US 11,670,073 B2
System and method for detection of carbonate core features from core images
Saleh Z. Alatwah, Dhahran (SA)
Assigned to SAUDI ARABIAN OIL COMPANY, Dhahran (SA)
Filed by Saudi Arabian Oil Company, Dhahran (SA)
Filed on Aug. 25, 2020, as Appl. No. 17/1,932.
Prior Publication US 2022/0067420 A1, Mar. 3, 2022
Int. Cl. G06V 10/50 (2022.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06F 18/40 (2023.01); G06F 18/21 (2023.01); E21B 49/02 (2006.01)
CPC G06V 10/50 (2022.01) [G06F 18/2163 (2023.01); G06F 18/40 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); E21B 49/02 (2013.01); E21B 2200/20 (2020.05)] 20 Claims
OG exemplary drawing
 
1. A method for detection of carbonate core features from core images, comprising:
collecting a variety of feature images corresponding to a variety of carbonate core features to define training sets of sample feature images;
calculating a mean value of image for each of the training sets corresponding to each of the plurality of carbonate core features;
training an artificial intelligence (AI) model to predict features and automatically create the carbonate core description based on images cropped randomly from the variety of carbonate core features;
separating an input carbonate core image into a plurality of first blocks, each of the plurality of first blocks having a carbonate core feature and a first block size;
inputting an image of each of the separated plurality of first blocks into artificial intelligence (AI) model;
subtracting a mean value of an image in each of the plurality of first blocks from the mean value of image calculated for each of the training sets to obtain subtracted values; and
generating the carbonate core description for the input carbonate core image based on the mean value of image calculated for each of the training sets that yields the lowest subtracted value.