| CPC G06V 20/54 (2022.01) [G06T 3/10 (2024.01); G06T 7/11 (2017.01); G06T 7/60 (2013.01); G06V 10/751 (2022.01); G06V 10/82 (2022.01); G06V 20/70 (2022.01); H04N 7/183 (2013.01); G06T 2207/20084 (2013.01)] | 20 Claims |
|
1. A method for monitoring a harbor performed by a computing device, the method comprising:
obtaining a preset first view transformation information used to transform an image having a first view attribute into an image having a second view attribute different from the first view attribute;
obtaining a harbor image having the first view attribute;
generating a segmentation image having the first view attribute and corresponding to the harbor image by performing an image segmentation using an artificial neural network trained to output information, from an input image, related to an object included in the input image, wherein the segmentation image includes a first pixel labeled to correspond to a ship and a second pixel labeled to correspond to a sea;
obtaining sea level information reflecting a height of a sea level;
updating the first view transformation information by reflecting the sea level information;
generating a transformed segmentation image having the second view attribute through a viewpoint transformation of the segmentation image using the updated first view transformation information; and
calculating berthing guide information of the ship based on the transformed segmentation image, wherein the berthing guide information includes at least one of information related to a distance to a wharf wall of the ship or information related to an approaching velocity of the ship to the wharf wall.
|