US 12,131,526 B2
Systems and methods for segmenting rock particle instances
Tetsushi Yamada, Cambridge, MA (US); and Simone Di Santo, Dhahran (SA)
Assigned to SCHLUMBERGER TECHNOLOGY CORPORATION, Sugar Land, TX (US)
Filed by Schlumberger Technology Corporation, Sugar Land, TX (US)
Filed on Jan. 7, 2022, as Appl. No. 17/647,407.
Prior Publication US 2023/0220761 A1, Jul. 13, 2023
Int. Cl. G06V 10/82 (2022.01); E21B 44/00 (2006.01); G01N 15/1433 (2024.01); G06N 3/08 (2023.01); G06T 7/00 (2017.01); G06T 7/73 (2017.01); G06V 10/77 (2022.01); G06V 10/774 (2022.01); G06V 20/10 (2022.01)
CPC G06V 10/82 (2022.01) [E21B 44/00 (2013.01); G06N 3/08 (2013.01); G06T 7/0004 (2013.01); G06T 7/74 (2017.01); G06V 10/7715 (2022.01); G06V 10/774 (2022.01); G06V 20/10 (2022.01); E21B 2200/22 (2020.05); G01N 15/1433 (2024.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30181 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
training, via an analysis and control system, a neural network model using a first set of photographs, wherein each photograph of the first set of photographs depicts a first set of objects and include one or more annotations relating to each object of the first set of objects;
manually arranging a plurality of cuttings in a relatively sparse configuration on a tray having a relatively vivid background color;
generating a second set of photographs that depict the plurality of cuttings arranged in the relatively sparse configuration on the tray having the relatively vivid background color;
automatically creating, via the analysis and control system, mask images corresponding to the plurality of cuttings depicted by the second set of photographs;
enabling, via the analysis and control system, manual fine tuning of the mask images corresponding to the plurality of cuttings depicted by the second set of photographs;
re-training, via the analysis and control system, the neural network model using the second set of photographs, wherein the re-training is based at least in part on the manual fine tuning of the mask images corresponding to the plurality of cuttings depicted by the second set of photographs; and
identifying, via the analysis and control system, one or more individual cuttings in a third set of photographs using the re-trained neural network model.