Back to results
US 12,705,899 B2Granted
Field vision control framework
AssigneeSchlumberger Technology Corporation
InventorsPrince Samuel, Debashis Gupta
Life of this patent
Estimating the term…
Abstract
A method can include receiving data for a field site by an edge framework gateway at the field site, where the data correspond to equipment operations at the field site and where the data include imagery data acquired by one or more cameras at the field site; performing image recognition on the imagery data to recognize gauges and to recognize content displayed on the gauges via execution of a machine learning model on the edge framework gateway; and issuing an instruction based at least in part on the recognized content displayed on the gauges.
Claims20
A method comprising:
- receiving data for a wellsite by an edge framework gateway at the wellsite, wherein the data correspond to equipment operations at the wellsite and wherein the data comprise imagery data acquired by one or more cameras at the wellsite;
- performing image recognition on the imagery data via execution of a machine learning model on the edge framework gateway, wherein performing the image recognition comprises:
- identifying a type of one or more gauges in the imagery data;
- accessing instructional information for the type of the one or more gauges, wherein the instructional information comprises a manual associated with the type of the one or more gauges;
- and
- recognizing content displayed on the one or more gauges in the imagery data using at least the instructional information;
- and
- issuing an instruction based at least in part on the recognized content displayed on the one or more gauges, wherein the instruction comprises a control action to control operation of one or more valves or pumps at the wellsite.
5 more elements
A system comprising:
- a processor;
- memory accessible to the processor;
- and
- processor-executable instructions stored in the memory and executable by the processor to instruct the system to:
- receive data for a wellsite by an edge framework gateway at the wellsite, wherein the data correspond to equipment operations at the wellsite and wherein the data comprise imagery data acquired by one or more cameras at the wellsite;
- perform image recognition on the imagery data via execution of a machine learning model on the edge framework gateway, wherein the processor-executable instructions executable by the processor to perform the image recognition further comprise instructions to:
- identify a type of one or more gauges in the imagery data;
- access instructional information for the type of the one or more gauges, wherein the instructional information comprises a manual associated with the type of the one or more gauges;
- and
- recognize content displayed on the one or more gauges in the imagery data using at least the instructional information;
- and
- issue an instruction based at least in part on the recognized content displayed on the one or more gauges, wherein the instruction comprises a control action to control operation of one or more valves or pumps at the wellsite.
9 more elements
One or more non-transitory computer-readable media comprising computer-executable instructions executable by a system to instruct the system to:
- receive data for a wellsite by an edge framework gateway at the wellsite, wherein the data correspond to equipment operations at the wellsite and wherein the data comprise imagery data acquired by one or more cameras at the wellsite;
- perform image recognition on the imagery data via execution of a machine learning model on the edge framework gateway, wherein the computer-executable instructions executable by the system to perform the image recognition further comprise instructions to:
- identify a type of one or more gauges in the imagery data, wherein the type of the one or more gauges comprises an analog gauge or a digital gauge and a gauge model of the analog gauge or digital gauge;
- access instructional information for the type of the one or more gauges, wherein the instructional information comprises a manual associated with the type of the one or more gauges;
- and
- recognize content displayed on the one or more gauges in the imagery data using at least the instructional information;
- and
- issue an instruction based at least in part on the recognized content displayed on the one or more gauges, wherein the instruction comprises a control action to control operation of one or more valves or pumps at the wellsite.
5 more elements
Dependencies are inferred from the claim text.