US 12,254,408 B2
Patterned substrate for use in image-based classification of rock cuttings
Richa Sharma, Cambridge, MA (US); and Quincy K. Elias, Hyde Park, MA (US)
Assigned to SCHLUMBERGER TECHNOLOGY CORPORATION, Sugar Land, TX (US)
Appl. No. 17/756,393
Filed by Schlumberger Technology Corporation, Sugar Land, TX (US)
PCT Filed Dec. 15, 2020, PCT No. PCT/US2020/065140
§ 371(c)(1), (2) Date May 24, 2022,
PCT Pub. No. WO2021/126866, PCT Pub. Date Jun. 24, 2021.
Claims priority of provisional application 62/948,504, filed on Dec. 16, 2019.
Prior Publication US 2023/0030441 A1, Feb. 2, 2023
Int. Cl. G01N 33/24 (2006.01); G06N 3/08 (2023.01); G06T 5/00 (2024.01); G06T 7/60 (2017.01); G06V 10/22 (2022.01)
CPC G06N 3/08 (2013.01) [G01N 33/24 (2013.01); G06T 7/60 (2013.01); G06V 10/22 (2022.01); G06T 2207/30181 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A method for automated classification of rock cuttings, comprising:
forming a substrate having a patterned top surface;
while using the patterned top surface of the substrate to support a rock cutting, producing, by an image acquisition system, a sequence of images of the rock cutting using focal plane stacking;
processing, by one or more computers, the sequence of images by aligning and merging the images to generate a sharpened image of the rock cutting;
inputting as training data, by the one or more computers, the sharpened image of the rock cutting into an automated texture-based deep learning classifier of rock cuttings; and
classifying a rock cutting using the automated texture-based deep learning classifier that has been trained using the sharpened image.