METHOD AND SYSTEM FOR LEARNING LANE CHANGEABLE TIME BASED ON CAMERA AND METHOD AND SYSTEM FOR PREDICTING TIME LANE CHANGEABLE TIME

摘要

Disclosed are a method for learning a lane changeable time point based on a camera and a system thereof and a method for predicting a lane changeable time point and a system thereof. According to an embodiment of the present invention, the method for learning a lane changeable time point comprises the steps of: receiving an image, in which a first or second label is assigned to an individual frame by being input through a camera in relation to a side-rear region of a vehicle, wherein the image, to which the first label is assigned, includes an image indicating a lane change in a blocked state and the image, to which the second label is assigned, includes an image indicating the lane change in a free state; assigning a label defining that a lane is changeable after n seconds (n is a natural number) in at least one frame before a switching time point in accordance with a frame per second (FPS) with respect to the switching time point when a state is switched from the blocked state to the free state; and learning a lane changeable time point based on the frames of the image and the label included in the frames.