Miftahuddin, Yusup and FAHRUDIN, NUR FITRIANTI and PRAYOGA, OCHAMAD FACHRY (2020) Algoritma Scale Invariant Feature Transform (SIFT) pada Deteksi Kendaraan Bermotor di Jalan Raya. MIND (Multimedia Artificial Intelligent Networking Database) Journal, 5 (1). pp. 54-64. ISSN 2528-0902

Text (An article in a journal, magazine, newspaper. Not necessarily peer-reviewed. May be an electronic-only medium, such as an online journal or news web) Algoritma Scale Invariant Feature Transform (SIFT) pada Deteksi Kendaraan Bermotor di Jalan Raya.pdf Download (957kB)

Abstract

The process of collecting vehicles still done manually and requires a lot of human resources. Therefore, we need a system that can detect and classify vehicles passing on the highway automatically. SIFT is an algorithm for identification of an image. The features will be compared using the K-Nearest Neighbor (KNN) method. In this study, system will be designed to detect the type of heavy vehicle using the SIFT method to measure the accuracy of success based on the value of lighting, number of objects, changes in rotation, and day night conditions. Dataset used was 100 heavy vehicle images. The system performance during daytime conditions gets an average precision value of 100%, a recall value of 54%, and an accuracy value of 78%. From the results of precision and recall, the f-measure value is 67 %. Keywords: SIFT, heavy vehicles, K-Nearest Neighbour.

Actions (login required)

View Item