Methods and devices for automated characterization of drill cuttings
AssigneeUniversity of Wyoming
InventorsAmin AMOOIE, Yanbin GONG, Mohammad PIRI
Life of this patent
Estimating the term…
Abstract
A method and system for image processing are disclosed. The method includes image processing by one or more processing units which includes obtaining a set of one or more raw images of a set of drill cutting samples, processing the set of one or more raw images to generate at least a trinary image dataset, a binary image dataset, or both, and extracting a set of features from at least the trinary image dataset, the binary image dataset, or both. The set of features are clustered to associate each of the drill cuttings with at least one representative class, and the set of features are output and the at least one representative class is associated with each of the drill cuttings.
Claims15
A method for image processing by one or more processing units, comprising:
- obtaining a set of one or more raw images of a set of drill cutting samples;
- processing the set of one or more raw images to generate at least a trinary image dataset, a binary image dataset, or both;
- extracting a set of features from at least the trinary image dataset, the binary image dataset, or both;
- clustering the set of features to associate each of the drill cuttings with at least one representative class;
- and
- outputting the set of features and the at least one representative class associated with each of the drill cuttings.
A non-transitory computer-readable medium comprising computer-executable instructions that, when executed by one or more processors, cause one or more processing units to:
- obtain a set of one or more raw images of a set of drill cutting samples;
- process the set of one or more raw images to generate at least a trinary image dataset, a binary image dataset, or both;
- extract a set of features from at least the trinary image dataset, the binary image dataset, or both;
- cluster the set of features to associate each of a drill cutting of the set of drill cuttings with at least one representative class;
- and
- output the set of features and the at least one representative class associated with each of a drill cutting of the set of drill cuttings.
An apparatus for image processing comprising a memory and one or more processing units, the one or more processing units configured to cause the apparatus to:
- obtain a set of one or more raw images of a set of drill cutting samples;
- process the set of one or more raw images to generate at least a trinary image dataset, a binary image dataset, or both;
- extract a set of features from at least the trinary image dataset, the binary image dataset, or both;
- cluster the set of features to associate each of a drill cutting of the set of drill cuttings with at least one representative class;
- and
- output the set of features and the at least one representative class associated with each of a drill cutting of the set of drill cuttings.
Dependencies are inferred from the claim text.