# US 7,609,874 B2 · System and method for prediction of pitting corrosion growth > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 7,609,874 B2 - Document ID: US7609874B2 - Status: Granted - Kind: B2 - Application number: 11/314,903 - Filed: 2005-12-21 - Published: 2007-06-21 - Granted: 2009-10-27 - Earliest claimed: 2005-12-21 - Assignees: Honeywell International Inc. - Inventors: Lalitha M. Eswara; Lokesh R. Boregowda; Thirumaran Ekambaram - Web page: https://patentgrep.com/patent/US7609874B2 ## CPC - G06V 10/54: relating to texture - G06V 2201/06: Recognition of objects for industrial automation - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A system and method predicts pitting corrosion growth using imaging technology. In an embodiment, a first module preprocesses images to locate seed points, and a second module models a corrosion life cycle of the seed points. In another embodiment, the seed points are identified with defined confidence levels, and the rate of change of features in corrosion images is analyzed to identify dominant life-cycle phases of corrosion and defining corrosion impact factor for corrosion quantification. ## Claims ### Claim 1 A computerized process comprising: capturing a first image in an image acquisition system, the image acquisition system including a processor; capturing successive images in the image acquisition system; preprocessing in the processor the first and successive images acquired by the image acquisition system to locate seed points, the preprocessing comprising subtracting said successive images from said first image; using said processor to subject output of said preprocessing to feature extraction; and using said processor to model a life cycle of corrosion resulting from said seed points; wherein said seed points are identified by identifying an expected direction of change in said seed points. ### Claim 2 (depends on 1) The process of claim 1, wherein said modeling comprises an analysis of the rate of change of said features of said corrosion. ### Claim 3 (depends on 1) The process of claim 1, wherein said preprocessing comprises: providing a reference image; providing an actual image; extracting potential patches and points from said actual image; calculating a correlation in the rate of change of said potential patches and points; identifying said patches and points as noise when there is no correlation; and identifying said patches and points as seed points when there is a correlation. ### Claim 4 (depends on 3) The process of claim 3, further comprising applying a threshold to said actual image in order to remove said noise. ### Claim 5 (depends on 1) The process of claim 1, wherein said seed points are identified by correlating a rate of change of features of said successive images. ### Claim 6 (depends on 1) The process of claim 1, wherein said preprocessing of said input corrosion images comprises: analyzing the pit shape of said corrosion; analyzing the pit size of said corrosion; analyzing the pit roughness of said corrosion; and analyzing the pit crack length of said corrosion. ### Claim 7 (depends on 6) The process of claim 6, wherein said pit shape is analyzed using deformable contours; said pit size is analyzed by measuring the diameter of the pit mouth; and said roughness is measured as a mean height of the features of said corrosion. ### Claim 8 (depends on 1) The process of claim 1, further comprising: weighting pits depending upon the phase that said pit is in; and determining the density of pits in a region; wherein said corrosion life cycle is modeled as a nucleation phase, a corrosion growth phase, and a crack propagation phase. ### Claim 9 (depends on 1) The process of claim 1, further comprising: quantifying fatigue by identifying a dominant phase of a plurality of pits; and calculating a corrosion impact factor. ### Claim 10 (depends on 9) The process of claim 9, wherein said corrosion impact factor is calculated as follows: CIF = ∑ i = 1 N ⁢ ⁢ 1 ⁢ ( W i * DPC i ) + ∑ g = 1 N ⁢ ⁢ 2 ⁢ ( W g * DPC g ) + ∑ c = 1 N ⁢ ⁢ 3 ⁢ ( W c * DPC c ) wherein The notation “i” indicates the nucleation phase; The notation “g” indicates the corrosion growth phase; The notation “c” indicates the crack propagation phase; The notation N1 represents the number of pits dominantly in the nucleation phase; The notation N2 represents the number of pits dominantly in the growth phase; The notation N3 represents the number of pits dominantly in the crack phase; The notation Wi is a weight derived from nucleation features and the dynamics and density of the pits; The notation Wg is a weight derived from growth features and the