YUV SYSTEM AND METHOD FOR AUTONOMOUS DRIVING BASED ON YUV-BASED SEGMENTATION AND AUGMENTATION

摘要

Disclosed are a system and a method for autonomous driving based on YUV-based segmentation and augmentation. The method for autonomous driving includes the steps of: receiving a driving image based on a first image type; converting the driving image based on the first image type into a driving image based on a second image type; and determining a parameter for controlling a vehicle by applying the driving image based on the second image type to a neural network. A first difference between an actual driving image and a virtual driving image in the space of the first image type can be greater than a second difference between an actual driving image and a virtual driving image in the space of the second image type.