KR100669625B1 - Assuming system of eyeball position using morphologic image processing and method thereof - Google Patents

Abstract

The present invention relates to an eye position estimation system and method for detecting the reflection noise of the pupil and processing the position of the eye by processing morphological operations using a specific structural element. An ambient dark area value is assigned to the pupil area of the reflective noise pixel of the pupil through an arithmetic processing to assign one of the image values of the pixels adjacent to the reflective noise by using a ring-shaped mask forming a diameter of a predetermined pixel size. A calculation processing module; And a pixel selection module for determining a difference between the eyeball image and the computed image and selecting a pixel area having a difference of an image value greater than or equal to a reference value.

According to the present invention, it is possible to detect only the reflection noise caused by the pupil through a minimized operation by using a morphological operation such as expansion operation and having a structural element that is different from the conventional one, thus avoiding a complicated preprocessing process and quickly irising It has the effect of processing recognition.

Description

Assuming system of eyeball position using morphologic image processing and method

1 is a block diagram schematically illustrating the components of an eyeball position estimation system using morphological computation according to an embodiment of the present invention.

2 is a view illustrating an original image of an eyeball area photographed by an image input module according to an embodiment of the present invention.

3 is a diagram illustrating a case in which reflection noise caused by a pupil is allocated to a dark value around by a calculation processing module according to an embodiment of the present invention;

4 is a diagram illustrating a screen result of subtracting an image subjected to expansion operation from an original image by a pixel selection module according to an exemplary embodiment of the present invention.

5 is a view comparing the form of the structural elements and the form of general structural elements according to an embodiment of the present invention.

6 is a diagram illustrating pixel values of an eyeball image in spatial coordinates according to an exemplary embodiment of the present invention, and showing a reflection noise and an entire eyeball image of a pupil at this time;

7 is a diagram illustrating each pixel value according to the calculation after spatial expansion processing of an eye image according to an exemplary embodiment of the present invention on spatial coordinates, illustrating the reflection noise of the pupil and the whole eye image at this time;

FIG. 8 is a diagram illustrating each pixel value after subtracting an expansion image processed by subtraction from an original eye image according to an embodiment of the present invention, and illustrating the reflection noise and the whole eye image of the pupil at this time .

9 is a flowchart illustrating an eyeball position estimation method using morphological calculation processing according to an embodiment of the present invention.

<Explanation of symbols for main parts of drawing>

100: Eye Position Estimation System Using Morphological Operation

110: image input module 120: operation processing module

130: control module 140: storage means

150: pixel selection module 160: position estimation module

The present invention relates to a system and method for estimating eye position as a pretreatment process for iris recognition.

Currently, active research is being conducted on a system for authenticating an individual using biological characteristics such as genetic traits, fingerprints, voices, veins, facial features, and eye irises.

Among them, the iris recognition field is expected to be used most in the future due to the advantages of high recognition rate, impossibility of counterfeiting, pattern data with a large amount of data, and no change factor.

The iris (Iris) refers to the area between the pupil and the white area, which is comb-like ligaments, red fibers, cilia-like protrusions, vascular forms, ring-shaped tissues, and corona-shaped ligaments surrounding the pupil. Because of its unique characteristics such as color, spot color, etc., the iris can be used to identify individuals.

In order to reduce the amount of data computation, a preprocessing process for finding the eye position is performed before the iris is recognized. This process can be performed in various ways. Conventionally, using the image information according to the shape of the eye to find the position of the eye or specular noise represented by the pupil (Specular noise component when the light is reflected in the pupil has a regular characteristic in size and pattern) And the like.

In this way, when the eye is located, the iris recognition system is precisely driven so that the subject's eye and the lens of the camera are optically matched, and the iris is precisely photographed to process the iris recognition process.

However, in case of finding the position of the eye by using the shape of the eye, the shape of the eye varies depending on factors such as makeup and artificial eyebrows, or the shape of the eye varies according to the characteristics of the individual, or due to wearing contact lenses or glasses. Since there are many obstacles such as the shape of the eye can be perceived differently, it is technically difficult to detect the image object called the eye uniformly through image processing.

