WO2006049396A1

WO2006049396A1 - Method and apparatus for distinguishing forged fingerprint using laser beam

Info

Publication number: WO2006049396A1
Application number: PCT/KR2005/003476
Authority: WO
Inventors: Chongsoo Kim; Soonwon Jung; Sunghyu Shin
Original assignee: Nitgen Co., Ltd.
Priority date: 2004-11-05
Filing date: 2005-10-19
Publication date: 2006-05-11
Also published as: KR20060040325A; KR100607579B1

Abstract

The present invention relates generally to a method and apparatus for distinguishing forged fingerprint (hereinafter, referred to as a replica fingerprint) from living human's real fingerprint (hereinafter, referred to as a bodily fingerprint), by using a laser beam. A method of the present invention comprises steps of: irradiating a laser beam to the object, detecting a surface scattering pattern image (speckle image) created where the laser is irradiated on the object, analyzing a distribution of grey levels of the speckle image, and determining whether or not the object being in contact with the fingerprint input window is a bodily fingerprint. The degree of surface scattering for a human skin is different from that for non-human materials, because the laser scattering characteristic is different between them. Thus, if the grey level differece between the maximum and the minimum over a whole area or a certain area can be compared with a predetermined reference value, it can determine whether the object now being in contact with the fingerprint input window is a bodily fingerprint or a replica fingerprint.

Description

METHOD AND APPARATUS FOR DISTINGUISHING FORGED FINGERPRINT

USING LASER BEAM

Technical Field

The present invention relates generally to a method and apparatus for distinguishing forged fingerprint (hereinafter, referred to as a replica fingerprint) from living human's real fingerprint (hereinafter, referred to as a bodily fingerprint), by using a laser beam.

Background Art

As fingerprint recognition apparatuses are broadly popularized in personal authentication fields such as access and settlement authentication, it is required to tighten security. As various fingerprint recognition algorithms are developed to enhance the accuracy of the fingerprint recognition, fingerprint replica producing techniques are also being developed accordingly.

There are several preceding patents relating to methods of distinguishing forged fingerprints.

First, Japanese Patent Laid-Open Publication No. Hei 11-45338 discloses a method of detecting a bioelectric potential generated in a human body to recognize forged fingerprints. However, in the above method, if only an electrode for potential detection is touched on a human body in a state of bringing a forged fingerprint into contact with a fingerprint input window, the forged fingerprint cannot be distinguished. Further, the above method requires additional hardware and operating units since it additionally needs a detection signal process and a frequency analysis procedure.

Second, Japanese Patent Laid-Open Publication No. Hei 10-290796 discloses a method of applying various stimulations to a human body and measuring types of reactions to the stimulations to distinguish forged fingerprints. However, the second method of analyzing biological reactions to external stimulations is too artificial, thus giving an unpleasant feeling to a user. Further, this method is problematic in that it is difficult to formally quantify the types of reactions to stimulations.

Third, Japanese Patent Laid-Open Publication No. 9-259272 discloses a method of detecting the existence of sweat glands and the number of them in a fingerprint image to distinguish forged fingerprints. This third method is impractical because even sweat glands can be copied as well as the patterns of a fingerprint according to current forged fingerprint producing techniques.

Fourth, Japanese Patent Laid-Open Publication No. Hei 7-308308 discloses a method of emitting specific wavelengths of light and detecting variations of oxygen density and blood flow using the amount of transmitted light to distinguish forged fingerprints. However, this method is limited in that if a forged fingerprint is produced using a material enabling the frequency of light emitted from the light emitting device to be easily transmitted therethrough and is put on a finger to be used, the forged fingerprint cannot be distinguished. Fifth, Japanese Patent Laid-Open Publication No. Hei 12-20684 discloses a device that uses two light sources for living body discernment, which irradiates probe light and a reference light in a fingerprint image input device at the finger contacted by the control surface. The probe light has the wavelength such that the absorbency index for oxidized hemoglobin is less than that for reduced hemoglobin in a human body. The reference light has the wavelength such that the absorbency index for oxidized hemoglobin is equal to that for reduced hemoglobin. Therefore, the extent of oxidized hemoglobin's absorbing the probe light is less than the extent of reduced hemoglobin's absorbing the probe light. On the other hand, the extent of oxidized hemoglobin's absorbing the reference light is equal to the extent of reduced hemoglobin's absorbing the reference light. Since a bodily finger naturally contains oxidized hemoglobin and reduced hemoglobin, the intensity of the probe light and the reference light after penetrating an object being in contact with a fingerprint input window is different from the intensity of those before entering the object, respectively in a bodily object and in a non-bodily object. Comparing the intensities of the detected leaving light and the incident light to the lookup table, it is possible to discern whether the object being in contact with a fingerprint input window is a bodily fingerprint or a replica fingerprint. In this prior art, however, it is important to select wavelengths of the probe light and the reference light. Thus, it is not easy to construct a hardware system, and also a software algorithm becomes quite complicated.

