US 11,669,949 B2
Apparatus and method for inferring contrast score of an image
Yevgeniy Gennadiy Guyduy, Bellmore, NY (US); and Tyler James Shaw, Bay Shore, NY (US)
Assigned to Canon U.S.A., Inc., Melville, NY (US); and Canon Kabushiki Kaisha, Tokyo (JP)
Filed by Canon U.S.A., Inc., Melville, NY (US)
Filed on Oct. 21, 2020, as Appl. No. 17/76,391.
Claims priority of provisional application 62/924,518, filed on Oct. 22, 2019.
Prior Publication US 2021/0118122 A1, Apr. 22, 2021
Int. Cl. G06K 9/00 (2022.01); G06T 7/00 (2017.01); G06N 5/04 (2023.01); G06N 7/01 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06T 7/0002 (2013.01) [G06N 5/04 (2013.01); G06N 7/01 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06T 2207/20076 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30168 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method of classifying a contrast level of an image, the method comprising:
obtaining an image from an image source;
extracting intensity values for each pixel of the obtained image;
calculating a probability distribution for the obtained image representing a number of pixels at each unique pixel value;
determining, from the calculated probability distribution, a spread value representing a series pixel values including at least a predetermined number of pixels at each pixel value in the series of pixel values;
classifying the obtained image into a member of one of three classes by
predicting, via a trained machine learning model having been trained using the calculated probability distribution and the determined spread value to predict which of the three classes the image belongs,
generating, for each of the three classes, a likelihood value representing a likelihood that the obtained image belongs in that class and,
classifying the obtained image into the one of the three classes based on the likelihood value meeting a predetermined threshold associated with the one of the three classes.