In addition, various noise components, including reflection noise, appear in the image of the eyeball area, and these noise components cover part of the eye so that the shape of the eye cannot be recognized. Sometimes it can't be.

For this reason, a method of finding the position of the eye by detecting a specific light and detecting a reflection noise generated only in the pupil, in particular, is used. In this case, various reflection noise components act as obstacles.

As the reflection noise components due to glasses and the like increase, it becomes difficult to detect the reflection noise components due to the pupils, and the computational processing requires complicated and time-consuming operations, thereby increasing the false recognition rate of the iris recognition system and causing inconvenience to the user.

On the other hand, the reflected noise component due to the pupil is always detected in the same pattern (small size, circle shape, etc.) in spite of environmental differences such as illumination, camera position, aperture value, shutter speed, etc. Reflective noise components are large in size and irregular in shape, and have various shapes and locations of occurrence.

Accordingly, there is a demand for a method of detecting eye position with minimal processing and high reliability by using reflection noise characteristics of the pupil.

Accordingly, the present invention provides an eyeball position estimation system that detects the reflected noise of the pupil and detects the eyeball position by performing signal processing using characteristics that are always detected in the same pattern without being affected by environmental differences. For that purpose.

In addition, the present invention is to provide a method for estimating the eyeball position to provide a specific constituent element and to handle the morphological calculation to cancel the reflection noise component caused by external factors and to detect only the reflection noise component by the pupil The purpose.

In order to achieve the above object, the eyeball position estimation system using the morphological calculation process according to the present invention uses the ring-shaped mask to form the eyeball image diameter of a predetermined pixel size from among the image values of the pixels adjacent to the reflection noise. An arithmetic processing module for allocating a peripheral dark region value to the pupillary reflective noise pixel region through an arithmetic processing for assigning any one value; And a pixel selection module for determining a difference between the eyeball image and the computed image and selecting a pixel area having a difference of an image value greater than or equal to a reference value.

In addition, the eye position estimation system using the morphological calculation process according to the present invention is characterized in that it further comprises a position estimation module for estimating the eye position by determining the selected pixel area as the reflection noise of the pupil.

delete

In addition, the structural element processed by the eyeball position estimation system using the morphological calculation process according to the present invention is characterized by forming a diameter of 1 pixel to 4 pixels.

delete

b(s, t) = max{f(s-x, t-y)│(x, y)

∈D_b}, (여기서, "f(x, y)"는 안구 영상을 의미하고, "f(s-x, t-y)"는 안구 영상의 화소가 x축 상에서 수치 s만큼 이동되고, y축 상에서 수치 t만큼 이동됨을 의미한다. 또한, "b(x, y)"는 구조적 요소를 의미하고, "D_b"는 b의 정의역을 의미한다. 또한, 연산"

"은 상기 구조적 요소 위치에 존재하는 화소들 중 최대값을 f(x, y)에 할당하는 연산을 의미함)"의 식으로 표현되는 것을 특징으로 한다.Further, the calculation processing performed by the eyeball position estimation system using the morphological calculation processing according to the present invention is "f.

b (s, t) = max {f (sx, ty) │ (x, y)

∈D _b }, (where “f (x, y)” refers to the eye image, and “f (sx, ty)” moves the pixel of the eye image by the number s on the x-axis and the number on the y-axis. "b (x, y)" means structural elements and "D _b " means domain of b.

Is an operation of allocating a maximum value of f (x, y) among the pixels existing at the structural element position.

In order to achieve the above object, the eye position estimation method using the morphological calculation process according to the present invention comprises the steps of photographing an eyeball image; An ambient dark area value is assigned to the pupil area of the reflective noise pixel of the pupil through an arithmetic processing to assign one of the image values of the pixels adjacent to the reflective noise by using a ring-shaped mask forming a diameter of a predetermined pixel size. Doing; Selecting a pixel region having a difference of an image value greater than a reference value by obtaining a difference between the eyeball image and the computed image; And determining the eye position by determining the selected pixel area as reflection noise of the pupil.

