JP4883987B2 - Face image collation device and collation face image registration method - Google Patents

Description

  The present invention relates to a face image collation device that collates a face image obtained by photographing a person to be collated with a face image acquired in advance and determines whether or not they are the same person, and a collation face image registration method used for the collation. More specifically, a face image collation device that performs collation based on an input face image obtained by photographing a person to be collated, and a face image acquired in advance and having different characteristics from the input face image, and a collation face used for the collation The present invention relates to an image registration method.

  In recent years, collation devices that perform personal authentication using image information obtained by photographing a face have been proposed. For example, a face image identification card that reads a face image printed on a personal identification (ID) card and simultaneously reads the image of the card user with an input device so as to match the face image of the card owner and the user. A collation processing method has been proposed (see Patent Document 1).

  In addition, it has been found that a person's face is characterized by individual features such as facial contours, positions of characteristic parts such as eyes and mouth, and eye shapes. Therefore, in such a collation device, each feature of a reference image (hereinafter referred to as a registered face image) registered in advance is extracted from the captured face image as a feature amount such as the position and shape of the eyes and mouth. The degree of similarity with the quantity is calculated, and collation is performed based on whether the degree of similarity satisfies a predetermined criterion.

  In many cases, a face image used for collation is taken using visible light (wavelength of about 400 nm to about 700 nm) or near infrared light (wavelength of about 700 nm to about 850 nm) as illumination light. .

  When photographing using a visible light source as an illumination light source, there is an advantage that no special photographing equipment is required and color information can also be used as a feature amount for collation. However, the brightness and color tone of the acquired image may vary due to the influence of the surrounding lighting environment. In order to avoid such an adverse effect, increasing the illuminance of the illumination light that strikes the face of the person is undesirable because it causes the subject to feel uncomfortable, such as dazzling.

  On the other hand, when shooting with a near-infrared light source as the illumination light source, even if it is set to have a relatively high illuminance, the subject will rarely feel dazzling, ensuring stable lighting conditions. It is easy and therefore the quality of the acquired image is stable, and there is a merit that it is easy to obtain high collation accuracy, and it is used more widely than that using visible light as an illumination light source.

  When the illumination light is near-infrared light, it is difficult to obtain color information, and since only the information related to brightness is included, the captured image is a monochrome image. On the other hand, when the registered face image used for collation is a color image taken under visible light, high collation accuracy cannot be obtained due to the difference in information and contrast contained in both. Therefore, when using a face image captured using near-infrared light for illumination as input (hereinafter referred to as an input face image), the registered face image may also be an image captured using near-infrared light for illumination. desirable.

  However, in order to photograph near infrared light as illumination light, a photographing device such as a near infrared illumination device and a camera having sensitivity in the near infrared region is required. Absent. Therefore, in order to prepare a registered face image, a special work is required such that the person to be photographed actually visits the collation device and takes a picture. If the number of registrants reaches several hundreds or thousands, the man-hours and time required for this work become enormous. Therefore, it is desirable to reduce the man-hours required for the above-described work by using a color image that is generally captured with a visible light photographing device. In the case of a color image, an existing image such as a face image taken separately for an identification card such as an employee ID card may be used.

  On the other hand, when a registered face image is a color image, a method and an apparatus for matching a face by converting it to a monochrome image have been proposed (see Patent Document 2). In the method and apparatus described in Patent Document 1, the color image is converted to monochrome so as to avoid the adverse effect of changes in face color information caused by changes in makeup and lighting, and to improve the recognition accuracy of collation. Yes. Similarly, a recognition device that recognizes a recognition target using a monochrome image of a color image is known in order to accurately recognize an IC position in mounting a semiconductor chip (see Patent Document 3). .

  However, an image obtained by simply converting a color image into a monochrome image has a contrast different from that of a face image taken by illuminating with near infrared light. Therefore, if a simple monochrome image is used for the registered face image and a near-infrared image is used for the input face image, it is not always possible to obtain a good recognition result.

