstract

In many areas of life, automated sistem technology has developed rapidly to support human performance. One such field is security sistems, especially for private vehicle drivers who require convenient and secure parking spaces. In parking sistems, ticket machines or human labor are usually used to record vehicle plate numbers as owner identification. In addition to creating traffic congestion due to the long process, the ticket sistem also does not guarantee the safety of the car from theft. Therefore, there is a need for an application for automatic recognition of vehicle plate numbers to facilitate the control of parking security sistems. Hence, there is an urgent need to build and design a prototype parking gate security sistem that uses digital image processing of barcodes and vehicle license plates. A webcam positioned in front of the gate will be used as a sensor to open and close the gate, taking a picture of the barcode on the vehicle. Once the image is captured, the characters on the license plate will be extracted and matched with a pre- built database. The identification data will be paired with the barcode read by the barcode reader to obtain additional characteristics of the vehicle, preventing counterfeiting of license plates. The processed data will be sent to the Arduino Uno microcontroller which will operate and control the servo motor if the data matches the reference data. Vehicle queues will be reduced and protection from theft will be increased thanks to the ability of this automated sistem to open and close parking gates more quickly and safely. It is expected that the application of this technology will provide a practical solution to improve the effectiveness and security of vehicle parking sistems.