US 12,406,439 B2
Digital outcrop model reconstructing method
Andre Luiz Durante Spigolon, Rio de Janeiro (BR); Luiz Gonzaga Da Silveira, Jr., São Leopoldo (BR); Caroline Lessio Cazarin, Rio de Janeiro (BR); Leonardo Bachi, São Leopoldo (BR); and Mauricio Roberto Veronez, São Leopoldo (BR)
Assigned to PETRÓLEO BRASILEIRO S.A.—PETROBRAS, Rio de Janeiro (BR); and UNIVERSIDADE DO VALE DO RIO DOSSINOS—UNISINOS, São Leopoldo (BR)
Filed by PETRÓLEO BRASILEIRO S.A.—PETROBRAS, Rio de Janeiro (BR); and UNIVERSIDADE DO VALE DO RIO DOS SINOS—UNISINOS, São Leopoldo (BR)
Filed on Nov. 28, 2023, as Appl. No. 18/521,656.
Claims priority of application No. BR 1020220244014 (BR), filed on Nov. 29, 2022.
Prior Publication US 2024/0177422 A1, May 30, 2024
Int. Cl. G06T 17/20 (2006.01); G06T 3/4023 (2024.01); G06T 7/10 (2017.01); G06V 10/24 (2022.01)
CPC G06T 17/205 (2013.01) [G06T 3/4023 (2013.01); G06T 7/10 (2017.01); G06V 10/24 (2022.01); G06T 2207/20212 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A Digital Outcrop Model (DOM) reconstructing method, comprising:
placing at least one image acquisition device for each image of a plurality of images to be captured;
capturing the plurality of images;
processing pairs of captured images;
aligning the plurality of images by identifying a plurality of homologous points between them;
generating at least one alignment sparse cloud;
performing densification of the at least one alignment sparse cloud to transform the at least one sparse cloud into at least one dense cloud;
performing a first segmentation of at least one dense cloud into at least two dense cloud portions, based on at least two categories of geological relevance levels,
wherein the at least two categories of geological relevance levels comprise a high geological relevance category and a low geological relevance category;
carrying out a first decimation of at least one dense cloud portion in the low geological relevance category;
joining the at least two dense cloud portions and obtaining at least one new dense cloud, wherein the at least two dense cloud portions comprise at least one dense cloud portion of the high geological relevance category and at least one dense cloud portion of the low geological relevance category, wherein at least one dense cloud portion of the low geological relevance category has undergone the step of carrying out a first decimation;
generating at least one three-dimensional triangular mesh from at least one new dense cloud;
performing a second segmentation of the three-dimensional triangular mesh into at least two three-dimensional triangular mesh portions, based on at least two categories of geological relevance levels, wherein the at least two categories of geological relevance levels comprise a high geological relevance category and a low geological relevance category;
carrying out a second decimation of at least one portion of a three-dimensional triangular mesh in the low geological relevance category;
performing a third decimation of at least one portion of a three-dimensional triangular mesh in the high geological relevance category;
carrying out a first texturing of at least one portion of a three-dimensional triangular mesh in the high geological relevance category, which has undergone the step of carrying out a third decimation;
carrying out a second texturing of at least one portion of a three-dimensional triangular mesh in the low geological relevance category, which has undergone the step of carrying out a second decimation; and
joining the at least two three-dimensional triangular mesh portions, wherein the at least two three-dimensional triangular mesh portions comprise the at least one three-dimensional triangular mesh portion of the high geological relevance category and the at least one three-dimensional triangular mesh portion of the low geological relevance category, which have been textured, and forming at least one object comprising at least one Digital Outcrop Model (DOM).