Deteksi Objek Kereta Api menggunakan Metode Faster R-CNN dengan Arsitektur VGG 16

ABSTRAKKereta merupakan sebuah alat transportasi umum yang sering digunakan oleh masyarakat untuk berpergian dari kota asal ke tujuan. Mereka membutuhkan akan sarana mempermudah aktifitas mereka. Namun kecelakaan di persimpangan jalan raya terlintasi kereta api memiliki angka cukup besar akibat kelalaian petugas menutup palang pintu api. Maka itu penelitian ini dibuat agar mengetahui keberadaan berdasarkan jarak dan tingkat cahayanya siang sampai malam hari. Sistem dibangun menggunakan metode Faster RCNN dengan model arsitektur VGG16 objek antara lokomotif gerbong cahaya terhadap objek. Setelah dilakukan pengujian paling dekat ±2 meter ±250 meter, diperoleh rata-rata akurasi sebesar 79,09%, 97,05%. memperoleh keakurasian deteksi 86,40%, 97,23%.Kata kunci: Deteksi Objek, RCNN, VGG, Kereta Api, Jarak, LuxABSTRACTRailway is a public transportation that often used by the to travel from home town destination city. They need facilitate their activities. But accidents at intersection of highway crossed train has considerable number due negligence officer close railway stopbars. Therefore, this study was made know existence trains based on distance and light level day night. The system built using method with architectural determine objects between locomotives carriages object. After testing closest meters meters, obtained an average accuracy for 79.09%, 97.05%. detection locomotive 86.40%, car 97.23%.Keywords: Object Detection, Railway, Distance, Lux