WO2007031607A1

WO2007031607A1 - Method of identifying a person

Info

Publication number: WO2007031607A1
Application number: PCT/FI2006/050394
Authority: WO
Inventors: Kari Kaila
Original assignee: Kari Kaila
Priority date: 2005-09-16
Filing date: 2006-09-15
Publication date: 2007-03-22
Also published as: FI20055494L; FI20055494A0

Abstract

The invention relates to a method of biometric identification of a person. The method comprises measuring a plural number of subsequent cross-sectional images of a person's finger (1) and using them for mathematically computing a three-dimensional model of the blood vessels (3) in the finger, the model being then used in the identification.

Description

METHOD OF IDENTIFYING A PERSON

BACKGROUND OF THE INVENTION

[0001] The invention relates to a method of biometrical identification of a person.

[0002] It is increasingly common to aim at identifying persons for reasons of security before allowing them to use a device or to enter an area, such as a country. At the moment biometric identification methods are being introduced into use. In these methods the aim is to implement identification based on a person's physical characteristic that is considered substantially unchanging and completely unique. Such identification methods include fingerprint recognition, iris recognition, different voice recognition methods, skull shape recognition, etc.

[0003] In fingerprint recognition a person places his/her finger on a read head, which reads the fingerprint pattern thus obtained and compares it with a pre-stored fingerprint pattern. The solution as such is fairly reliable, because there has been no evidence indicating that different persons could have fingerprints that are so similar that they would not be distinguishable from one another. The biggest problem with this identification method is that it has been shown that fingerprint recognition devices can be successfully deceived, and so far a reliable solution to overcome this problem has not been presented.

[0004] Iris recognition, in turn, is based on placing the eye to a measuring device that reads the features of the iris and compares them with the data on the iris stored in a memory. Compared with the fingerprint method, the deceiving of this method is more difficult, but nevertheless fully possible with similar means. Therefore this identification method is not fully reliable either.

[0005] By combining fingerprint identification and iris identification it is possible to provide a considerably more reliable identification method. However, although a significantly greater effort is required to deceive this method, it is still possible. Moreover, the system in question is more laborious to use and more time-consuming. It is therefore not applicable to mass situations, for example for identifying passengers at an airport.

[0006] Different voice recognition methods identify the speaker's voice and analyse certain features from it. These are then compared with a previously recorded voice pattern. This method is significantly easier to deceive than the two first ones.

[0007] One relatively recent recognition method involves measuring an image of a person's skull in a specific way and comparing it with an image made of the skull of the same person. This method is fairly complicated to implement efficiently and not very easy to apply to practice in normal air traffic conditions, for example.

SUMMARY OF THE INVENTION

[0008] It is an object of the invention to provide a biometric measuring method, which allows the measurement to be taken reliably and quickly and which is extremely difficult to deceive.

[0009] The method of the invention is characterized by imaging a cross-section of a person's protrusion, such as a finger, at multiple locations by means of a measurement distinguishing different parts of soft tissue in a human body; by mathematically computing the measured cross-sectional images to produce a three-dimensional model of the blood vessels in the person's protrusion in question; and by using this computed model, compared with another model previously obtained from the same protrusion by corresponding imaging and computing, as a biometric identifier. An embodiment of the invention is characterized by computing on the basis of the measurements a medial line of the blood vessels, thereby allowing the impact of contraction and expansion due to temperature acting on the blood vessels to be practically eliminated.

[0010] An essential idea of the invention is to measure at sufficiently frequent intervals cross-sections of a fingertip, or the end of some other suitably protruding organ of a person, by applying a measurement, such as magnetic resonance imaging, which distinguishes different parts of the soft tissue in a human body from one another, the cross-sectional images being then used for mathematically determining a three-dimensional model of the blood vessels in the protrusion. This three-dimensional model of the blood vessels obtained by the measurements is then compared with a pre-stored model to allow the person in question to be identified. An advantage of the invention is that it is easy and rapid to carry out, because placing the finger onto the measuring zone is all that is required. Further, this method is practically impossible to deceive, because it is simply not possible to build a three-dimensional model of blood vessels in the same way as false fingerprints or iris patterns can be produced. Identification may easily be obtained from more than one finger, or toes, for example, of one and the same person, if considered necessary when investigating into the causes of an accident, or the like. Further, suitable measuring devices based on magnetic resonance are in practice already commercially available, which means that the implementation of the invention mainly requires further development of the related computer programming and modelling. A further advantage of the invention is that the measurements may be carried out on the thicker blood vessels only, disregarding those known as capillaries, thereby enabling to obtain a measurement that is sufficiently accurate but requires less space and computing capacity.

