US 12,236,566 B2
Anomaly detection device, anomaly detection method, and anomaly detection program
Yun Xiang, Tokyo (JP); and Satoshi Ito, Kawaski Kanagawa (JP)
Assigned to Kabushiki Kaisha Toshiba, Tokyo (JP)
Filed by KABUSHIKI KAISHA TOSHIBA, Tokyo (JP)
Filed on Mar. 2, 2022, as Appl. No. 17/685,269.
Claims priority of application No. 2021-100236 (JP), filed on Jun. 16, 2021.
Prior Publication US 2022/0405911 A1, Dec. 22, 2022
Int. Cl. G06T 7/00 (2017.01); G06F 3/14 (2006.01); G06T 7/73 (2017.01)
CPC G06T 7/0002 (2013.01) [G06F 3/14 (2013.01); G06T 7/73 (2017.01); G06T 2207/10016 (2013.01); G06T 2207/20076 (2013.01); G06T 2207/20081 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An anomaly detection device, comprising:
a processor configured to:
acquire input data including image data,
derive a first anomaly degree corresponding to a difference between first feature data derived from the input data using a trained deep model trained by using natural data and second feature data derived from the input data using a prediction model trained by using target data;
derive a second anomaly degree corresponding to an estimated relative positional relationship between a first region in the image data and a second region in the image data based on the second feature data, at least a part of the second region not overlapping with the first region, and
calculate a total anomaly degree for the input data from the first anomaly degree and the second anomaly degree.