Sixth, there has been also developed a technology to distinguish forged fingerprints by a method, which recognizes pulsation of a fingertip using a pressure sensor. However, this fifth method is limited in that it cannot cope with other information similar to the pulsation.

In addition to the above examples, there are recently developed technologies of thinly producing forged fingerprints using materials such as silicone, film, etc, Accordingly, if such forged fingerprints are directly put on human bodies, it is very difficult to distinguish forged fingerprints from authentic fingerprints even though any of the above methods is employed.

Disclosure of Invention Technical Problem

The present invention has been developed to improve the conventional replica fingerprint distinguishing methods, by adopting simplified hardware and software constructions. This invention has been conceived, based on the fact that a laser beam creates a unique speckle marking when it is irradiated to a human body and is scattered thereon, wherein the speckle marking varies according to the materials of the object.

Therefore it is an object of the present invention to provide a method and apparatus for distinguishing a replica fingerprint from a bodily fingerprint, by adopting a laser source to utilize the above laser characteristic.

Technical Solution

First, a surface scattering characteristic of a laser beam will be briefly discussed. When a laser beam is irradiated on the surface of an object, it is scattered thereon while it simultaneously penetrates inwards. If the amount of penetration is comparatively less, the amount of surface scattering becomes more.This surface scattering creates a unique scattering marking which is formed by myriad of minute particles. This marking is called as "speckle," which is unique to a laser beam.

However, since a laser beam more easily penetrates in a bodily skin texture and is well absorbed therein, the amount of surface scattering is lessened, and thus the speckle marking is created obscurely. The reason why a laser beam is well abosorbed in the bodily texture is that the bodily texture contains fills such as melanin, hemoglobin, etc. Especially, a laser beam of 600-750nm is maximally abosorbed in the bodily texture.

To sum up, the surface scattering marking of a laser beam is obscure for the surface of bodily texture, while it is created definitely on the materials used for replica fingerprints, such as rubber, silicone, gelatin, film, etc. This is exemplified in Fig. 1.

Referring to Fig. 1, it can be noted that each of a human fingerprint, rubber, silicone, and film creates different speckle markings. Distribution graph of grey levels at right side of each item details the differences. The surface scattering markings, speckles, are unique to the relevant materials. Therefore, by analyzing the speckle marking obtained from irradiating a laser beam to an object to be fingerprint-authenticated, the object can be determined as a bodily fingerprint or a replica fingerprint. That is, since the speckle marking is obscure for a bodily fingerprint as in Fig. 1, by irradiating a laser beam before being undergone the fingerprint authentication process and analyzing a specke marking, system can judge whether the object now being in contact with a fingerprint input window is a bodily fingerprint or a replica fingerprint.

Analyzing the speckle marking can be performed by a variety of known image processing techniques. For example, distribution of gray levels calculated from an image can be utilized for analysis of the speckle. As described below, the difference between maximum value and minimum value of grey levels within a certain area (e.g., within a vertical line at a given position) can be calculated. The difference between the maximum and minimum values for a bodily fingerprint will be less than that for a replica fingerprint.

Advantageous Effects The present invention can distinguish a bodily fingerprint from replicas only by adding a simple hardware and software construction. In addition, the present invention is highly reliable in terms of security. If it is installed in a concealed place so as not to be seen from outside, a user is prevented from knowing of the existence of the forged fingerprint distinguishing apparatus and whether it works. Description of Drawings

The above and other objects, features, and advantages of the present invention will become more apparent from the following description when taken in conjunction with the accompanying drawings, in which:

Fig. 1 shows various surface scattering markings created when a laser beam is irradiated to a variety of objects, and

Fig. 2 is a construction diagram of a replica fingerprint distinguishing apparatus according to the present invention.

Best Mode for Invention

Hereafter, a best mode for the present invention will be described in detail with reference to the attached drawings. A method of distinguishing a replica fingerprint will be first explained. First, a laser beam is irradiated to an object to be fingerprint-authenticated, which is in contact with a fingerprint input window of a prism, also including a fingerprint exit face from which a fingerprint image leaves therefrom. Then, a surface scattering marking image (speckle image) created on the object where the laser beam is irradiated is detected. The detected speckle image's grey level distribution is analyzed to calculate the difference between the maximum value and the minimum value in the grey level distribution to determine whether or not the object being in contact with the fingerprint input window is a bodily fingerprint. As stated before, the degree of surface scattering for a human skin is different from that for non-human materials, because the laser scattering characteristic is different between them. Specifically, comparing the maximum grey level with the minimum grey level over a certain area (such as over a certain linear interval of the speckle image area), the difference between them for a bodily fingerprint is less than that for a replica fingerprint (see Fig. 1). Thus, the grey level differece between the maximum and the minimum over a whole area or a certain area can be compared with a predetermined reference value in order to determine whether the object now being in contact with the fingerprint input window is a bodily fingerprint or a replica fingerprint. Here, the predetermined reference value may be set as an average of the grey levels of a speckle marking gotten from a human skin.

