motorcycle-license-plate Computer Vision Project

Deteksi Plat Nomor Motor Indonesia

Warna plat nomor indonesia

• hitam
• putih
• merah

Kondisi citra/gambar

• dalam satu citra bisa terdapat 1, 2, lebih banyak objek/plat nomor
• posisi angle/sudut citra pada objek
• jarak pada objek dekat dan jauh
• lingkungan malam, dominan siang
• citra yang gelap dan terlalu terang
• occlusion/yang menghalangi objek

License Plate Detection Motorcycle Indonesia

This project aims to develop a machine learning model that can detect and recognize the license plates of motorcycles in Indonesia. The dataset consists of a collection of images of motorcycles with various license plate colors, such as black, white, and red with black and white being the most dominant. The images also vary in terms of the number of objects, the angle of view, the lighting condition, and the presence of occlusions. The model will be trained and evaluated using the YOLOv8 algorithm, which is a state-of-the-art object detection method. The expected outcome of this project is a model that can accurately and efficiently detect and recognize the license plates of motorcycles in Indonesia, regardless of the image quality and complexity.