US 12,169,966 B2
Method for optimizing detection of abnormalities in images, terminal device, and computer readable storage medium applying the method
Chung-Yu Wu, New Taipei (TW); Guo-Chin Sun, New Taipei (TW); and Chih-Te Lu, New Taipei (TW)
Assigned to HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed by HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed on May 19, 2022, as Appl. No. 17/748,259.
Claims priority of application No. 202111441530.6 (CN), filed on Nov. 30, 2021.
Prior Publication US 2023/0169762 A1, Jun. 1, 2023
Int. Cl. G06K 9/00 (2022.01); G06N 3/045 (2023.01); G06V 10/74 (2022.01); G06V 10/77 (2022.01); G06V 10/82 (2022.01); G06V 10/98 (2022.01)
CPC G06V 10/98 (2022.01) [G06N 3/045 (2023.01); G06V 10/761 (2022.01); G06V 10/7715 (2022.01); G06V 10/82 (2022.01)] 20 Claims
OG exemplary drawing
 
11. A terminal device comprises:
a storage medium; and
a processor,
wherein the storage medium stores computer programs, and
the processor executes the computer programs to implement the following:
forming a generative adversarial network (GAN) based on training images; wherein the training images are normal images; the GAN generates a first image similar to the training image, confirms a first similarity ratio between the first image and the training image, and generates a generation parameter of the GAN or an identification parameter of the GAN based on the first similarity ratio between the first image and the training image for training the GAN;
obtaining testing images; wherein the testing images comprises the normal images and abnormal images;
determining a second similarity ratio between second image corresponding to the testing image generated by the trained GAN and the testing image;
determining whether the second similarity ratio is larger than a specified threshold value;
confirming the testing image as a normal image when the second similarity ratio is larger than the specified threshold value; and
confirming the testing image as an abnormal image when the second similarity ratio is less than or equal to the specified threshold value.