BRIEF DISCLOSURE OF THE FIGURES

[0011] In the following the invention will be described in greater detail with reference to the accompanying drawings, which schematically depict a three-dimensional model of the blood vessels in a human fingertip.

Figure 1 is a schematic side view of blood vessels in a human finger; and

Figure 2 is a schematic top view of blood vessels in a human finger.

DETAILED DESCRIPTION OF THE FIGURES

[0012] Figure 1 is a schematic side view of blood vessels in a human finger, Figure 2 being a top view of blood vessels in the finger. The finger is schematically depicted with a dotted line 1 , the nail on the finger being indicated by reference number 2. Further, the figures provide a side view of a vascular system 3, which is different for each individual. Reference numeral 3a indicates, by way of example, a section of a measured blood vessel.

[0013] In a measurement based on magnetic resonance imaging the finger is pushed into a measuring device that measures cross-sectional images of the fingers at predetermined intervals in a manner known per se. The measurement is carried out using a strong magnetic field and radio frequency electro-magnetic radiation. The combined effect of these two produces the cross-sectional image of the fingers. In the measurement the organ being measured, such as a finger, is subjected to a fixed magnetic field, in addition to which pulses of radio frequency electro-magnetic radiation are transmitted to the target. The magnetic field sets the atomic kernels of the target being imaged into a specific pattern based on the magnetic field. The radio frequency electro-magnetic radiation, in turn, can be used to offset the kernels from the state caused by the magnetic field, thus allowing the energy generated by their return to be measured and, as a result, an image to be created. The measurement as such, both in relation to humans and animals, is fully known per se and therefore does not need to be disclosed in any more detail in this context.

[0014] After cross-sectional images have been taken from the finger at sufficiently frequent intervals and on a sufficiently long distance, up to the first articulation in the finger, for example, the measurement is stopped and by means of a modelling program built into a computer a mathematical model of the blood vessels in the person's fingertips is created. The human vascular system is totally unique for each individual and also for each finger of every individual. Consequently, by comparing a mathematical model obtained by the measurement to a model previously measured from the same person, it is easy to confirm that the person in both cases is the same. The three- dimensional models of different persons differ from one another to such a significant extent that there is no possibility of confusion and false identification. If for some exceptional reason an extremely high certitude of identification is desired, a model of the blood vessels in one of the person's other fingers may be measured for identification and compared with a previously stored model of that finger. After this, there is simply no possibility for an identification error.

[0015] As regards the possibility of a fraud in identification, it should be noted that there is no technical way to build a model of a blood vessel of a finger. Therefore it is only the person in question who qualifies for the identification procedure. Further, in situations where this kind of identification is required, there is control staff present and therefore it is not possible to force a person to provide the identification. In practice this prevents the use of violence in the identification procedure.

[0016] The invention is disclosed in the above specification and the drawings by way of example only, the invention not being restricted thereto in any way. An essential aspect is that a person's finger or toe is measured at sufficiently frequent intervals by means of a measurement, such as magnetic resonance imaging (MRI), which distinguishes different parts of soft tissue in the human body from one another and the measured cross-sectional patterns are then used to produce a three-dimensional model of the blood vessels in the finger or toe to be used for identifying the person.

Claims

1. A method for biometric identification of a person carried out by comparing a biometrically measured feature of the person with previously measured data on the same feature, characterized by imaging a cross-section of a person's protrusion, such as a finger (1), at multiple locations by means of a measurement distinguishing different parts of soft tissue in a human body; by mathematically computing the measured cross- sectional images to produce a three-dimensional model of the blood vessels (3) in the person's protrusion in question; and by using this computed model, compared with another model previously obtained from the same protrusion by corresponding imaging and computing, as a biometric identifier..

2. A method according to claim 1, characterized by computing a mathematical median line of the blood vessels (3a) in the vascular system (3) and using this three-dimensional model formed by the computed median lines as the model.

3. A method according to claim 1 or 2, characterized in that the imaging is carried out at predetermined distances in the longitudinal direction of the protrusion.

4. A method according to any one of the previous claims, characterized in that the imaging is carried out by means of a measurement based on magnetic resonance.