| CPC A01K 29/002 (2025.08) [G01S 15/8993 (2013.01); G06T 7/50 (2017.01); G06T 7/90 (2017.01); G06V 10/56 (2022.01); G06V 40/10 (2022.01); G06T 2207/10028 (2013.01); G06T 2207/10136 (2013.01); G06T 2207/20081 (2013.01)] | 13 Claims |

|
1. A method for calculating a backfat thickness of a sow by a processor of a device, the method comprising:
recognizing a dorsal angle measurement region using frame data;
recognizing a plurality of three-dimensional coordinate values related to at least three points within the dorsal angle measurement region; and
calculating a dorsal angle value of the sow for calculating the backfat thickness of the sow using the plurality of three-dimensional coordinate values;
wherein the recognizing the dorsal angle measurement region using the frame data comprises:
inputting at least one of the frame data and body data acquired based on the frame data into a dorsal angle measurement region detection model, wherein the dorsal angle measurement region detection model is pre-trained to select a region having the highest accuracy for dorsal angle measurement among a plurality of candidate dorsal angle measurement regions; and
recognizing the dorsal angle measurement region based on dorsal angle measurement region information output from the dorsal angle measurement region detection model.
|