Hereinafter, an eyeball position estimating system and an eyeball position estimating method using morphological calculation processing according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram schematically illustrating the components of an eyeball position estimation system 100 using morphological computation according to an embodiment of the present invention.

Referring to FIG. 1, the eyeball position estimation system 100 using the morphological calculation process according to an embodiment of the present invention includes an image input module 110, a calculation processing module 120, a pixel selection module 150, and a position estimation module. It comprises a 160, the control module 130 and the storage means 140.

Usually, the process of finding the position of the eyeball is preceded to avoid unnecessary computation before performing the iris recognition. The image input module 110 photographs the eyeball area to find the position of the eyeball, and the photographed eyeball image is stored. Stored at 140.

For reference, when the image input module 110 captures the eyeball image and the correct position of the eyeball is determined, another image input device may be further driven to accurately detect the eyeball position to capture the iris recognition image.

The control module 130 transmits an enable signal to sequentially function the image input module 110, the operation processing module 120, the pixel selection module 150, and the position estimation module 160. The image data, arithmetic processing data, and position tracking data, etc. captured in 140 are stored and managed.

First, the arithmetic processing module 120 performs a morphological operation on the photographed eyeball image by a structural element, so that the reflection noise caused by the pupil is allocated to a dark value of the surroundings, and the other dark portions are reduced or eliminated. .

The structural element has a kind of mask form and at the same time means a sub-image function (see FIG. 5).

2 to 4 are schematic diagrams illustrating a process of processing an eyeball image according to an exemplary embodiment of the present invention.

According to Figures 2 to 4, the basic structure of the eye is shown, the eye can be largely divided into a hinja (a), iris (b) and pupil (c).

FIG. 2 shows the original image of the eyeball area captured by the image input module 110, and FIG. 3 shows that the reflection noise c1 due to the pupil is assigned to the dark value around by the arithmetic processing module 120. Is a diagram illustrating the case. 3, it can be seen that portions other than the reflection noise caused by the pupil (including other reflection noise) are relatively unchanged.

4 is a diagram illustrating a result screen of the pixel selection module 150 subtracting an expansion-processed image from an original image.

According to FIG. 4, only the reflection noise component c1 due to the pupil can be extracted with a relatively high image value compared to the surroundings by subtracting the computed image shown in FIG. 3 from the original image shown in FIG. 2. It can be confirmed.

Various morphological operations can be used to process the eyeball image. In the present invention, a division operation is used.

The expansion operation means performing a set operation of a predetermined function having structural elements with respect to the origin pixel of the original image, and performing the set operation continuously by being moved while being overlapped by at least one element.

Here, the set operation of a predetermined function is to assign a value of a pixel to a maximum value among image values of adjacent pixels. When an expansion operation is processed, the output image is brighter than the input image when all values of structural elements are positive. The dark parts are reduced or eliminated depending on the relationship between the set operation and the shape defined in the structural element. Therefore, bright parts of the input image become relatively large.

For this reason, when the expansion operation is performed on an eyeball image, the noise component of the pupil is generated by the reflection of light when the eye region is photographed. For example, the reflection noise component has a high image value such as brightness compared to the surrounding pixel region), and the pixel region corresponding to the reflection noise is emphasized.

As described with reference to Figs. 2 to 4, the present invention processes the dilation operation to assign the surrounding dark area values to the pupil area of the reflective noise pixel (whereas the non-pupillary reflection noise areas are brightened by the dilation operation). The area is enlarged) Since only the reflection noise of the pupil is extracted by subtracting the calculated image from the original image, it is necessary to distinguish the form of the above-described structural element from the conventional art.

5 is a view showing a comparison between the shape of the structural element and the form of the general structural element according to an embodiment of the present invention.

According to (a) of FIG. 5, the general structural element A has a square shape, and when the general structural element is moved (A ′) and the set operation is performed, the reflection noise region c1 of the pupil is different from the reflection noise. The effect is to swell like an area. That is, a general structural element is to process a number of operations even inside the pupillary reflection noise region.

