# US 2019/0377928 A1 · Method for processing an image of a papillary impression > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0377928 A1 - Document ID: US20190377928A1 - Status: Published application - Kind: A1 - Application number: 16/429,875 - Filed: 2019-06-03 - Published: 2019-12-12 - Earliest claimed: 2018-06-08 - Assignees: Commissariat A L'energie Atomique Et Aux Energies Alternatives - Applicants: Commissariat A L'energie Atomique Et Aux Energies Alternatives - Inventors: Regis PERRIER; Mathilde BOURJOT - Web page: https://patentgrep.com/patent/US20190377928A1 ## CPC - G06V 40/1353: {Extracting features related to minutiae or pores} - G06V 40/1347: {Preprocessing; Feature extraction} - G06V 40/1382: {Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger} - G06V 10/993: {Evaluation of the quality of the acquired pattern} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract The invention relates to a method of processing a papillary print comprising: a) a step to detect a singular zone on the image, the singular zone being characterised by a position (p), and an orientation (θ) representative of a value of the gradient of the intensity of the image; b) a step to extract control patches, each control patch comprising one of the singular zones detected in step a); c) a step to project control patches in a reference base, so as to determine the projection coordinates of each control patch on said reference base, projection of a control patch on the reference base forming a projected patch; d) a step to calculate a difference in projection of control patches relative to the projected patches. ## Claims ### Claim 1 Method for processing an image of a papillary print designed to be executed by a computer, comprising: a) a step for detecting at least one singular zone on the image, each singular zone being characterised by a position (p), and an orientation (θ) representative of a value of the gradient of the intensity of the image at position (p); b) a step for extracting patches, called control patches, each control patch comprising one of the singular zones detected in step a); c) a step for projecting control patches in a base of reference patches, called reference base, so as to determine the projection coordinates of each control patch on said reference base, projection of a control patch on the reference base forming a projected patch, the reference patches base comprising at least two reference patches each associated with a different type of singular zone; d) a step for calculating a difference in projection of control patches relative to the projected patches. ### Claim 2 (depends on 1) Method according to claim 1, in which the reference patches base comprises reference patches associated with at least two order 2 and/or 3 singular zones. ### Claim 3 (depends on 1) Method according to claim 1, in which step b) for extracting control patches also includes a step to orient said patches along a principal axis so as to align the singular zones with said axis. ### Claim 4 (depends on 1) Method according to claim 1, in which step a) for detecting singular zones on the image comprises detection of a variation in the contrast of the intensity of the enrolment image. ### Claim 5 (depends on 4) Method according to claim 4, in which detection of the contrast variation includes the calculation of spatial derivatives of the intensity of each position of the enrolment image. ### Claim 6 (depends on 4) Method according to claim 4, in which step a) includes use of the Harris method. ### Claim 7 (depends on 1) Method according to claim 1, in which each control patch extracted in step b) is written in vector form, the vector form including intensities of each pixel of the patch considered, for components. ### Claim 8 (depends on 1) Method according to claim 1, in which step d) for calculating the projection difference includes the calculation of an error called the construction error, said construction error including the calculation of the normalised sum of the differences between each control patch and its reconstructed patch. ### Claim 9 (depends on 8) Method according to claim 8, in which an enrolment image is qualified as a quality image when the construction error is less than a predetermined value. ### Claim 10 (depends on 1) Method according to claim 1, in which the method also includes a step e) for quantifying discriminating information contained in the enrolment image, said discriminating information relating to singular points actually detected in step a). ### Claim 11 (depends on 10) Method according to claim 10, in which step e) includes the evaluation of the dispersion of at least two projection coordinates, for the set of control patches, relative to at least two reference patches called principal reference patches. ### Claim 12 (depends on 1) Method according to claim 1, in which the singular zones detected in step a) comprise order 2 or higher characteristics. ### Claim 13 (depends on 1) Method according to claim 1, in which the reference base is constructed from at least one reference image, of at least one papillary print. ### Claim 14 (depends on 13) Method according to claim 13, in which construction of the reference base comprises the following steps: a1) a step for detecting singular zones on the at least one reference image, each singular zone called a primitive singular zone being characterised by a position p1, and an orientation θ1 representative of a value of the gradient of the intensity of the reference image at position p1; b1) a patch extraction step to extract primitive patches, each of the primitive patches being centred on one of the primitive singular zones detected in step a1); c1) a step for analysing the set of primitive patches into principal components oriented so as to extract the reference base and the associated reference eigenvalues. ### Claim 15 (depends on 14) Method according to claim 14, in which the singular zones detected in step a1) comprise order 1 or higher characteristics. ### Claim 16 (depends on 14) Method according to claim 14, in which step b1) for extracting primitive patches also includes a step to orient said patches along the principal axis so as to align the primitive singular zones with said axis. ### Claim 17 (depends on 1) Computer program comprising instructions that, when the program is run on a computer, lead to the implementation of the method according to claim 1. ### Claim 18 (depends on 17) Electronic device on which the computer program according to claim 17 is installed, and provided with a papillary prints sensor for which print images acquired by the print sensor will be processed by said computer program.