KR100920251B1 - A method for restoring infrared vein image blurred by skin scattering - Google Patents

Abstract

PURPOSE: A method for restoring infrared vein image blurred by skin scattering is provided to improve the visibility improvement on the medical field. CONSTITUTION: The method for restoring infrared vein image comprises as follows. The image of the blood vessel having relatively dark to the skin is acquired using the near infrared ray irradiation apparatus and infrared camera. The virtual blood vessel image is produced based on the blood vessel thickness of the general man. The unknown parameter of the point distribution function is estimated by the line profile matching between the second vein image and the first vein image. The point distribution function is produced based on the presumed unknown parameter. The first vein image of which boundary is vague is restored by convolution of the point distribution function of the Gaussian.

Description

A method for restoring infrared vein image blurred by skin scattering}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vein recognition method that is in the spotlight in the field of bio-metrics, and more particularly, to an infrared vein image restoration method for solving a phenomenon in which a blood vessel image acquired by infrared rays is blurred by skin scattering.

The blood inside the vein has a characteristic of absorbing light having a near infrared wavelength due to unoxidized hemoglobin components. By using this point, if an image is acquired by using a near infrared ray irradiation apparatus and an infrared camera, an image of a blood vessel having a darker brightness than the skin can be obtained. The vein image obtained in this way has the advantage that the user's rejection and inconvenience and the recognition time is fast in the medical field as well as in the bio-recognition field for personal authentication, and the use rate thereof is rapidly increasing.

In the related art, after detecting the vein position of a finger, a method of extracting a feature based on the extracted vein and using the same may be obtained from the following data.

[1] T. Yanagawa, S. Aoki, and T. Ohyama, “Human Finger Vein Images Are Diverse And Its Patterns Are Useful for Personal Identification”, MHF 2007-12, pp. 1-7, 2007.

[2] N. Miura, A. Nagasaka, and T. Miyatake, "Feature Extraction of Finger-Vein Patterns Based on Repeated Line Tracking and Its Application to Personal Identification", Machine Vision and Applications, vol. 15, pp. 194-203, 2004.

[3] Z. Zhang, S. Ma, and X. Han, "Multiscale Feature Extraction of Finger-Vein Patterns Based on Curvelets and Local Interconnection Structure Neural Network", The 18th International Conference on Pattern Recognition (ICPR'06), pp. . 145-148, 2006.

[4] N. Miura, A. Nagasaka, and T. Miyatake, "Extraction of Finger-Vein Patterns Using Maximum Curvature Points in Image Profiles", IEICE TRANS. INF. & SYST., Vol. E90-D, no. 8, pp. 1185-1194, 2007.

In other words, the concept that the human finger can be identified by the finger vein has been proved by Yanagawa's research [1]. According to Yanagawa's research, each person's vein pattern of ten fingers is different, and one finger vein pattern is suitable for personal authentication with similar degrees of freedom as the highly reliable iris pattern [1].

In addition, Miura proposed a method to extract vein patterns with unbounded borders by vein line tracing starting from various locations on the finger image, and the experimental results showed an average recognition error of 0.145%. ms) [2].

Zhang proposed a finger vein pattern extraction method based on curvelet and local interconnection structure Neural Networks, and the average error rate was 0.128% based on Miura's matching method [3].

Meanwhile, a recent study by Miura suggested a method for efficiently extracting vein patterns of various thicknesses, based on the fact that the thickness of the pattern may vary due to changes in vein blood flow depending on weather or human condition. Recognition error of 0.0009% showed excellent performance [4].

The conventional camera-based finger vein recognition method has a problem that the vein area may not be extracted accurately by the thick finger skin. That is, in the case of the vein image acquired using the near infrared irradiation device and the infrared camera, there is a problem in that the image acquired by scattering the infrared rays irradiated by the finger skin layer positioned between the camera lens and the vein is blurred.

However, current studies are only studying the method of accurately distinguishing the skin area and the vein area from the acquired vein image, and the vein image is solved by solving the phenomenon that the vein image irradiated by infrared rays is blurred by skin scattering. There is no ongoing research to improve the quality of the products.

However, the functions and methods are different, but the existing data for restoring the image blurred by the scattering layer includes the following data.

[5] Publication (10-2008-0069591: Scatter correction)

[6] Koichi Shimizu, Koji Tochio and Yuji Kato, "Transcutaneous Fluorescent Imaging with a Depth-Dependent Point Spread Function", Proceeding of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, pp. 3141-3144

[5] is a method of reconstructing a blurred image by acquiring a point diffusion function for scattering flow corresponding to each part when acquiring a 3D pixel image of a radiation-based tooth structure. The problem is that information about the point spread function must also be obtained. Therefore, this method cannot be used to reconstruct an already acquired vein image. In addition, since this method uses a 3D pixel image acquisition technique, there is a problem that it is not appropriate to be applied to a method of extracting a vein pattern using a vein image using a 2D pixel image acquisition technique.

