# US 11,443,542 B2 · Fingerprint recognition method and electronic interactive apparatus thereof

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

- Number: US 11,443,542 B2
- Document ID: US11443542B2
- Status: Granted
- Kind: B2
- Application number: 17/143,170
- Filed: 2021-01-07
- Published: 2021-11-25
- Granted: 2022-09-13
- Earliest claimed: 2020-05-19
- Assignees: Industrial Technology Research Institute
- Applicants: Industrial Technology Research Institute
- Inventors: Jian-Lung Chen; Chih-Chia Chang; Yu-Hsin Lin; Yu-Ju Chao; Ting-Hsun Cheng
- Web page: https://patentgrep.com/patent/US11443542B2

## CPC

- G06V 40/1335: {Combining adjacent partial images (e.g. slices) to create a composite input or reference pattern; Tracking a sweeping finger movement}
- G06F 3/041: Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
- G06V 40/1347: {Preprocessing; Feature extraction}
- G06V 40/1365: {Matching; Classification}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A fingerprint recognition method is provided. The method includes obtaining a plurality of fingerprint images by sensing a finger of a user, respectively calculating geometric center points corresponding to the fingerprint images, and calculating positions and offsets of the fingerprint images according to the geometric center points. The method also includes filling signals in the fingerprint images into a part of pixels in a pixel array according to the positions and the offsets of the fingerprint images, and obtaining signals of other pixels in the pixel array by inputting the signals filled in the part of pixels in the pixel array into an artificial intelligence engine. The method further includes generating a candidate fingerprint image and recognizing a user based on the candidate fingerprint image.

## Claims

### Claim 1

A fingerprint recognition method, comprising:
detecting a touch signal corresponding to a user on a touch display;
sensing a finger of the user to obtain a plurality of fingerprint images through a fingerprint sensor disposed on the touch display, wherein the fingerprint sensor has a plurality of sensing units and the finger of the user moves on the sensing units;
respectively calculating geometric center points corresponding to the fingerprint images;
calculating positions and offsets of the fingerprint images according to the geometric center points corresponding to the fingerprint images;
filling signals in the fingerprint images into a part of pixels in a pixel array according to the positions and the offsets of the fingerprint images;
determining whether a signal quantity of the pixel array is sufficient;
if the signal quantity of the pixel array is sufficient, obtaining signals of other pixels in the pixel array by inputting the signals filled in the part of pixels in the pixel array into an artificial intelligence engine;
generating a candidate fingerprint image according to the pixel array; and
recognizing the user according to the candidate fingerprint image.

### Claim 2 (depends on 1)

The fingerprint recognition method of claim 1, wherein a trajectory of the finger of the user moving on the sensing units is a closed line segment or an unclosed line segment.

### Claim 3 (depends on 1)

The fingerprint recognition method of claim 1, wherein the step of respectively calculating the geometric center points corresponding to the fingerprint images comprises:
recognizing pixel points having the signals in each of the fingerprint images; and
calculating the geometric center points of the fingerprint images according to the pixel points having the signals in each of the fingerprint images.

### Claim 4 (depends on 1)

The fingerprint recognition method of claim 1, wherein the step of respectively calculating the geometric center points corresponding to the fingerprint images comprises:
recognizing pixel points having the signals in each of the fingerprint images; and
calculating the geometric center points of the fingerprint images according to a contour of the pixel points having the signals in each of the fingerprint images.

### Claim 5 (depends on 1)

The fingerprint recognition method of claim 1, further comprising:
storing the candidate fingerprint image and a timestamp corresponding to the candidate fingerprint image in a database.

### Claim 6 (depends on 5)

The fingerprint recognition method of claim 5, further comprising:
detecting another touch signal on the touch display;
sensing a plurality of second fingerprint images through the fingerprint sensor;
respectively calculating geometric center points corresponding to the second fingerprint images;
calculating positions and offsets of the second fingerprint images according to the geometric center points corresponding to the second fingerprint images;
filling signals in the second fingerprint images into a part of pixels in a second pixel array according to the positions and the offsets of the second fingerprint images;
obtaining signals of other pixels in the second pixel array by inputting the signals filled in the part of pixels in the second pixel array into the artificial intelligence engine;
generating a second candidate fingerprint image according to the second pixel array;
comparing whether the second candidate fingerprint image is identical to the candidate fingerprint image stored by the database; and
if the second candidate fingerprint image is not identical to the candidate fingerprint image, stopping to respond to the another touch signal.

### Claim 7 (depends on 6)

The fingerprint recognition method of claim 6, further comprising:
when no other touch signal corresponding to the user is received for a period of time, clearing the recorded candidate fingerprint image from the database.

### Claim 8 (depends on 1)

The fingerprint recognition method of claim 1, further comprising:
if the signal quantity of the pixel array is insufficient, sensing other fingerprint images corresponding to the finger of the user through the fingerprint sensor disposed on the touch display.

### Claim 9 (depends on 8)

The fingerprint recognition method of claim 8, wherein the step of determining whether the signal quantity of the pixel array is sufficient comprises:
calculating and obtaining a fitting curve according to the geometric center points of the fingerprint images, and calculating a length of the fitting curve;
calculating and obtaining an extended fitting curve extended from an end point of the fitting curve to an edge of the fingerprint sensor, and calculating a length of the extended fitting curve;
if the length of the fitting curve is less than the length of the extended fitting curve, determining that the signal quantity of the pixel array is insufficient;
if the length of the fitting curve is not less than the length of the extended fitting curve, determining whether a new pixel point is present by searching backward from the end point of the fitting curve along the fitting curve according to the length of the extended fitting curve;
if the new pixel point is present, determining that the signal quantity of the pixel array is insufficient; and
if the new pixel point is absent, determining that the signal quantity of the pixel array is sufficient.

