Method for processing picture of automatic driving vehicle
Disclosed herein is a method for processing a picture of an automatic driving vehicle. The method for processing a picture of an automatic driving vehicle according to an embodiment of the present disclosure includes obtaining a first image from a camera while a vehicle on which the camera is driving, generating a prediction signal for predicting a sign of occurrence of White-Out from the first image, when a quantity of light which is a threshold brightness or higher from the first image, storing the first image during a first period when the prediction signal is generated, and predicting and correcting a second image based on the first image stored during the first period and a correction model, when the second image in which the White-Out occurs in the first image is detected on a specific time during the first period, and the correction model is a result of learning the second image by processing high dynamic range (HDR) image. At least one of the automatic driving vehicle according to the present disclosure, a user terminal and a server may be linked with artificial intelligence, robot, Augmented Reality (AR), virtual reality (VR), and the like.
1. A method for processing a picture of an automatic driving vehicle, the method comprising:
obtaining a first image from a camera while a vehicle is moving, wherein the camera is mounted on the vehicle;
based on a quantity of light being detected from the obtained first image matching or exceeding a brightness threshold, generating a prediction signal for predicting an occurrence of White-Out from the first image, wherein an area in which the occurrence of White-Out occurs and a remaining area are distinguished in the first image;
storing the obtained first image during a first time period after the prediction signal is generated, wherein an image quality of the area in which the occurrence of White-Out occurs is stored as a higher image quality than the image quality of the remaining area; and
based on a second image in which the White-Out occurs in the obtained first image being detected at a specific time during the first time period, predicting and correcting the second image based on the stored first image and a correction model by:
extracting a portion that causes White-Out in the stored first image based on the second image being corrected with high dynamic range (HDR),
selecting a portion in which correction is required through the extracted portion that causes White-Out, and
adjusting a degree of correction according to the selected portion that causes White-Out,
wherein the correction model is a result of training the second image by processing an HDR image,
wherein a moving object is extracted from the first image,
wherein a position and a motion of the extracted object are determined, and
wherein an image quality of the area in which the White-Out occurs is stored as a higher image quality than an image quality of the remaining area based on the object being included in the area in which the White-Out occurs.
2. The method for processing a picture of an automatic driving vehicle of claim 1 , wherein the portion in which correction is required is selected by providing a priority according to a range in which a portion to be corrected is configured in the portion that causes White-Out.
3. The method for processing a picture of an automatic driving vehicle of claim 1 , based on the object being included in the selected portion that causes White-Out, the second image is predicted and corrected in greater detail than otherwise.