[6] is a method of restoring a blood vessel image obtained after injecting a fluorescent substance into a mouse body, and designing a point diffusion function by applying a fixed thickness of the skin layer to improve the quality of the image. It is a way. In other words, the blood vessel image obtained by this method is an image obtained from the head of the rat. The thickness of the skin layer is so thin that a point diffusion function can be calculated by applying a skin layer having a fixed thickness. However, in the case of a human finger, since the thickness of the skin layer is thick and the thickness of each person is different from each other, designing a point diffusion function using the thickness of the fixed skin layer causes a large error. In addition, this method is a method of acquiring blood and images by injecting a fluorescent material into the body, which is different from a general vein image acquisition device using infrared illumination and a camera, and applies a fixed thickness without estimating the thickness of the skin layer that causes scattering. In the case of applying this to a human finger, there is a problem in that the quality of the acquired image is deteriorated.

As described above, methods for overcoming a phenomenon in which a subject is blurred by scattering layers such as teeth and skin in radiation-based three-dimensional images [5] or medical fluorescent-based two-dimensional camera images [6] have been studied. Since both point diffusion functions must be acquired at the time of image acquisition, they cannot be applied to images already acquired, or the absorption, scattering coefficients, and thickness of the scattering layer required for the point diffusion function design are simply estimated by fixed empirical methods. Due to this, the accuracy of the point spread function estimation is inferior.

Therefore, the present invention has been made to solve the above problems, and in order to improve the recognition accuracy in the vein recognition method, which has recently been spotlighted in the field of bio-recognition, the blurred blood vessel image scattered by the skin layer is identified by skin optics. The purpose of the present invention is to provide an infrared vein image reconstruction method for designing a point diffusion function of a blood vessel image based on possible skin layer thickness, scattering, and absorption coefficient, and applying the same to a blood vessel image to improve vein image quality.

Another object of the present invention is to restore the blurred blurry image of the vein due to the scattering of light by the skin layer when acquiring a blood vessel image irradiated by infrared rays to increase the visibility to be applied to various medical fields through the analysis of veins To provide an infrared vein image restoration method that can be.

Another object of the present invention is to calculate the thickness of the skin layer, absorption coefficient, scattering coefficient based on the skin optics in consideration of the thickness of the finger, and to design a point diffusion function using this to improve the quality of the already acquired vein image To provide an infrared vein image restoration method that can be improved.

Another object of the present invention is to restore the image flow on the basis of the point diffusion, so that the boundary between the vein and the skin becomes clear, the bio-recognition method using the vein processed by an automated algorithm (palm veins, finger veins, facial veins, etc.) The present invention provides a method for restoring infrared vein images that can increase recognition rate.

In order to achieve the above object, the infrared vein image restoration method blurred by the scattering of the skin according to the present invention is characterized by (a) an image of a blood vessel having a relatively darker brightness than the skin using a near infrared irradiation device and an infrared camera. Acquiring a first vein image; acquiring a second vein image; (b) acquiring a second vein image based on a blood vessel thickness of a general person; and (c) the first vein image and the second vein image. Estimating an unknown variable of a point spread function through line profile matching between the steps; (d) calculating a point spread function based on the estimated unknown variable; and (e) using a Gaussian basis using the calculated point spread function. And restoring the first vein image having an ambiguous vein boundary by restoring filtering using a convolution of a point diffusion function.

(조명의 세기) 및 산란계층의 두께(d) 중 적어도 하나 이상인 것을 특징으로 한다.Preferably the unknown variable in step (c)

(Intensity of illumination) and the thickness d of the scattering layer.

Preferably step (d)

는 피부층의 흡수계수를,

는 피부층의 산란계수를,

는

의 제곱근을,

는 흐려진 면적에 대한 레디얼(radial) 거리를, d 는 산란계층의 두께를 나타내는 것을 특징으로 한다.To calculate the point spread function using the formula of

Is the absorption coefficient of the skin layer,

The scattering coefficient of the skin layer,

Is

Square root of

Is the radial distance to the blurred area, and d is the thickness of the scattering layer.

Preferably, the step (e) is to restore the first vein image using the resultant damage image calculated through the calculated point diffusion function and the second vein convolution.

As described above, the method for restoring a blurred infrared vein image by scattering of the skin according to the present invention is based on skin layer thickness, scattering and absorption coefficients, which can be identified by skin optics. The point diffusion function of is designed and applied to the vascular image to improve the quality of the vein image. In addition, the improvement of the vascular image quality can be expected to improve the visibility in the medical field using the vein pattern, the recognition rate in the bio-recognition field using the vein image.

