# US 2023/0267643 A1 · Interface for Digitization and Detection of Rip Currents Within Optical Imagery by Way of a Fuzzy Set > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0267643 A1 - Document ID: US20230267643A1 - Status: Published application - Kind: A1 - Application number: 18/112,472 - Filed: 2023-02-21 - Published: 2023-08-24 - Earliest claimed: 2022-02-22 - Assignees: The Government of the United States of America, as represented by the Secretary of the Navy - Applicants: The Government of the United States of America, as represented by the Secretary of the Navy - Inventors: Corey Maryan; Christopher Michael; Steven Dennis; Sarah Trimble - Web page: https://patentgrep.com/patent/US20230267643A1 ## CPC - G06T 7/73: using feature-based methods - G01C 13/006: {horizontal movement} - G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components - G06V 10/763: {Non-hierarchical techniques, e.g. based on statistics of modelling distributions} - G06T 2207/10032: Satellite or aerial image; Remote sensing - G06T 2207/20061: Hough transform - G06T 2207/20081: Training; Learning - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method of identifying a rip current that includes receiving an image comprising a body of water, and identifying, in the image an estimated location of a rip current associated with the body of water, wherein the identification comprises a plurality of displayed vertices based on least one of a rip length or a rip width. The method may include digitizing the estimated location of the rip current, the estimated location having highest, lower, and no confidence boundaries. The method may include generating, based on the digitized estimated location, a set of fuzzy-set scheme labels for one or more pixels in the image based on a respective confidence boundary, where the fuzzy-set scheme labels are based on a pixel interpolation associated with the lower confidence boundary, and generating, based on the generated fuzzy-set scheme labels, a refined digitizing of the estimated location of the rip current. ## Claims ### Claim 1 A method of identifying a rip current, comprising: receiving, by a computing device, an image comprising a body of water; identifying, by the computing device, in the image an estimated location of a rip current associated with the body of water, wherein the identification comprises a plurality of displayed vertices based on least one of a rip length or a rip width; digitizing, by the computing device and based on the displayed vertices, the estimated location of the rip current, the estimated location comprising a highest confidence boundary, a lower confidence boundary, and a no confidence boundary, generating, by the computing device and based on the digitized estimated location, a set of fuzzy-set scheme labels for one or more pixels in the image based on a respective confidence boundary, wherein the fuzzy-set scheme labels are based on a pixel interpolation associated with the lower confidence boundary; and generating, by the computing device and based on the generated fuzzy-set scheme labels, a refined digitizing of the estimated location of the rip current associated with the body of water. ### Claim 2 (depends on 1) The method of claim 1, wherein the digitizing comprises using a machine learning model to digitize the estimated location. ### Claim 3 (depends on 1) The method of claim 1, wherein generating the refined digitizing of the estimated location comprises using a machine learning model to digitize the estimated location. ### Claim 4 (depends on 1) The method of claim 1, further comprising closing one or more Bezier curves around the vertices, wherein generating the set of fuzzy-set scheme labels is based on the one more Bezier curves. ### Claim 5 (depends on 1) The method of claim 1, wherein identifying the estimated location comprises generating a cross-segment of a segment of the rip length and a segment of the rip width, wherein the cross-segment comprises four vertices associated with the rip length and the rip width and is used to digitize the rip current. ### Claim 6 (depends on 5) The method of claim 5, further comprising closing one or more Bezier curves around endpoints of the cross-segment, wherein generating the set of fuzzy-set scheme labels comprises gradually decreasing the fading the fuzzy labels from a cross point of the cross-segment. ### Claim 7 (depends on 5) The method of claim 5, further comprising determining an orientation of the rip current based on one or more angles associated with the cross-segment. ### Claim 8 (depends on 1) The method of claim 1, further comprising determining an orientation of the rip current via feature extraction of the fuzzy-set labeling. ### Claim 9 (depends on 1) The method of claim 1, wherein generating the set of fuzzy-set scheme labels comprises one or more of the following: cross-point detection, orientation detection, initial vertex placement, or vertex placement optimization. ### Claim 10 (depends on 9) The method of claim 9, wherein cross-point detection comprises a thresholded local maxima, orientation detection comprises a partial Hough transform, initial vertex placement comprises a local search for a rough boundary associated with the rip current, and vertex placement optimization comprises a local max coefficient of determination. ### Claim 11 (depends on 1) The method of claim 1, further comprising performing a water-based operation based on the refined digitizing of the estimated location of the rip current associated with the body of water. ### Claim 12 (depends on 1) The method of claim 1, wherein generating the set of fuzzy-set scheme labels comprises identifying an orientation associated with the rip current. ### Claim 13 (depends on 1) The method of claim 1, wherein the image comprises a satellite image. ### Claim 14 (depends on 1) The method of claim 1, wherein the image comprises a photograph. ### Claim 15 (depends on 1) The method of claim 1, wherein digitizing the estimated location of the rip current comprises generating (1) a cross point pixel denoting an area of rip current with highest confidence, (2) a threshold curve designating a rough boundary of pixels of the rip current, and (3) a root curve designating pixels of the image outside of the rip current. ### Claim 16 (depends on 1) The method of claim 1, further comprising training a classifier based on input from a user, the input from the user being associated with the identified estimated location of a rip current and the digitized estimated location. ### Claim 17 (depends on 16) The method of claim 16, further comprising generating, via the classifier, the refined digitizing of the estimated location of the rip current associated with the body of water.