### Claim 10 (depends on 1)

The fingerprint recognition method of claim 1, wherein the plurality of sensing units disposed on a once-stretchable substrate having a strain detector, and the method further comprises, before the step of detecting the touch signal corresponding to the user on the touch display,
calculating a stretch rate near the strain detector;
estimating a stretch of each place on the once-stretchable substrate;
estimating distances between the plurality of sensing units; and
calculating a position of at least one deformed pixel needed to be complement with a signal in the pixel array,
wherein the other pixels include the at least one deformed pixel.

### Claim 11

An electronic interactive apparatus, comprising:
a processor;
a touch display, coupled to the processor, and configured to detect a touch signal corresponding to a user;
a fingerprint sensor, coupled to the processor, and configured to sense a finger of the user to obtain a plurality of fingerprint images, wherein the fingerprint sensor has a plurality of sensing units and the finger of the user moves on the sensing units; and
a storage device, coupled to the processor,
wherein the fingerprint sensor respectively calculates geometric center points corresponding to the fingerprint images, and calculates positions and offsets of the fingerprint images according to the geometric center points corresponding to the fingerprint images,
wherein the fingerprint sensor is further configured to determine whether a signal quantity of the pixel array is sufficient, if the signal quantity of the pixel array sufficient, the fingerprint sensor fills signals in the fingerprint images into a part of pixels in a pixel array according to the positions and the offsets of the fingerprint images, and obtains signals of other pixels in the pixel array by inputting the signals filled in the part of pixels in the pixel array into an artificial intelligence engine,
wherein the fingerprint sensor generates a candidate fingerprint image according to the pixel array,
wherein the processor recognizes the user according to the candidate fingerprint image.

### Claim 12 (depends on 11)

The electronic interactive apparatus of claim 11, wherein a trajectory of the finger of the user moving on the sensing units is a closed line segment or an unclosed line segment.

### Claim 13 (depends on 11)

The electronic interactive apparatus of claim 11, wherein the fingerprint sensor recognizes pixel points having the signals in each of the fingerprint images, and calculates the geometric center points of the fingerprint images according to the pixel points having the signals in each of the fingerprint images.

### Claim 14 (depends on 11)

The electronic interactive apparatus of claim 11, wherein the fingerprint sensor recognizes pixel points having the signals in each of the fingerprint images, and calculates the geometric center points of the fingerprint images according to a contour of the pixel points having the signals in each of the fingerprint images.

### Claim 15 (depends on 11)

The electronic interactive apparatus of claim 11, wherein the storage device stores a database, wherein the processor stores the candidate fingerprint image and a timestamp corresponding to the candidate fingerprint image in the database.

### Claim 16 (depends on 15)

The electronic interactive apparatus of claim 15, wherein the touch display detects another touch signals,
wherein the fingerprint sensor senses a plurality of second fingerprint images, respectively calculates geometric center points corresponding to the second fingerprint images, and calculates positions and offsets of the second fingerprint images according to the geometric center points corresponding to the second fingerprint images,
wherein the fingerprint sensor fills signals in the second fingerprint images into a part of pixels in a second pixel array according to the positions and the offsets of the second fingerprint images, and obtains signals of other pixels in the second pixel array by inputting the signals filled in the part of pixels in the second pixel array into the artificial intelligence engine,
wherein the fingerprint sensor generates a second candidate fingerprint image according to the second pixel array,
wherein the processor compares whether the second candidate fingerprint image is identical to the candidate fingerprint image stored by the database, and
if the second candidate fingerprint image is not identical to the candidate fingerprint image, the processor stops to respond to the another touch signal.

### Claim 17 (depends on 16)

The electronic interactive apparatus of claim 16, wherein when no other touch signal corresponding to the user is received by the touch display for a period of time, the processor clears the recorded candidate fingerprint image from the database.

### Claim 18 (depends on 11)

The electronic interactive apparatus of claim 11, wherein
if the signal quantity of the pixel array is insufficient, the fingerprint sensor senses other fingerprint images corresponding to the finger of the user.

### Claim 19 (depends on 18)

The electronic interactive apparatus of claim 18, wherein
the fingerprint sensor is further configured to calculate and obtain a fitting curve according to the geometric center points of the fingerprint images, and calculate a length of the fitting curve,
the fingerprint sensor is further configured to calculate and obtain an extended fitting curve extended from an end point of the fitting curve to an edge of the fingerprint sensor, and calculate a length of the extended fitting curve;
if the length of the fitting curve is less than the length of the extended fitting curve, the fingerprint sensor determines that the signal quantity of the pixel array is insufficient,
if the length of the fitting curve is not less than the length of the extended fitting curve, the fingerprint sensor determines whether a new pixel point is present by searching backward from the end point of the fitting curve along the fitting curve according to the length of the extended fitting curve,
if the new pixel point is present, the fingerprint sensor determines that the signal quantity of the pixel array is insufficient, and
if the new pixel point is absent, the fingerprint sensor determines that the signal quantity of the pixel array is sufficient.

### Claim 20 (depends on 11)

The electronic interactive apparatus of claim 11, further comprising:
a strain detector, coupled to the processor, wherein the plurality of sensing units and the strain detector disposed on a once-stretchable substrate,
wherein the strain detector is configured to measure a strain amount of the once-stretchable substrate,
wherein the processor, based on the strain amount of the once-stretchable substrate, calculates a stretch rate near the strain detector, estimates a stretch of each place on the once-stretchable substrate, estimates distances between the plurality of sensing units, and calculates a position of at least one deformed pixel needed to be complemented with a signal in the pixel array, wherein the other pixels include the at least one deformed pixel.