# US 10,373,301 B2 · Structural hot spot and critical location monitoring system and method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,373,301 B2 - Document ID: US10373301B2 - Status: Granted - Kind: B2 - Application number: 14/036,308 - Filed: 2013-09-25 - Published: 2015-03-26 - Granted: 2019-08-06 - Earliest claimed: 2013-09-25 - Assignees: Sikorsky Aircraft Corporation - Applicants: Sikorsky Aircraft Corporation - Inventors: Zaffir A. Chaudhry; Alan Matthew Finn; Ziyou Xiong; Hongcheng Wang; Patrick Louis Clavette - Web page: https://patentgrep.com/patent/US10373301B2 ## CPC - G06T 7/0004: {Industrial image inspection} - G01M 5/0033: {by determining damage, crack or wear} - G06T 7/001: {using an image reference approach} - G01M 5/0091: {by using electromagnetic excitation or detection} - G01M 11/081: {by using a contact-less detection method, i.e. with a camera} - G06T 2207/10024: Color image - G06T 2207/10028: Range image; Depth image; 3D point clouds - G06T 2207/10048: Infrared image - G06T 2207/20076: Probabilistic image processing - G06T 2207/30164: Workpiece; Machine component - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G01M: TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR ## Abstract A method for detecting a crack in a structural component includes receiving, with a processor, signals indicative of at least one image for a critical location in the structural component; determining, with the processor, at least one shape in the at least one image, the at least one shape being representative of a structure of the critical location; representing, with the processor, at least one region around the structure into a matrix; and applying, with the processor, image processing on the matrix to detect cracks in the at least one region of the structural component. ## Claims ### Claim 1 A method for detecting cracks in a structural component, comprising: applying a detection algorithm that includes a low level image processing and a high level image processing; identifying, using the low level image processing algorithm, a critical location of a structural component by detecting repetitive shapes in an area of the structural component where load is transferred; directing, by the detection algorithm, a camera toward the critical location in response to determining the critical location; receiving, with a processor, signals indicative of at least one camera image for the critical location in the structural component, wherein the camera image of the critical location includes a location where metal fatigue of the structural component is likely to occur; determining, with the processor, at least one shape in the at least one image, the at least one shape being representative of a structure of the critical location; transmitting, with the processor, the camera image for analysis of the identified critical location based on analytical modeling, experiential information and historical information; producing, with the processor, one or more images representing a large input matrix; wherein the images represents at least one region around the structure of the critical location in the matrix; applying, with the high level image processing, a contextual high level driven model analysis on the matrix, wherein the contextual model includes steps of image processing on the matrix including performing a robust principle component analysis using an output from a contextual model applied to the matrix, to decompose the matrix into a normal component and a sparse component, the sparse component containing information pertaining to cracks or deformation of the structural component; and applying a thresholding scheme, with the processor, to the sparse component to detect cracks in the at least one region of the structure of the critical location. ### Claim 2 (depends on 1) The method of claim 1, wherein the applying of the contextual model further comprises: applying at least one of a Geometric Layout Context model to the matrix, a Force Context model to the matrix, and a Fretting Context model to the matrix. ### Claim 3 (depends on 2) The method of claim 2, wherein the applying of the Geometric Layout Context model further comprises applying prior information related to an area around the critical location. ### Claim 4 (depends on 2) The method of claim 2, wherein the applying of the Force Context model further comprises: determining the crack in a direction orthogonal to an application of force. ### Claim 5 (depends on 2) The method of claim 2, wherein the applying of the Fretting Context model further comprises: predicting coefficient of friction characteristics within a rivet. ### Claim 6 (depends on 1) The method of claim 1, further comprising: determining a sparse component from the matrix, wherein the sparse component is representative of cracks in the structural component. ### Claim 7 (depends on 1) The method of claim 1, wherein the receiving of the signals indicative of the image further comprises: capturing with a camera two-dimensional or three-dimensional images of the structural component. ### Claim 8 (depends on 1) The method of claim 1, wherein the applying of the image processing further comprises: implementing at least one of a Hough transform algorithm or a Transform Invariant Low-rank Textures algorithm to the image. ### Claim 9 (depends on 1) The method of claim 1, wherein the applying of the contextual model further comprises at least one of: determining a change in a feature in the at least one image relative to historical information regarding the feature for the at least one image; determining an estimated value for a displacement change in the at least one image relative to a historical estimated value for the displacement change in the at least one image; or applying an image matching algorithm to the at least one image and historical information for the at least one image. ### Claim 10 A system for detecting cracks in a critical location of a structural component, comprising: an image processing computer including a processor and memory; and a camera that receives signals indicative of at least one image for the critical location; apply a detection algorithm that includes a low level image processing and a high level image processing; identify, using the low level image processing algorithm, a critical location of a structural component by detecting repetitive shapes in an area of the structural component where load is transferred; direct, by the detection algorithm, a camera toward the critical location in response to determining the critical location; receive, with a processor, signals indicative of at least one camera image for the critical location in the structural component, wherein the camera image of the critical location includes a location where metal fatigue of the structural component is likely to occur; determine, with the processor, at least one shape in the at least one image, the at least one shape being representative of a structure of the critical location; transmit, with the processor, the camera image for analysis of the identified critical location based on analytical modeling, experiential information and historical information; produce, with the processor, one or more images representing a large input matrix; wherein the images represents at least one region around the structure of the critical location in the matrix; apply, with the high level image processing, a contextual high level driven model analysis on the matrix, wherein the contextual model includes steps of image processing on the matrix including performing a robust principle component analysis using an output from a contextual model applied to the matrix, to decompose the matrix into a normal component and a sparse component, the sparse component containing information pertaining to cracks or deformation of the structural component; and apply a thresholding scheme, with the processor, to the sparse component to detect cracks in the at least one region of the structure of the critical location. ### Claim 11 (depends on 10) The system of claim 10, wherein the system is configured to apply at least one of a Geometric Layout Context model to the matrix, a Force Context model to the matrix, and a Fretting Context model to the matrix. ### Claim 12 (depends on 10) The system of claim 10, wherein the system is configured to determine the crack in a direction orthogonal to an application of force in the Force Context model. ### Claim 13 (depends on 10) The system of claim 10, wherein the system is configured to determine a sparse component from the matrix, wherein the sparse component is representative of cracks in the structural component. ### Claim 14 (depends on 10) The system of claim 10, wherein the system is configured to capture with the camera two-dimensional or three-dimensional images of the structural component. ### Claim 15 (depends on 10) The system of claim 10, wherein the system is configured to implement at least one of a Hough transform algorithm or a Transform Invariant Low-rank Textures algorithm to the image. ### Claim 16 (depends on 10) The system of claim 10, wherein the system is configured for at least one of determine a change in a feature in the at least one image relative to historical information regarding the feature for the at least one image; determine an estimated value for a displacement change in the at least one image relative to a historical estimated value for the displacement change in the at least one image; or apply an image matching algorithm to the at least one image and historical information for the at least one image.