  Therefore, it is desired to develop a face image collation apparatus that registers face images suitable for collation based on color face images that can be easily obtained, and performs collation using the registered face images, and a method for registering face images. ing.

Japanese Patent Laid-Open No. 5-233783 JP 2000-331167 A JP 2002-26075 A

  In view of the above problems, an object of the present invention is to provide a face image collation apparatus and a face image registration method capable of registering a face image suitable for collation using a color face image.

  Another object of the present invention is to provide a face image collation device and a face image registration method that facilitate registration of face images for collation.

  A first invention for solving such a problem is a face image collation device that collates a face image inputted by a person to be collated with a registered face image by photographing the person to be collated under a predetermined environment. A face image input unit that inputs a face image, a color face image acquisition unit that acquires a color face image of a person to be verified, and an estimated face image that can be compared with a face image shot under a predetermined environment based on the color face image An estimation unit to be calculated, a collation unit that collates the input face image acquired by the face image input unit with the estimated face image and determines whether or not both face images belong to the same person, and a registration that stores the registered face image An image storage unit, and a face image registration unit that registers an estimated face image in the registered image storage unit as a registered face image when the collation unit determines that the person is the same person, and does not register unless the collation unit determines that the person is the same person Provided is a face image collating device.

  According to such a face image matching device, a color face image that is generally easy to acquire and has a larger amount of information than a monochrome image is acquired. An estimated face image suitable for face image matching is created from the acquired color face image. If the face image photographed under a predetermined environment and the estimated face image are collated, and the person is the same person, the estimated face image is registered as a registered face image used for subsequent face image collation.

  That is, when an estimated face image inappropriate for collation is created from a color face image for some reason, registration of an estimated face image unsuitable for collation as a registered face image can be prevented. Note that the color face image acquisition unit, the estimation unit, the face image input unit, and the registered image storage unit are not necessarily provided in the same device as the collation unit, and can transmit and receive predetermined information via a network, for example. You may provide in another apparatus. In addition, the predetermined environment refers to a face image shooting environment when the person to be verified matches the face image, and differs depending on the type of illumination, the type of image sensor, the type of filter, and the like.

  A second invention for solving such a problem is a face image collation device that collates a face image input by a person to be collated with a registered face image by photographing the person to be collated under a predetermined environment. A face image input unit that inputs a face image, a color face image acquisition unit that acquires a color face image of a person to be verified, and an estimated face image that can be compared with a face image shot under a predetermined environment based on the color face image An estimation unit to be calculated, a collation unit that collates the input face image acquired by the face image input unit with the estimated face image and determines whether or not both face images belong to the same person, and a registration that stores the registered face image An image storage unit, and a face image registration unit that registers an input face image as a registered face image in the registered image storage unit when the collation unit determines that the person is the same person, and that does not register unless the collation unit determines the same person. Provided is a face image collating device.

  According to such a face image matching device, a color face image that is generally easy to acquire and has a larger amount of information than a monochrome image is acquired. An estimated face image suitable for face image matching is created from the acquired color face image. If a face image shot under a predetermined environment is collated with an estimated face image and the same person is registered, the input face image shot under the predetermined environment is registered as a registered face image used for subsequent face image matching. .

  That is, an estimated face image artificially created from a color face image is not used as a registered face image, but an input face image photographed under a predetermined environment is registered, so that a more appropriate face image can be registered. . Note that the color face image acquisition unit and the estimation unit are not necessarily provided in the same device as the verification unit, as in the first invention. In addition, the predetermined environment refers to a face image shooting environment when the person to be verified matches the face image, and differs depending on the type of illumination, the type of image sensor, the type of filter, and the like.

