Live facial recognition method and system

1. A live facial recognition method, comprising:

capturing a visual image and obtaining temperature information of a person under recognition, facial features being derived according to the visual image;

comparing the facial features of the visual image with corresponding facial features of a facial feature database to obtain a difference therebetween; and

determining whether facial temperature conforms to a live facial recognition criterion according to the temperature information, if the difference is less than a predetermined threshold value;

wherein the live facial recognition criterion comprises:

performing mapping between the visual image and the temperature information;

normalizing the visual image to catch a recognition area in the visual image;

catching a corresponding recognition area in the temperature information according to mapping relationship between the visual image and the temperature information;

dividing the recognition area of the temperature information into a plurality of sub-areas;

selecting a feature area in the visual image and a corresponding feature area in the temperature information, the feature area of the temperature information including at least one said sub-area;

comparing temperature of the feature area of the temperature information with temperature of at least one neighboring sub-area; and

asserting the live facial recognition criterion if comparing result conforms to predetermined facial temperature distribution.

2. The method of claim 1 , wherein the predetermined facial temperature distribution comprises one or more of the following requirements:

facial temperature is higher than ambient temperature;

temperature above eyes is higher than temperature below the eyes;

above the eyes, eyebrow temperature is lower than temperature of other area;

below the eyes, nose temperature is lower than temperature of other area;

the facial temperature is higher than 28° C.;

eye temperature is the highest in a face;

nose temperature is the lowest in the face;

temperature above nose is higher than temperature below the nose; and

temperature of the eyes with glasses is lower than other temperature in the face, but higher than the ambient temperature.

3. The method of claim 1 , wherein the temperature information comprises a temperature image, the visual image and the temperature image are captured by an image capture device and a temperature detection device respectively, and the image capture device is separate from the temperature detection device.

4. The method of claim 1 , wherein the temperature information comprises a temperature image, the visual image and the temperature image are respectively captured by an image capture device and a temperature detection device at a same time, and the image capture device and the temperature detection device are integrated.

5. The method of claim 1 , wherein the temperature information comprises plural temperature data, the visual image and the plural temperature data are captured by an image capture device and a temperature detection device respectively, and the temperature detection device includes a single temperature sensor that is separate from the image capture device.

6. The method of claim 5 , wherein the temperature sensor obtains the plural temperature data at specific points of a face according to facial positions detected in the visual image.

7. The method of claim 1 , wherein the facial feature database is generated by the following steps:

capturing an image of a person under test, resulting in a captured image;

performing face detection on the captured image;

catching a face image from the captured image according to results of the face detection;

extracting facial features from the face image;

numericalizing the facial features to generate facial feature values; and

building a model according to the facial feature values, thereby generating the facial feature database.

8. A live facial recognition method, comprising:

capturing a visual image and obtaining temperature information of a person under recognition, facial features being derived according to the visual image;

comparing the facial features of the visual image with corresponding facial features of a facial feature database to obtain a difference therebetween; and

determining whether facial temperature conforms to a live facial recognition criterion according to the temperature information, if the difference is less than a predetermined threshold value;

wherein the live facial recognition criterion comprises:

normalizing a temperature information to catch a recognition area in the temperature information;

dividing the recognition area into a plurality of sub-areas;

selecting a feature area in the temperature information, the feature area including at least one said sub-area;

comparing temperature of the feature area of the temperature information with temperature of at least one neighboring sub-area; and

asserting the live facial recognition criterion if comparing result conforms to predetermined facial temperature distribution.

9. The method of claim 8 , wherein the predetermined facial temperature distribution comprises one or more of the following requirements:

facial temperature is higher than ambient temperature;

temperature above eyes is higher than temperature below the eyes;

above the eyes, eyebrow temperature is lower than temperature of other area;

below the eyes, nose temperature is lower than temperature of other area;

the facial temperature is higher than 28° C.;

eye temperature is the highest in a face;

nose temperature is the lowest in the face;

temperature above nose is higher than temperature below the nose; and

temperature of the eyes with glasses is lower than other temperature in the face, but higher than the ambient temperature.

