Skripsi
DETEKSI KEPADATAN KENDARAAN MENGGUNAKAN METODE K-NEAREST NEIGHBOR DIOPTIMASI DENGAN ALGORITMA GENETIK UNTUK SISTEM TRANSPORTASI DI KOTA PALEMBANG
Growth factors such as economic, social, political, and cultural developments have become one of the traffic problems in Palembang City that has caused the occurrence of traffic density. To avoid getting stuck in traffic density, it is necessary to detect the density of the vehicle. The study uses the Closed-Circuit Television (CCTV) version of Pan, Tilt, and Zoom (PTZ) that can record various traffic incidents, one of which is the density of road conditions. YOLO (You Only Look Once) version 8 is used to detect, classify, and count vehicles on the traffic with a collection of 1462 image files in.jpg format and 96 CCTV videos of the Palembang City Communications Service. The accuracy of the model reached 70.5% and the accuracy of the image test data was 77.64%. The accurateness of the test data with CCTV video reached 87.13%. The density detection of this vehicle is using the K�Nearest Neighbor (KNN) method which has been optimized with the Genetic algorithm. With the KNN method, the accuracy reaches 89.9% with 7 videos having detection errors. Subsequently optimized by the Genetical algorithm, it produces an accuracy of 87.5% with 12 videos with detection error. The accuracy has been reduced by 2.15% due to several factors, one of which is the setting of parameters in the genetic algorithm.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407003977 | T149245 | T1492452024 | Central Library (REFERENCES) | Available but not for loan - Not for Loan |
No other version available