In the first invention and the second invention, when the face image registration unit determines that the matching person is the same person, the face image registration unit registers the estimated face image or the input face image as a registered face image in the information storage medium, and It is preferable not to register unless it is determined as the same person.
According to such a configuration, since the registered face image is stored in the information storage medium, it is not necessary to store the face image in the face image matching device. Further, for a person to be collated using a plurality of face image collation devices, the face image used for collation can be used in a plurality of devices simply by registering the face image used in the self-owned information storage medium.

In the first invention and the second invention, it is preferable that the color face image acquisition unit further includes a network communication unit, and acquires the color face image of the person to be verified through the network communication unit.
According to this configuration, when the color face image is centrally managed by the central management device in advance, the color face image can be acquired by connecting to the central management device via the network. For this reason, the registration of the face image can be saved.

In the first and second inventions, it is preferable that the color face image acquisition unit further includes a data reading unit from the information storage medium, and acquires the color face image of the person to be verified from the information storage medium. is there.
According to such a configuration, when a color face image is stored in advance in a portable information storage medium, the face image can be registered without the need for troublesome network connection or the like.

In the first invention and the second invention, the estimation unit includes a pixel value extraction unit that extracts, for each pixel, color gradation information included in the pixel value of the pixel of the color face image, and a skin color reflection characteristic for each pixel. Based on the correlation between the color gradation information based on the pixel value of the near infrared region and the color gradation information extracted by the pixel value extraction unit, the estimation means for calculating the estimated pixel value at the wavelength of the near infrared region, It is preferable to have estimated image creating means for constructing an estimated face image from pixels having estimated pixel values estimated by the estimating means.
According to such a configuration, when a near-infrared face image is used as the predetermined environment, the near-infrared face image is estimated based on the skin color reflection characteristics using the color gradation information of the color face image, so that it is more suitable for collation. An estimated face image can be created.

  Further, the third invention for solving the above-described problem is a color face image acquisition step for acquiring a color face image of a person to be verified, and a face image photographed under a predetermined environment based on the color face image. An estimation step for calculating an estimated face image, a face image input step for photographing a person to be verified under a predetermined environment and inputting a face image, and an input face image and an estimated face image acquired in the face image input step A collation process for collating and determining whether both face images belong to the same person, and a face image that is registered if it is determined that the face is the same person in the collation process and is not registered unless it is determined to be the same person And a face image registration method for collation including the registration step.

  ADVANTAGE OF THE INVENTION According to this invention, it became possible to provide the face image collation apparatus and the face image registration method which can register the face image suitable for collation using a color face image.

  In addition, it is possible to provide a face image collation apparatus and a face image registration method that facilitate the registration of collation face images.

Hereinafter, a first embodiment in which the present invention is applied to a face image matching apparatus will be described with reference to the drawings.
FIG. 1 is a diagram showing an overall system configuration of an example in which a face image matching device 1 to which the present invention is applied is used for electric lock control at an entrance. In the figure, the system captures a face image of a person to be verified, compares the face image with a registered face image, and if a person can be identified, a plurality of face image verification devices 1 that unlock the electric lock 4 of the entrance door, and face image verification From a centralized management apparatus 2 that centrally manages color face images as a database for registration as face images for verification in the apparatus 1, and from a network 3 that can connect the face image verification apparatus 1 and the centralized management apparatus 2 Composed.

  The central management apparatus 2 is configured by a server, and a color face image of the person to be verified is stored in the database 21. FIG. 5 shows the contents stored in the database 21, and stores the employee number, ID number, and color face image of the person to be verified in association with each other.

  Here, a color face image will be described. A color face image is a digital image that includes a human face as a subject and has information that can be expressed in color for each pixel. The color image color system includes an RGB color color system, a YUV color color system, a YIQ color color system, a CMY color color system, and the like, and the color system can be mutually converted. Such color color system identification information is stored in the header of an image file. By reading and referring to this identification information by a computer, a color image can be expressed with an appropriate color gradation on a display or printer. Can do. The color face image here is a pixel value including color gradation information for each component such as R (red), G (green), and B (blue) components in the RGB color system for each pixel. have.