10. A live facial recognition system, comprising:

a processor;

an image capture device that is controlled by the processor to capture a visual image of a person under recognition, facial features being derived according to the visual image;

a temperature detection device that is controlled by the processor to obtain temperature information of the person under recognition; and

a storage device that stores a facial feature database;

wherein the processor compares the facial features of the visual image with corresponding facial features of the facial feature database to obtain a difference therebetween; and the processor determines whether facial temperature conforms to a live facial recognition criterion according to the temperature information, if the difference is less than a predetermined threshold value;

wherein the live facial recognition criterion comprises:

performing mapping between the visual image and the temperature information;

normalizing the visual image to catch a recognition area in the visual image;

catching a corresponding recognition area in the temperature information according to mapping relationship between the visual image and the temperature information;

dividing the recognition area of the temperature information into a plurality of sub-areas;

selecting a feature area in the visual image and a corresponding feature area in the temperature information, the feature area of the temperature information including at least one said sub-area;

comparing temperature of the feature area of the temperature information with temperature of at least one neighboring sub-area; and

asserting the live facial recognition criterion if comparing result conforms to predetermined facial temperature distribution.

11. The system of claim 10 , further comprising a memory unit that stores a computer program, by which the processor performs live facial recognition.

12. The system of claim 10 , wherein the predetermined facial temperature distribution comprises one or more of the following requirements:

facial temperature is higher than ambient temperature;

temperature above eyes is higher than temperature below the eyes;

above the eyes, eyebrow temperature is lower than temperature of other area;

below the eyes, nose temperature is lower than temperature of other area;

the facial temperature is higher than 28° C.;

eye temperature is the highest in a face;

nose temperature is the lowest in the face;

temperature above nose is higher than temperature below the nose; and

temperature of the eyes with glasses is lower than other temperature in the face, but higher than the ambient temperature.

13. The system of claim 10 , wherein the image capture device is separate from the temperature detection device, and the temperature detection device obtains a temperature image representing the temperature information.

14. The system of claim 10 , wherein the image capture device and the temperature detection device are integrated, and respectively capture the visual image and a temperature image at a same time, the temperature image representing the temperature information.

15. The system of claim 10 , wherein the image capture device is separate from the temperature detection device, and the temperature detection device includes a single temperature sensor.

16. The system of claim 15 , wherein the temperature sensor obtains plural temperature data at specific points of a face according to facial positions detected in the visual image, the plural temperature data representing the temperature information.

17. A live facial recognition system, comprising:

a processor;

an image capture device that is controlled by the processor to capture a visual image of a person under recognition, facial features being derived according to the visual image;

a temperature detection device that is controlled by the processor to obtain temperature information of the person under recognition; and

a storage device that stores a facial feature database;

wherein the processor compares the facial features of the visual image with corresponding facial features of the facial feature database to obtain a difference therebetween; and the processor determines whether facial temperature conforms to a live facial recognition criterion according to the temperature information, if the difference is less than a predetermined threshold value;

wherein the live facial recognition criterion comprises:

normalizing a temperature information to catch a recognition area in the temperature information;

dividing the recognition area into a plurality of sub-areas;

selecting a feature area in the temperature information, the feature area including at least one said sub-area;

comparing temperature of the feature area of the temperature information with temperature of at least one neighboring sub-area; and

asserting the live facial recognition criterion if comparing result conforms to predetermined facial temperature distribution.

18. The system of claim 17 , wherein the predetermined facial temperature distribution comprises one or more of the following requirements:

facial temperature is higher than ambient temperature;

temperature above eyes is higher than temperature below the eyes;

above the eyes, eyebrow temperature is lower than temperature of other area;

below the eyes, nose temperature is lower than temperature of other area;

the facial temperature is higher than 28° C.;

eye temperature is the highest in a face;

nose temperature is the lowest in the face;

temperature above nose is higher than temperature below the nose; and

temperature of the eyes with glasses is lower than other temperature in the face, but higher than the ambient temperature.