However, as shown in (b) of FIG. 5, the structural element B according to the present invention has a ring shape, and the structural element according to the present invention is moved (B`) and the reflection operation of the pupil is performed when the set operation is performed. Unlike the other reflection noise regions (or other screen regions), the region c1 is assigned a peripheral dark region value. Referring to FIG. 5B, the structural element B according to the present invention is defined to match the size of the reflected noise region c1 of the pupil.

Usually, since the reflection noise region c1 due to the pupil forms a diameter of 1 to 4 pixels, the ring-shaped structural element B also has a diameter r of 1 to 4 pixels. It is preferably defined to be.

Next, the above-described expansion operation will be described in detail.

fb(s, t) = max{f(s-x, t-y)│(x, y)∈D_b}

fb (s, t) = max {f (sx, ty) │ (x, y) ∈D _b }

Here, first, "f (x, y)" means a pixel of the eyeball image,

Second, "f (s-x, t-y)" means that the pixels of the eyeball image are shifted by the number s on the x-axis and shifted by the number t on the y-axis.

Third, "b (x, y)" means structural elements,

Fourth, "D _b " means the domain of b,

"은 상기 구조적 요소 위치에 존재하는 화소들 중 최대값을 f(x, y)에 할당하는 연산을 의미한다.Fifth, morphological operations

"Means an operation of allocating a maximum value of f (x, y) among pixels existing at the structural element position.

FIG. 6 is a diagram illustrating pixel values of an eyeball image in spatial coordinates according to an exemplary embodiment of the present invention, and the reflected noise and the entire eyeball image of the pupil at this time.

According to Fig. 6, bright numerical pixel areas appear in two places D1 and E1, where the first bright pixel area D1 represents the reflection noise area by the pupil and the second bright pixel area E1 is another factor. It shows the reflection noise area by.

It can be seen that the size of the reflection noise area D1 due to the pupil is smaller than that of the reflection noise area due to other factors (1 to 4 pixels).

FIG. 7 is a diagram illustrating a pixel value of each pixel according to the calculation after the expansion operation is performed on the eye image according to an exemplary embodiment of the present invention, and the reflection noise and the whole eye image of the pupil at this time.

According to FIG. 7, through the structural elements and the expansion operation according to the present invention, it is confirmed that the first bright pixel area D2 has been assigned the dark value of the surroundings and the original brightness value is significantly reduced. Rather, (E2) shows that the figure has expanded.

It can be seen that other screen regions are relatively less affected by the surrounding image values than the reflective noise region D2 of the pupil.

FIG. 8 is a diagram illustrating each pixel value after subtracting an expansion image processed by subtraction from an original eye image according to an embodiment of the present invention, and illustrating the reflection noise and the whole eye image of the pupil at this time to be.

According to FIG. 8, by subtracting the image shown in FIG. 7 from the image shown in FIG. 6, the numerical value of the second bright pixel region E3 is almost close to “0”, and the value of the first bright pixel region D3 is reduced. The figure almost maintains the value on the original image.

Therefore, since only the reflection noise region D3 of the pupil has a bright pixel value on the final processed eye image, it is possible to easily detect the position of the reflection noise D3 due to the pupil.

As described above, the pixel selection module 150 subtracts the original eyeball image and the morphologically processed image to select a pixel area having a difference in image value greater than or equal to a reference value.

When a pixel region having a difference in image value greater than or equal to a reference value is selected, the position estimation module 160 determines the eye position by determining the selected pixel region as reflection noise of the pupil, and the estimated eye position information is stored in the storage means ( 140).

9 is a flowchart illustrating a method for estimating eyeball position using morphological computation according to an embodiment of the present invention.

First, when the subject places the eyeball area on the image input module 110, the image input module 110 photographs the eyeball area. The photographed eyeball image is stored in the storage unit 140 (S100).