  FIG. 2 is a functional block diagram of the face image matching device 1 according to the first embodiment of the present invention. The face image matching device 1 is a color face that acquires a color face image from a centralized management device 2 connected to a network 3 and an operation display unit 101 of a liquid crystal display with a touch panel (LCD) that displays human operation inputs and various guidance. A communication unit 102 that functions as an image acquisition unit, and a near-infrared face that irradiates a person to be verified with near-infrared light, captures the face of the person to be verified, and acquires a near-infrared face image that is an input face image An image input unit 103, a registered image storage unit 105 that stores a registered face image that is a face image of a person who is permitted to pass through an entrance in advance, and outputs an unlocking or locking signal to the electric lock 4. An output unit 106 and a control unit 104 that performs various controls of the face image matching device 1 are configured.

  The operation display unit 101 is composed of an LCD touch panel. When the person to be collated gives a collation instruction on the touch panel or when a color face image is instructed to be acquired, various operation signals such as ID number input / operation selection are displayed. Perform operation input. In addition, the operation display unit 101 appropriately displays to the operator information indicating that verification is being performed, a verification result, an operation guide, and the like. In the present embodiment, the operation display unit 101 is described using an LCD touch panel, but various key switches and LEDs may be employed.

  The communication unit 102 serving as a color face image acquisition unit is connected to the central management apparatus 2 via the network 3 and includes a communication interface and a driver that perform communication using a predetermined communication protocol. In the present embodiment, a color face image is acquired from the centralized management apparatus 1 via the network 3, but the communication unit 102 is an RFID (Radio Frequency IDentification) reader / writer as shown in FIG. A color face image configured by the card R / W 112 and stored in the IC card may be acquired from the non-contact card R / W 112. Needless to say, a contact type may be used as the card R / W 112.

  The near-infrared face image input unit 103 includes a near-infrared illumination light source that illuminates near-infrared light (not shown), an image sensor such as a CCD element or a C-MOS element that is sensitive to light in the near-infrared region, and the front surface of the image sensor. And a visible light cut filter that has sufficient transmission characteristics in the near infrared region and blocks visible light, and an input circuit that captures the captured near infrared face image. In the present embodiment, near-infrared face images are acquired for face image matching in consideration of the glare of the person to be collated, but in spite of this, matching without using near-infrared face images When applying to processing, a face image suitable for the collation may be acquired.

  The registered image storage unit 105 includes a rewritable nonvolatile memory such as a flash memory or an EEPROM, and stores an ID number and a registered face image for each person to be verified.

  When the collation result is OK, the output unit 106 outputs an unlocking or locking signal to the electric lock 4 via an electric lock control device (not shown) installed at the entrance door.

  The control unit 104 includes a CPU, a ROM, a RAM, and the like, and performs overall control of the face image matching device 1 according to a program. In particular, an image estimation unit 104a that performs an image estimation process for creating an estimated face image from a color face image, a collation unit 104b that collates / determines whether the person to be collated is the same person based on the acquired face image, It has at least a face image registration unit 104c for registering the estimated face image that has been verified to the registered image storage unit 105. In addition, the image estimation unit 104a includes a pixel value extraction unit that extracts, for each pixel, color gradation information included in the pixel value of the pixel of the color face image, color gradation information based on skin color reflection characteristics, and near infrared rays for each pixel. An estimation means for calculating an estimated pixel value at a wavelength in the near-infrared region based on the correlation with the pixel value of the area and the color gradation information extracted by the pixel value extraction unit, and an estimated pixel estimated by the estimation means Estimated image creation means for constructing an estimated face image from pixels having values. In addition, about other processes, such as display control and communication control, since the technique currently implemented normally is used, detailed description is abbreviate | omitted.

Next, a processing flow of the face image matching device 1 will be described with reference to FIG.
First, it is assumed that the face image matching device 1 is turned on and set in a state in which communication with the centralized management device 2 is possible. The centralized management device 2 stores in advance in the database 21 color face images of persons who have been permitted to enter and exit.

