Method of determining, in an image, at least one zone liable to represent at least one finger of an individual
View Patent ↗A method of detecting zones liable to contain at least one finger print in an image of at least one portion of hand of an individual includes a step of acquiring the image, a step of determining a plurality of zones of the acquired image and a first selection step. The step of determining a plurality of zones determines that each zone is a continuous zone of points of the acquired image which each exhibit a luminous intensity level greater than or equal to a selected luminous intensity threshold value. The first selection step selects, from among the zones determined, the zones which are liable to represent at least one print of a finger.
1. A method of detecting zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ) liable to contain at least one finger print in an image of at least one portion of hand of an individual, the method comprising:
a) a step of acquiring the image of at least one illuminated portion of hand with the aid of an optical sensor and without contact with this hand;
b) a step of determining a plurality of zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d ) of the acquired image, each zone thus determined being a continuous zone of points of the acquired image which each exhibit a luminous intensity level (I) greater than or equal to a luminous intensity threshold value (Vsn, Vs 1 ) selected for the determination of the zone thus determined; and
c) a first selection step for selecting, from among the zones determined, the zones which are liable to represent at least one print of a finger, this selection being carried out with the aid of at least one filtering criterion of a first type, chosen from among a finger shape identification criterion and/or a finger print standard texture identification criterion and/or a finger standard color identification criterion and in which a set of threshold values (Vsn, Vs, Vs 2 , Vs 3 ) of luminous intensity (I) lying between a minimum threshold value (Vsmin) of intensity (I) and a maximum threshold value (VsMax) of intensity (I) is selected and the step of determining a plurality of zones of the acquired image is carried out with the aid of each of the threshold values (Vs 1 , Vs 2 , Vs 3 ) of the set of threshold values (Vsn), and then the plurality of the zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ) of the captured image which have been determined with each of the threshold values (Vs 1 , Vs 2 , Vs 3 ) of the set of threshold values (Vsn) is stored.
2. The method of detection according to claim 1 , further comprising a second selection step comprising filtering the previously selected zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ) on completion of the first selection step, with the aid of at least one filtering criterion of a second type, chosen from among:
a criterion of maximum separation between zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ), this maximum separation criterion being determined so as to retain, in a set of retained zones, only zones previously selected on completion of the first selection step, which are mutually neighbouring and which are liable to represent solely portions of prints of fingers of one and the same hand; and/or
a criterion of areal ratio between mutually neighbouring zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ), this areal ratio criterion being determined so that the ratio between areas of mutually neighbouring zones lies between predetermined minimum and maximum values of areal ratio, the mutually neighbouring zones which meet this areal ratio criterion being retained in a set of retained zones; and/or
a criterion of relative orientation between mutually neighbouring zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ), the mutually neighbouring zones which meet this orientation criterion being retained in a set of retained zones.
3. The method of detection according to claim 1 , further comprising, when the acquired image is a color image and, generating a grey level image on the basis of the color image acquired to determine the levels of luminous intensity of the points of the acquired image.
4. The method of detection according to claim 1 , further comprising comparing characteristics of each of the zones (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ) of the set of retained zones with characteristics of images of fingerprints contained in a prerecorded database and, in case of a match between characteristics of one of the zones of the set of retained zones and of one of the fingerprints of the prerecorded database, storing an identification link representative of this match.
5. The method of detection according to claim 1 , wherein, in order to carry out the step of acquiring the image, the illuminated hand is displaced facing a principal optical axis (X 1 ) of the optical sensor.
6. The method of detection according to claim 5 , further comprising, in order to carry out the step of acquiring the image, illuminating the hand with an optical means of artificial lighting exhibiting a principal lighting axis (X 2 ) at least at the moment of the acquisition of the image by the optical sensor.
7. The method of detection according to claim 6 , wherein the optical sensor and the optical lighting means are mechanically interlinked and are designed so that the optical axis (X 1 ) and the principal lighting axis (X 2 ) are mutually parallel.
8. The method of detection according to claim 1 , wherein the acquired image arises from a series of images belonging to a video stream generated with the said optical sensor.
9. The method of detection according to claim 8 , wherein at least one image of the video stream is displayed on a display means and a graphical symbol, indicating a site of the image where the said at least one zone (Z 1 d ) of the acquired image which is liable to represent at least one finger print is situated is superimposed on the displayed image of the video stream.
10. The method of detection according to claim 8 , wherein the video stream comprises a succession of images, each of the images of this succession is used in turn as acquired image, the zones liable to contain at least one finger print are determined in each of the images of the succession of images of the video stream.
11. The method of detection according to claim 9 , wherein the graphical symbol is a frame.
12. The method of detection according to claim 1 , further comprising:
a step of determining at least one zone in the image by using a device, the device comprising a local group of means assembled together and a remote group of means for identifying individuals on the basis of characteristics of fingerprints, said local group of means comprising said optical sensor, at least a memory, at least a processor, telecommunication means adapted to exchange data with said remote group of means, and at least a program executed by the processor to implement said step of determining a plurality of zones of the acquired image and said first selection step for selecting, from among the zones determined, the zones which are liable to represent at least one print of a finger, said local group of means being further arranged to transmit to the remote group of means data containing characteristics representative of the prints represented on the determined zones of the image.
13. A determination device for determining, in an image, at least one zone (Z 1 a , Z 1 b , Z 1 c , Z 1 d , Z 2 a , Z 2 b , Z 2 d , Z 3 a ) of this image which is liable to represent at least one finger print of an individual, this device being adapted to implement the method according to claim 1 .
14. The determination device according to claim 13 , wherein the device comprises an intelligent telephone.