Subsequently, the calculation processing module 120 performs an expansion operation while moving the structural element having a ring shape on the photographed eye image (S200).

As mentioned above, the structural element used in the present invention is defined as a ring shape that forms a diameter of 1 to 4 pixels in size.

At this time, the reflection noise area due to the pupil is assigned to the dark value of the surroundings, which is assigned a significantly lower value than the image value of the original pupil reflection noise.

On the other hand, the reflection noise area due to other factors has an effect of expanding, and even though some other areas of the screen are assigned to the dark value of the surroundings, unlike the reflection noise caused by the pupil, low values of many differences are not allocated. Do not.

When the expansion operation is performed on the entire area of the eyeball image (S300), the pixel selection module 150 subtracts the expansion operation image from the original eyeball image (S400).

In the subtracted image, only the pupil reflection noise region remains as a region having a high value because the pupil reflection noise region allocated to the dark value of the surrounding is subtracted from the original pupil reflection noise having a high value.

The pixel selection module 150 selects a pixel area whose difference in the image value is greater than or equal to the reference value as a result of subtraction, and only the pupil reflection noise area corresponds to an area having a value larger than the reference value (S500).

The pupil reflection noise region having a constant size and image value can also be detected at a high detection rate by structural elements and expansion calculations set according to these characteristics, and the position estimation module 160 can be detected by the detected pupil reflection noise region. Estimate the correct eye position (S600).

Finally, when the correct eye position is estimated, the iris recognition process is performed (S700).

The present invention has been described above with reference to the preferred embodiments, which are merely examples and are not intended to limit the present invention, and those skilled in the art to which the present invention pertains do not depart from the essential characteristics of the present invention. It will be appreciated that various modifications and applications are not possible that are not illustrated above. For example, each component specifically shown in the embodiment of the present invention can be modified. And differences relating to such modifications and applications will have to be construed as being included in the scope of the invention defined in the appended claims.

According to the eye position estimation system using the morphological calculation process according to the present invention, it is possible to detect only the reflection noise due to the pupil through a minimized operation by using a morphological operation such as expansion operation and having structural elements that are different from the conventional one. As a result, it is possible to avoid complex pretreatment and to rapidly process iris recognition.

In addition, according to the present invention, not only the reflection noise caused by external factors but also the image areas having different brightness values can be effectively excluded according to the environment such as the angle of illumination, the reflectance of the face surface, the camera lens optical axis, and the like. It can be improved.

Claims (7)

The dark area around the pupil area of the reflective noise pixel of the pupil through an arithmetic processing to assign an eye image an image value of pixels adjacent to the reflective noise using a ring-shaped mask forming a diameter of a predetermined pixel size. An operation processing module for assigning a value; And And a pixel selection module for determining a difference between the eyeball image and the computed image and selecting a pixel area having a difference of an image value greater than or equal to a reference value. The method of claim 1, And a position estimation module for estimating the eye position by judging the selected pixel region as reflection noise of the pupil. delete The method of claim 1, And the predetermined pixel size forms a diameter of 1 pixel to 4 pixels. delete The method of claim 1, wherein the operation processing f

b (s, t) = max {f (sx, ty) │ (x, y)

∈D _b }, (where “f (x, y)” refers to the eye image, and “f (sx, ty)” moves the pixel of the eye image by the number s on the x-axis and the number on the y-axis. In addition, "b (x, y)" means a structural element, "D _b " means a domain of b. Also, operation " Eyeball position estimation system using a morphological calculation process, characterized in that "is the operation of assigning the maximum value of f (x, y) of the pixels present in the structural element position). Photographing an eyeball image; An ambient dark area value is assigned to the pupil area of the reflective noise pixel of the pupil through an arithmetic processing to assign one of the image values of the pixels adjacent to the reflective noise by using a ring-shaped mask forming a diameter of a predetermined pixel size. Doing; Selecting a pixel region having a difference of an image value greater than a reference value by obtaining a difference between the eyeball image and the computed image; And Estimating the eye position by determining the selected pixel region as the reflection noise of the pupil.

Images