  In this state, when the ID number of the person to be verified is input from the operation display unit 101 (step S401), the control unit 104 determines whether a registered face image corresponding to the ID number exists in the registered image storage unit 105. Is determined (step S402). If a registered face image exists in step S402, the process proceeds to step S404, the corresponding registered face image is read from the registered image storage unit 105, and the process proceeds to step S406 (step S404). If the registered face image does not exist in step S402, color face image acquisition processing is executed (step S403). The color face image acquisition process S403 accesses the database 21 of the centralized management apparatus 2 from the communication unit 102 (color face image acquisition unit) via the network 3, and uses the input ID number as a key and the database shown in FIG. The color face image corresponding to the ID number is downloaded from. In this flow, the description about the case where the face image corresponding to the ID number does not exist in the database 21 of the centralized management apparatus 2 is omitted. The operation is displayed on the operation display unit 101 and the process is terminated.

The process proceeds from the color face image acquired in step S403 to an estimated image construction process that is a near-infrared face image (step S405).
Here, with reference to FIG.7 and FIG.8, the process which estimates a near-infrared face image from a color face image is demonstrated in detail. FIG. 8 shows a specific process of the estimated image configuration process S405.

The color system identification information in the header of the color face image file acquired in step S403 is read, and it is determined whether or not the color system is an RGB color system (step S801).
If the color system is not the RGB color system, the process proceeds to step S802, where the color system is converted to the RGB color system, and the process proceeds to step S803.
If the color system is an RGB color system, the process proceeds directly to step S803.

In step S803, the pixel value extraction unit of the image estimation unit 104a sequentially specifies the pixels constituting the face image. Each pixel constituting the specified color face image is a color in which each color component of R (red), G (green), and B (blue) is expressed by 8 bits (256 gradations of pixel values 0 to 255). It has a pixel value including gradation information. For each component of R, G, and B, for example, the amount of light received by a photoelectric conversion element such as a CCD sensor or a C-MOS sensor, which is monochromatic with a color filter centered at wavelengths of 660 nm, 550 nm, and 440 nm, respectively, is 8 bits. It is digitalized.
In step S804, I _r , I _g , and I _b that are pixel values of the target pixel are extracted, and the process proceeds to step S805.

In step S805, the estimation unit of the image estimation unit 104a substitutes the extracted pixel values I _r , I _g , and I _b into Equation (1) described later to calculate an estimated pixel value I _ir .
Here, Formula (1) will be described. First, prior to the description of Equation (1), the reflection characteristics of human skin will be described with reference to FIG.

  A graph of the reflection characteristics of human skin is shown in FIG. In FIG. 7, the horizontal axis represents the wavelength, and the vertical axis represents the reflectance for each wavelength. A graph 61 indicated by a solid line represents the reflection characteristics of human skin. As is apparent from the graph, the reflectance of the skin occupying most of the human face increases in the order of B (blue), G (green), and R (red). Further, it can be seen that in the range of wavelengths from 400 nm to about 750 nm, the reflectance of the skin increases proportionally as the wavelength increases.

Therefore, the reflectance of near-infrared light is obtained by approximation with a straight line obtained using, for example, the least square method from the reflectances of wavelengths 440 nm, 550 nm, and 660 nm, as shown by the graph 62 shown by the dotted line in FIG. Can do.
That is, the correlation between the reflectances of B, G, and R, which are color gradations based on the skin color reflection characteristics, and the reflectance in the near infrared region is shown in the graph 62.

  From this, the intensity of each wavelength component of the illumination light based on an image obtained by photographing a face with illumination light containing visible light of a plurality of wavelengths, preferably illumination light containing three wavelengths of visible light separated from each other by several tens of nanometers In addition, from the reflected light intensity from the skin corresponding to each illumination wavelength, it is possible to approximately estimate the light intensity due to skin reflection when illuminated with near infrared light (for example, wavelength 760 nm). An infrared face image can be created.

