-
公开(公告)号：US20190138786A1
公开(公告)日：2019-05-09
申请号：US15997917
申请日：2018-06-05
Applicant: SIGHTLINE INNOVATION INC.
Inventor： Wallace TRENHOLM , Mark ALEXIUK , Hieu DANG , Siavash MALEKTAJI , Kamal DARCHINIMARAGHEH
Abstract: A method and system for analysis of an object of interest in a scene using 3D reconstruction. The method includes: receiving image data comprising a plurality of images captured of the scene, the image data comprising multiple perspectives of the scene; generating at least one reconstructed image by determining three-dimensional structures of the object from the imaging data using a reconstruction technique, the three-dimensional structures comprising depth information of the object; identifying the object from each of the reconstructed images, using a trained machine learning model, by segmenting the object in the reconstructed image, segmentation comprises isolating patterns in the reconstructed image that are classifiable as the object, the machine learning model trained using previous reconstructed multiple perspective images with identified objects; and outputting the analysis of the reconstructed images.
-
公开(公告)号：US20190019061A1
公开(公告)日：2019-01-17
申请号：US15997966
申请日：2018-06-05
Applicant: SIGHTLINE INNOVATION INC.
Inventor： Wallace TRENHOLM , Mark ALEXIUK , Hieu DANG , Siavash MALEKTAJI , Kamal DARCHINIMARAGHEH
Abstract: A method and system for increasing data quality of a dataset for semi-supervised machine learning analysis. The method includes: receiving known class label information for a portion of the data in the dataset; receiving clustering parameters from a user; determining a data cleanliness factor, and where the data cleanliness factor is below a predetermined cleanliness threshold: assigning data without class label information as a data point to a cluster using the clustering parameters, each cluster having a cluster class label associated with such cluster; and determining a measure of assignment, and where the measure of assignment for each data point is below a predetermined assignment threshold, receiving a class label for such data points, otherwise, assigning the respective cluster class label to each data point with the respective measure of assignment below the predetermined assignment threshold; and otherwise, outputting the dataset with associated class labels for machine learning analysis.
-
公开(公告)号：US20200167656A1
公开(公告)日：2020-05-28
申请号：US16613843
申请日：2018-05-16
Applicant: SIGHTLINE INNOVATION INC.
Inventor： Wallace TRENHOLM , Mark ALEXIUK , Hieu DANG , Siavash MALEKTAJI , Kamal DARCHINIMARAGHEH
Abstract: A system and method for non-destructive optical coherence tomography (OCT) is provided. The system includes: an input interface for receiving OCT data including at least a C-scan; a processing unit executable to detect a feature on a surface or subsurface of the object, trained using a training set and configured to: separate the C-scan into A-scans; using a neural network, successively analyze each A-scan to detect the presence of an A-scan feature associated with the object; separate the C-scan into B-scans; segment each of the B-scans to determine thresholds associated with the object; using a neural network, successively analyze each segmented B-scan to detect the presence of an B-scan feature associated with the object; convert the C-scan to one or more two-dimensional representations; and using a neural network, detect the presence of an C-scan feature associated with the object.
-
公开(公告)号：US20190139214A1
公开(公告)日：2019-05-09
申请号：US16006043
申请日：2018-06-12
Applicant: SIGHTLINE INNOVATION INC.
Inventor： Wallace TRENHOLM , Lorenzo PONS , Mark ALEXIUK , Hieu DANG , Siavash MALEKTAJI , Kamal DARCHINIMARAGHEH
Abstract: A method and system for analysis of interferometric domain optical coherence tomography (OCT) data of an object. The method includes: receiving the OCT data comprising one or more A-scans; successively analyzing each of the one or more A-scans, using a trained feed-forward neural network, to detect one or more features associated with the object by associating A-scan raw data with a descriptor for each of the one or more features, the feed-forward neural network trained using previous A-scans with one or more known features; generating location data associated with the one or more features for localizing the one or more features in the one or more A-scans; and outputting the feature detection and the location data.
-
-
-
Search Results
4
records
(took 0.009 seconds)
