SISTEM PENGENALAN JENIS KENDARAAN BERMOTOR PADA SISTEM PARKIR MENGGUNAKAN METODE HISTOGRAM OF GRADIENT DAN K-NEAREST NEIGHBOR
Daftar Isi:
- Vehicle detection system is one of the most important technologies. because, it has many applications in the field of vehicles such as vehicle traffic monitoring, calculation of parked vehicles, different types of vehicles, and so on. In this skripsi the author conducted an experiment in which to distinguish the types of two-wheeled and four-wheeled motorized vehicles in one parking gate using Histogram Of Gradient (HOG) as a step used for object detection. HOG describes features based on local histograms of the orientation of the gradient that is rated with gradient magnitude and gradient direction and K-Nearest Neighbor (KNN) to classify images of different types of vehicles. So it can be analyzed how the system can recognize the type of vehicle that is the object of two-wheeled vehicles and four-wheeled vehicles. The stages of this vehicle detection are image pre-processing consisting of resizing and grayscale processes, and vehicle type detection using feature extraction process with Histogram Of Gradient and classification process with K-Nearest Neighbor. From 60 vehicle images which are divided into two-wheeled and four-wheeled vehicles, the test results with an accuracy of 81.69% are obtained.