dynamics and density of the pits; The notation Wc is a weight derived from crack features and the dynamics and density of the pits; The notation DPCi is the phase contribution of a pit whose dominant phase is the nucleation phase; The notation DPCg is the phase contribution of a pit whose dominant phase is the corrosion growth phase; and The notation DPCc is the phase contribution of a pit whose dominant phase is the crack propagation phase. ### Claim 11 (depends on 1) The process of claim 1, wherein a three dimensional (3D) rendered corrosion image is generated from said input corrosion images. ### Claim 12 A computer-readable medium embodying a program comprising instructions thereon for executing a process comprising: capturing a first image in an image acquisition system, the image acquisition system including a processor; capturing successive images in the image acquisition system; preprocessing the first and successive images to locate seed points, the preprocessing comprising subtracting said successive images from said first image; subjecting output of said preprocessing to feature extraction; and modeling a life cycle of corrosion resulting from said seed points; wherein said seed points are identified by identifying an expected direction of change in said seed points. ### Claim 13 (depends on 12) The computer-readable medium of claim 12, wherein said modeling comprises an analysis of the rate of change of said features of said corrosion; and further wherein said preprocessing comprises: providing a reference image; providing an actual image; extracting potential patches and points from said actual image; calculating a correlation in the rate of change of said potential patches and points; identifying said patches and points as noise when there is no correlation; and identifying said patches and points as seed points when there is a correlation. ### Claim 14 (depends on 12) The computer-readable medium of claim 12, further comprising instructions for: quantifying fatigue by identifying a dominant phase of a plurality of pits; and calculating a corrosion impact factor; wherein said corrosion impact factor is calculated as follows: CIF = ∑ i = 1 N ⁢ ⁢ 1 ⁢ ( W i * DPC i ) + ∑ g = 1 N ⁢ ⁢ 2 ⁢ ( W g * DPC g ) + ∑ c = 1 N ⁢ ⁢ 3 ⁢ ( W c * DPC c ) wherein The notation “i” indicates the nucleation phase; The notation “g” indicates the corrosion growth phase; The notation “c” indicates the crack propagation phase; The notation N1 represents the number of pits dominantly in the nucleation phase; The notation N2 represents the number of pits dominantly in the growth phase; The notation N3 represents the number of pits dominantly in the crack phase; The notation Wi is a weight derived from nucleation features and the dynamics and density of the pits; The notation Wg is a weight derived from growth features and the dynamics and density of the pits; The notation Wc is a weight derived from crack features and the dynamics and density of the pits; The notation DPCi is the phase contribution of a pit whose dominant phase is the nucleation phase; The notation DPCg is the phase contribution of a pit whose dominant phase is the corrosion growth phase; and The notation DPCc is the phase contribution of a pit whose dominant phase is the crack propagation phase. ### Claim 15 A computerized method to predict corrosion comprising: using a processor for: capturing a plurality of images; comparing said captured images with a reference image; subtracting said captured images from said reference image; correlating between captured images to differentiate between corroded and non-corroded images; constructing sub-regions of said captured images; identifying regions of interest in said sub-regions to characterize corrosion seed points based on rates of change of features of said corrosion; identifying a dominant phase of corrosion at said seed points based on said rate of change of features; identifying the effect of said features on each of said seed points on proximate seed points and state of corrosion of said sub-region; and calculating an impact factor to predict the state of corrosion of said sub-regions. ### Claim 16 (depends on 15) The method of claim 15, wherein said subtraction is weighted. ### Claim 17 (depends on 15) The method of claim 15, wherein said correlating comprises higher-order statistical calculations. ### Claim 18 (depends on 15) The method of claim 15, wherein said sub-regions comprise a density of pits and a dominant phase of pits. ### Claim 19 (depends on 15) The method of claim 15, wherein said features comprise pit shape, pit size, pit roughness, crack length; and density of pits.