# US 12,002,250 B2 · Method for processing digital images

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,002,250 B2
- Document ID: US12002250B2
- Status: Granted
- Kind: B2
- Application number: 17/421,486
- Filed: 2020-01-08
- Published: 2022-03-03
- Granted: 2024-06-04
- Earliest claimed: 2019-01-08
- Assignees: Surys
- Applicants: Surys
- Inventors: Gaël Mahfoudi; Mohammed Amine Ouddan
- Web page: https://patentgrep.com/patent/US12002250B2

## CPC

- G06V 10/42: Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
- G06T 7/13: Edge detection
- G06V 10/426: Graphical representations
- G06V 10/462: {Salient features, e.g. scale invariant feature transforms [SIFT]}
- G06V 20/95: {Pattern authentication; Markers therefor; Forgery detection}
- G06V 40/165: {using facial parts and geometric relationships}
- G06V 40/171: {Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships}
- G06V 40/172: {Classification, e.g. identification}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method for processing a candidate digital image includes defining a set of noteworthy points in the candidate digital image. A set of at least three noteworthy points is selected to comprise a notable departure point, a notable arrival point, and a third notable point not aligned with the notable departure point and the notable arrival point. A set of at least one route, between the notable departure point and the notable arrival point, is defined. The route passes through all of the selected notable points. Local characteristics, of the pixels located along the route, are extracted. The signal, corresponding to the variation in the magnitude of the local characteristics as a function of each pixel along each defined route, is recorded in the form of a fingerprint.

## Claims

### Claim 1

A method for processing a candidate digital image, the method comprising:
defining a set of notable points on the candidate digital image;
selecting a set of at least three notable points, comprising:
a notable departure point,
a notable arrival point, and
a third notable point, distinct from the notable departure point and the notable arrival point, and not aligned with the notable departure point and the notable arrival point;
defining a set of at least one route between the notable departure point and the notable arrival point, the route being constrained by all of the selected notable points;
extracting local characteristics of pixels located along the defined at West one route;
recording, in fingerprint form, a signal corresponding to a variation in the local characteristics of each pixel along each of the defined at least one route;
comparing the fingerprint of the candidate digital image to a reference fingerprint; and
generating a signal, the value of which depends on the result of the comparison.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
defining a set of notable points on a reference digital image;
selecting a set of at least two notable points on the reference digital image;
defining a set of at least one route passing through the notable points selected on the reference digital image;
determining at least one local characteristic of each pixel along each of the defined at least one route passing through the notable points selected on the reference digital image; and
recording, as the reference fingerprint, a signal corresponding to a variation in the local characteristics, determined with regard to the reference digital image, as a function of each pixel along each of the defined at least one route on the reference digital image, the notable points selected on the reference digital image being the same as the notable points selected on the candidate digital image.

### Claim 3 (depends on 1)

The method of claim 1, wherein defining the set of the at least one route comprises defining a linear or curved route between two consecutive notable points.

### Claim 4 (depends on 1)

The method of claim 1, further comprising, prior to recording the fingerprint, filtering at least one among the candidate digital image and a reference digital image with a band-pass filter or a low-pass filter.

### Claim 5 (depends on 1)

The method of claim 1, wherein defining the set of notable points comprises applying at least one among algorithms based on:
a Harris detector,
a scale-invariant feature transform,
speeded-up robust features,
an algorithm for detecting biometric points,
a derivation operator, and
an edge-detecting algorithm.

### Claim 6 (depends on 1)

The method of claim 1, wherein comparing the fingerprint of the candidate digital image to the reference fingerprint comprises computing a Pearson correlation coefficient.

### Claim 7 (depends on 1)

The method of claim 1, wherein extracting the local characteristics comprises extracting at least one among: light intensity,
intensity of one or two predetermined colors,
a high-frequency density above a predetermined threshold in a set of pixels around the pixels on the defined at least one route,
an orientation of the edges of shapes in the candidate digital image and crossed by a curve of the defined at least one route at each point on the defined at least one route, and a direction of a lightness gradient at each point on the defined at least one route.

### Claim 8 (depends on 1)

The method of claim 1, wherein defining the set of at least one route between the notable departure point and the notable arrival point comprises defining a plurality of routes between the notable departure point and the notable arrival point.

### Claim 9 (depends on 1)

A non-transitory computer-readable medium comprising a computer program comprising program-code instructions for executing, on a computer, the method of claim 1.

### Claim 10 (depends on 3)

The method of claim 3, wherein defining the set of the at least one route comprises defining the curved route, defining the curved route comprising constraining the curved route with a set of Bezier curves that pass through the selected notable points or that minimize a distance to the selected notable points.

### Claim 11 (depends on 4)

The method of claim 4, wherein the filtering comprises decomposing the signal, which is recorded as the fingerprint, via a discrete cosine transform.

### Claim 12 (depends on 5)

The method of claim 5, wherein defining the set of notable points comprises applying the at least one among algorithms based on the derivation operator, the derivation operator being a difference-of-Gaussian operator.

### Claim 13 (depends on 5)

The method of claim 5, wherein defining the set of notable points comprises applying the at least one among algorithms based on the edge-detecting algorithm, the edge-detecting algorithm being a Laplacian-of-Gaussian edge-detecting algorithm.