As stated before, if the wavelength of a laser beam is 600-750nm, the difference of grey levels between a bodily fingerprint and a replica fingerprint becomes maximum. Thus far, a grey level distribution method has been introduced, however, such a method using grey levels is merely one embodiment for the present invention. To distinguish a bodily fingerprint from a replica fingerprint by analyzing the speckle image, a variety of known image processing techniques can be utilized by those who ordinarily skilled in the art. In the meantime, as in Fig. 3, the fingerprint acquisition apparatus of the present invention is composed of a prism 10 including a fingerprint input window 12 and a fingerprint exit face 14 from which a fingerprint image leaves therefrom; a laser source 30 for irradiating a laser beam 32 to an object being in contact with the fingerprint input window 12; a focusing lens 50 for focusing a speckle marking leaving the fingerprint exit face 14 when the laser source 30 irradiates the laser beam to the object; an image sensor 60 for imaging the speckle marking focused from the focusing lens 50; and an image processor 70 for processing the speckle image signal and for determining that the object being in contact with the fingerprint input window 12 is a bodily fingerprint or a replica fingerprint, based upon the grey level of the speckle image. The image processor 70 can distinguish a bodily fingerprint from a replica fingerprint by analyzing the difference between the maximum value and the minimum value of the grey level distribution of the speckled image. The analysis of the grey level distribution may be perfomed over the whole area of the speckle image acquired, but also may be performed in only a certain intervals for the saving of time or memory capacity.

Therefore, the image processor 70 is comprised of a means for calculating an area value of the scattered pattern image acquired by irradiating a laser beam to the object, and a means for determining a replica fingerprint if the calculated area value is less than the reference value. Further, if the image processor 70 determines that the object being in contact with the fingerprint input window is a replica fingerprint, it can stop further fingerprint authentication procedures; while if it determines the bodily fingerprint, further procedures, i.e., extracting fingerprint minutiae and authenticating a fingerprint, can go on.

In Fig. 3, an element 40 shows a laser controller for controlling turning on and off the laser source 30, and elements 20 and 22 respectively shows a light source for acquiring a fingerprint image and a light beam irradiated from this, which are indispensable to the optical fingerprint acquisition apparatuses.

Like the above, by simply irradiating a laser beam to an object to be fingerprint- authenticated to obtain a scattered image before performing a fingerprint image acquiring step, a replica fingerprint can be easily but perfectly distinguished.

While the invention has been shown and described with reference to certain embodiments to carry out this invention, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method of distinguishing forged fingerprint (referred to as a replica fingerprint) from living human's real fingerprint (referred to as a bodily fingerprint), by acquiring a fingerprint image from an object being in contact with a fingerprint input window of a fingerprint acquisition apparatus, the method comprising steps of: irradiating a laser beam to the object, detecting a surface scattering pattern image (speckle image) created where the laser is irradiated on the object, analyzing a distribution of grey levels of the speckle image, and determining whether or not the object being in contact with the fingerprint input window is a bodily fingerprint.

2. The method of claim 1, wherein the step of determining a bodily fingerprint includes a step of comparing the difference between the maximum level and the minimum level of the grey levels of the speckle image, over a predetermined area, and determining whether or not the object being in contact with the fingerprint input window is a bodily fingerprint, by utilizing the difference.

3. The method of claim 1, wherein a wavelength of the laser beam is approximately 600-750nm.

4. An apparatus for distinguishing forged fingerprint (referred to as a replica fingerprint) from living human's real fingerprint (referred to as a bodily fingerprint), by acquiring a fingerprint image from an object being in contact with a fingerprint input window of a fingerprint acquisition apparatus, the apparatus comprising: a prism including a fingerprint input window and a fingerprint exit face from which a fingerprint image leaves therefrom; a laser source for irradiating a laser beam to an object being in contact with the fingerprint input window ; an image sensor for imaging a speckle image leaving the fingerprint exit face when the laser source irradiates the laser beam to the object; and an image processor for analyzing a distribution of grey levels of the speckle image signal and for determining, based upon the grey level, that the object being in contact with the fingerprint input window is a bodily fingerprint or a replica fingerprint,

5. The apparatus of claim 4, wherein the image processor comprises means for comparing the difference between the maximum level and the minimum level of the grey levels of the speckle image, over a predetermined area, whereby the image processor determines whether or not the object being in contact with the fingerprint input window is a bodily fingerprint, by utilizing the difference.

6. The apparatus of claim 4, wherein a wavelength of the laser beam is approximately 600-750nm.