Other objects, features and advantages of the present invention will become apparent from the following detailed description of embodiments with reference to the accompanying drawings.

Referring to the accompanying drawings, a preferred embodiment of the infrared vein image restoration method blurred by the scattering of the skin according to the present invention will be described. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various different forms, only the embodiments to complete the disclosure of the present invention and complete the scope of the invention to those skilled in the art. It is provided to inform you. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiments of the present invention and do not represent all of the technical idea of the present invention, various equivalents that may be substituted for them at the time of the present application It should be understood that there may be water and variations.

The reason why the boundary between the vein and the skin is blurred by scattering of the skin layer can be explained by the drawings.

As shown in FIG. 1A, assuming that there is no skin layer, the border of the vein will appear clearly in the image. However, in practice, since the skin layer surrounds the vein as shown in FIG. 1B, the boundary between the vein and the skin is vaguely observed in an image obtained by scattering light emitted by the skin layer.

The blurring of the boundary by the scattering layer through the skin layer may be explained by the convolution of the Gaussian-based point diffusion function as shown in FIG.

2 is a diagram illustrating a model for explaining a blur phenomenon caused by a scattering layer. Equation 1 is an equation for calculating a point spread function. At this time, d represents the thickness of the skin layer.

))는 도 2와 상기 수학식 1에 기반하여 추정한다. Referring to FIG. 2 and Equation 1, the point spread function P (

)) Is estimated based on FIG. 2 and Equation 1 above.

Therefore, it is assumed that there is no scattering layer, and the vein thickness in the image is assumed to be the thickness of a general human blood vessel, and a virtually generated clear vein image is estimated as shown in FIG. In this case, the line profile of the vein image is estimated based on the thickness of the vein of a general person corresponding to 1mm ~ 2mm.

On the other hand, the line profile of the actually obtained vein image is blurred as shown in FIG. 3 (b) because a scattering layer corresponding to the skin layer exists as shown in FIG. 1 (b).

As a result, an unknown parameter of the point spread function is estimated through line profile matching between the actually acquired blurred vein image as shown in FIG. 3 (b) and the virtually obtained vein image as shown in FIG. 3 (a).

는 피부층의 흡수계수를,

는 피부층의 산란계수를,

는

의 제곱근을,

는 흐려진 면적에 대한 레디얼(radial) 거리를, d 는 산란계층(손가락 피부)의 두께를 나타낸다. 이처럼 점 확산 함수(P(

))를 추정하기 위한 수학식 1에 적용되는 피부층의 흡수계수(

), 피부층의 산란계수(

), 레디얼 거리(

)들은 피부 광학 분야에서 표준화된 상수를 적용하여 사용할 수 있다. In more detail, in Equation 1

Is the absorption coefficient of the skin layer,

The scattering coefficient of the skin layer,

Is

Square root of

Is the radial distance to the blurred area and d is the thickness of the laying layer (finger skin). Like this, the point spread function (P (

Absorption coefficient of skin layer applied to Equation 1 for estimating

), The scattering coefficient of the skin layer (

), Radial distance (

) Can be used by applying a standardized constant in the field of skin optics.

(조명의 세기)는 표준화된 값으로 알려진 바가 없으며, 또한 산란계층(손가락 피부)의 두께(d)는 사람마다 손가락 두께가 서로 다르므로 정해진 상수 값을 바로 사용할 수가 없다.However

(Light intensity) is not known as a standardized value, and since the thickness d of the scattering layer (finger skin) is different from one finger to another, a constant value cannot be used directly.

및 d 값을 변경해 가면서 실제 취득된 정맥 영상인 도 3(b)가 상기 산란계층이 없다고 가정하여 추정된 도 3(a)와 같은 형태로 라인 프로파일 매칭을 통해 가장 유사하게 나타나는

및 d 의 값을 구하게 된다. Thus above

And FIG. 3 (b), which is an actually acquired vein image while changing the d value, appears most similarly through line profile matching in the form as shown in FIG.

And the value of d .

값 및 d 값을 수학식 1에 적용하여 최종적인 점 확산 함수(P(

))를 결정하게 된다.And

Value and d are applied to Equation 1 to obtain a final point spread function (P (

)).

))가 결정되면, 취득된 정맥 영상에 상기 결정된 최종적인 점 확산 함수를 이하 수학식 2에 적용하여 영상을 복원함으로써 흐려진 정맥 영상의 품질을 향상시킨다.The final point spread function (P (

When)) is determined, the final vein diffusion function determined in the acquired vein image is applied to Equation 2 below to reconstruct the image to improve the quality of the blurred vein image.

