SISTEM OTOMATISASI PENGHITUNGAN KENDARAAN (VEHICLE COUNTING) BERBASIS CONVOLUTIONAL NEURAL NETWORK

Mochamad Ghifari ‘Azmi, NIM.: 16650051 (2020) SISTEM OTOMATISASI PENGHITUNGAN KENDARAAN (VEHICLE COUNTING) BERBASIS CONVOLUTIONAL NEURAL NETWORK. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (SISTEM OTOMATISASI PENGHITUNGAN KENDARAAN (VEHICLE COUNTING) BERBASIS CONVOLUTIONAL NEURAL NETWORK) 16650051_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (SISTEM OTOMATISASI PENGHITUNGAN KENDARAAN (VEHICLE COUNTING) BERBASIS CONVOLUTIONAL NEURAL NETWORK) 16650051_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (7MB) | Request a copy

Abstract

The growth of motor vehicle users is increasing every year. As vehicle users grow, road traffic volume also increases. To manage traffic properly, monitoring of traffic flow is certainly needed. Traffic flow monitoring can be seen from the number of motorized vehicles passing a road. But the process of counting the number of motorized vehicles manually is limited to the accuracy of human vision. Currently, almost every intersection in Indonesia, especially CCTV / ATCS cameras have been installed, therefore the author sees the possibility of utilizing the video from the camera to be processed in order to generate traffic flow monitoring data automatically. This research was started by collecting data from CCTV / ATCS of the Department of Transportation of Sukoharjo Regency and after that the image preprocessing was done with cropping. Followed by object detection, tracking and counting or vehicle counting. The vehicle calculation in this video is done using the convolutional neural network architecture. The results of this study conclude that if you want to count cars, be it trucks, buses, minibuses, sedans or pickups, the COCO dataset is suitable to use the truck, car and bus classes. Meanwhile, to calculate the motorbike is not suitable, and finally, if you want to count all the vehicles passing on the road, be it bicycles, motorbikes, trucks, buses, pickups and sedans, the COCO dataset is also not suitable for application in the Sukoharjo area and in general Indonesia.

Share this knowledge with your friends :

Actions (login required)

View Item