Smart Parking System Based on Haar Cascade Classifier and SIFT Method

Parking is a non-temporary state of motion of a vehicle. Motorcycles are the most common type of vehicle in the parking area. The rate of motorcycle use in developing countries from year to year are constantly increasing. In Indonesia, motorcycle use from 2000 to 2018 has been increased by approximately 20% with an average increase of about 204 million per year. Parking systems in Indonesia are mostly still using security system which is quite fragile, one of which is by recording the license plate of the vehicle as a ticket for the conditions when exiting the parking lot. With the use of the system, the level of security from the parking lot is still very low and will trigger a crime in the vehicle, namely motorcycle theft. Therefore, a solution is needed to improve and assist the entrance and exit systems of the parking area, in order to create a safe-parking space. Through this work, an intelligent parking system based on RFID E-KTP (Electronic Identity) and Vision Processing are designed to improve the security of the parking area. Haar Cascade is used to detect and capture the picture of motorcycle license plates and then SIFT method (Scale Invariant Feature Transform) is implemented for the matching process of the image. In addition, it is also equipped with RFID which is used to retrieve E-KTP Id data from motorcycle users who will enter the parking area. Both data are used as a condition to be able to exit and enter the parking area. The success rate of the appliance at the entrance is 89% with an average time processing of 13.25 seconds, while at the exit has a 100% success rate with an average time of 19.75 seconds.