Equation 2 is a reconstruction equation using a convolution of a Gaussian-based point spread function, and the resultant corrupted image is calculated through the convolution of the point spread function and the original undamaged image.

In this case, o (x, y) denotes a resultant corrupted image in a spatial domain, h (x, y) denotes a point spread function, and i (x, y) denotes an original uncorrupted image. And n (x, y) represents noise. In addition, the O (u, v) is a resultant image in the frequency domain, the H (u, v) is a point spread function, the I (u, v) is a resultant damage in the frequency domain An image, and N (u, v) represents noise.

Therefore, as described above, when the point spread function h (x, y) or H (u, v ) is obtained, a given corrupted image o (x, y) or O (u, v ) is used. The original uncorrupted image i (x, y) or I (u, v) can be obtained using a reconstruction filter.

In this case, an inverse filter, a Wiener filter, and a Consort Least Square (CLS) filter may be used as the reconstruction filter.

))를 수학식 2를 통해 복원된 영상의 차이를 나타낸 실시예로서, 도 4(a)와 같이 피부층에 의해 산란되어 흐려진 혈관 영상을, 복원 필터를 통해 복원함으로써, 도 4(b)와 같이 정맥과 피부 사이의 경계가 명확해진 것을 확인할 수 있다.4A and 4B illustrate a point spread function P (

As an embodiment showing the difference of the image restored through the equation (2), as shown in FIG. It can be seen that the boundary between the vein and the skin becomes clear.

Although the technical spirit of the present invention described above has been described in detail in a preferred embodiment, it should be noted that the above-described embodiment is for the purpose of description and not of limitation. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the technical idea of the present invention. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1A and 1B are diagrams for explaining the reason why the boundary between the vein and the skin is blurred by scattering of the skin layer.

2 is a diagram illustrating a model for explaining a blur phenomenon caused by a scattering layer.

3 (a) and 3 (b) are embodiments showing virtually generated clear vein images and actual acquired blurred vein images

))를 통해 복원전과 복원후의 영상의 차이를 나타낸 실시예4 (a) (b) shows the estimated point spread function P (

An embodiment showing the difference between the image before and after restoration through))

Claims (7)

(a) a first venous image acquisition step of acquiring an image of a blood vessel having a relatively darker brightness than the skin by using a near infrared ray irradiation device and an infrared camera; (b) a second venous image acquisition step of generating a virtual blood vessel image based on a blood vessel thickness of a general person; (c) estimating an unknown variable of a point spread function through line profile matching between the first vein image and the second vein image; (d) calculating a point spread function based on the estimated unknown variable; (e) restoring the first vein image having an ambiguous vein boundary by restoring filtering using a convolution of a Gaussian-based point spreading function using the calculated point spreading function. How to restore a blurred infrared vein image. The method of claim 1, In step (b), the blood vessel thickness is 1mm or more, 2mm or less thickness of the infrared vein image restoration method blurred by the scattering of the skin, characterized in that. The method of claim 1, The unknown variable in step (c)

(Intensity of illumination) and the thickness ( d ) of the scattering layer is at least one or more infrared vein image restoration method blurred by the scattering of the skin. The method of claim 1, wherein step (d)

To calculate the point spread function using the formula of

Is the absorption coefficient of the skin layer,

The scattering coefficient of the skin layer,

Is

Square root of

The method of restoring the infrared vein image blurred by the scattering of the skin, characterized in that the radial (radial) to the blurred area, d represents the thickness of the scattering layer. The method of claim 1, wherein step (e) A method of restoring an infrared vein image blurred by scattering of skin, characterized in that the first vein image is restored using the resultant image obtained by the calculated point diffusion function and the convolution of the second vein image. The method of claim 5, wherein step (e)

Reconstructing the first vein image using the equation of, wherein o (x, y) is the resultant deterioration image in the spatial domain, and h (x, y) is the point spread function. , wherein i (x, y) is the original non-damaged image, and wherein n (x, y) denotes a noise, and the O (u, v) is the result of damaged image in the frequency domain, the H (u , v) represents a point spread function, I (u, v) represents a resultant image in the frequency domain, and N (u, v) represents noise Infrared vein image restoration method blurred by. The method of claim 6, wherein step (e) When the point spread function h (x, y) or H (u, v) is obtained, the original uncorrupted image i is obtained by using a given corrupted image o (x, y) or O (u, v) . (x, y) or I (u, v) ) is obtained using a reconstruction filter, The restoration filter used at this time is any one of an inverse filter (inverse filter), Wiener filter (wiener filter), CLS (Constraint Least Square) filter, the infrared vein image restoration method blurred by the scattering of the skin.

Images