Systems and methods for registering headset system
Arrangements described herein relate to systems, apparatuses, and methods for a headset system that includes a transducer, a camera configured to capture image data of a subject, and a controller configured to detect one or more fiducial markers placed on the subject using the captured image data, and register the transducer with respect to the subject based on the detected fiducial markers.
1. A method for operating a transducer of a headset system that collects ultrasound data from a subject, the method comprising:
detecting at least one fiducial marker placed on the subject using image data, the image data comprising two images of the subject from different viewpoints; and
determining a scanning area of the subject bounded by the at least one fiducial marker; and
restricting the transducer from traveling beyond the scanning area while the transducer collects the ultrasound data.
2. The method of claim 1 , further comprising determining a plurality of candidate regions using the image data based on machine vision.
3. The method of claim 2 , wherein the plurality of candidate regions are identified using the image data based on a Maximally Stable Extremal Regions (MSER) algorithm.
4. The method of claim 2 , wherein each of the plurality of candidate regions comprises a list of pixels.
5. The method of claim 2 , further comprising
sorting the plurality of candidate regions into candidate region pairs;
determining features for each of the candidate region pairs;
determining a probability that each of the candidate region pairs corresponds to the at least one fiducial marker; and
selecting two or more candidate regions based on the probability.
6. The method of claim 5 , wherein sorting the plurality of candidate regions into the candidate region pairs comprises determining permutations of pairings of the plurality of candidate regions.
7. The method of claim 5 , wherein each of the features comprise one or more of a location, a size, a shape, an image intensity, or a local texture.
8. The method of claim 7 , wherein the probability that each of the candidate region pairs corresponds to the at least one fiducial marker is determined based on the one or more of the location, the size, the shape, the image intensity, or the local texture.
9. The method of claim 5 , wherein the probability that each of the candidate region pairs corresponds to the at least one fiducial marker is determined based on a weight assigned to each of the features.
10. A non-transitory computer-readable medium having computer-readable instructions, such that when executed, causes a processor of a headset system to operate a transducer of the headset system that collects ultrasound data from a subject by:
detecting at least one fiducial marker placed on the subject using image data, the image data comprising two images of the subject from different viewpoints; and
determining a scanning area of the subject bounded by the at least one fiducial marker; and
restricting the transducer from traveling beyond the scanning area while the transducer collects the ultrasound data.
11. The non-transitory computer-readable medium of claim 10 , wherein the processor is further caused to determine a plurality of candidate regions using the image data based on machine vision.
12. The non-transitory computer-readable medium of claim 11 , wherein the plurality of candidate regions are identified using the image data based on a Maximally Stable Extremal Regions (MSER) algorithm.
13. The non-transitory computer-readable medium of claim 11 , wherein each of the plurality of candidate regions comprises a list of pixels.
14. The non-transitory computer-readable medium of claim 11 , the processor is further caused to:
sort the plurality of candidate regions into candidate region pairs;
determine features for each of the candidate region pairs;
determine a probability that each of the candidate region pairs corresponds to the at least one fiducial marker; and
select two or more candidate regions based on the probability.
15. The non-transitory computer-readable medium of claim 14 , wherein sorting the plurality of candidate regions into the candidate region pairs comprises determining permutations of pairings of the plurality of candidate regions.
16. The non-transitory computer-readable medium of claim 14 , wherein each of the features comprise one or more of a location, a size, a shape, an image intensity, or a local texture.
17. The non-transitory computer-readable medium of claim 16 , wherein the probability that each of the candidate region pairs corresponds to the at least one fiducial marker is determined based on the one or more of the location, the size, the shape, the image intensity, or the local texture.
18. The non-transitory computer-readable medium of claim 15 , wherein the probability that each of the candidate region pairs corresponds to the at least one fiducial marker is determined based on a weight assigned to each of the features.
19. A system, comprising:
a controller configured to:
detect at least one fiducial marker placed on the subject using image data, the image data comprising two images of the subject from different viewpoints; and
determine a scanning area of the subject bounded by the at least one fiducial marker; and
restrict the transducer from traveling beyond the scanning area while the transducer collects the ultrasound data.
20. The system of claim 19 , wherein the processor is further caused to determine a plurality of candidate regions using the image data based on machine vision.