なお、数式（１）において、Ｒ、Ｇ、Ｂの各波長Ｗ_r、Ｗ_g、Ｗ_bは、例えば４４０ｎｍ、５５０ｎｍ、６６０ｎｍの値を用いる。また、実際のカラー画像の作成時においては、必ずしもＲ、Ｇ、Ｂの各波長は上記の値ではないが、一般的な環境で撮影されるデジタル画像では、Ｒ、Ｇ、Ｂの各波長の値を上記の値と仮定しても、顔画像の照合に使用するという目的下では、十分な照合精度を得ることができる擬似近赤外線顔画像を作成できる。 Assume that the wavelength of R is w _r , the wavelength of G is w _g , the wavelength of B is w _b, and the pixel values I _r , I _g , and I _{b of} R, G, and B are the estimated pixel values I in the near infrared region. The relationship with _ir is as follows.

In Equation (1), R, G, and B wavelengths W _r , W _g , and W _b are values of 440 nm, 550 nm, and 660 nm, for example. Also, when an actual color image is created, the R, G, and B wavelengths are not necessarily the above values. However, in a digital image taken in a general environment, the R, G, and B wavelengths are not included. Even if the value is assumed to be the above value, a pseudo near-infrared face image capable of obtaining sufficient matching accuracy can be created for the purpose of being used for face image matching.

In step S805, the values of the wavelengths w _ir , W _r , W _g , and W _b and the pixel values I _r , I _g , and I _b of the target pixel are input to Equation (1) and estimated. A pixel value I _ir is calculated.
In step S806, the estimated image creating means of the image estimating unit 104a stores the estimated pixel value I _ir calculated in step S805 as the pixel value of the corresponding pixel of the estimated image frame stored in advance. Then, by storing the pixel values of all the target pixels, the configuration of the estimated image of the near-infrared face image is completed.

  In step S807, it is determined whether or not all target pixels have been processed. If there is an unprocessed target pixel, the processes in steps S803 to S806 described above are repeatedly executed, and estimation is performed in step S808. The face image is stored in the buffer.

Returning to FIG. 4, when the estimated face image is buffered, near-infrared face image acquisition processing is executed in step S406.
In the near-infrared face image acquisition process S406, the near-infrared face image input unit 103 acquires a near-infrared face image of the person to be verified, which is captured using near-infrared light illumination (step S406).

Next, the collation unit 104b collates the estimated face image or registered face image acquired in step S404 or step S405 with the near-infrared face image acquired in step S406 (step S407).
That is, the matching unit 104b processes the estimated face image and the near-infrared face image, and extracts feature amounts such as the positions of the corners of the eyes, the eyes, the lips, the circumscribed rectangular areas of the eyes, the nose, and the mouth, and the shape of the face area. To do.

  In order to extract feature amounts, the position of the face is first detected from the image. The face position can be detected, for example, by preparing a plurality of templates for the entire face in advance, performing template matching while scanning the image using these templates, and detecting a place with the highest degree of matching. The template stores a plurality of face sizes, inclinations with respect to the image, and skin part contrast or pixel values with variations. Also, instead of preparing a template for the entire face, prepare templates for specific parts such as eyes, nose, and mouth, and scan the face image while maintaining the positional relationship between the templates of the specific parts. May be.

  When the position of the face in the image is detected, one or more feature amounts are extracted. At that time, for example, in order to detect the position of the corner of the eye as one of the feature amounts, template matching is performed by scanning the face range using a standard eye template, Determine the position of the template. Then, based on the template and the position of the template and the position of the corner of the eye in the template, the position of the corner of the corner can be obtained.

  When the feature amounts of the estimated face image and the near-infrared face image are extracted, the similarity between both images is calculated based on the feature amounts. In the case where the extracted feature amount is a position such as the corner of the eye, the head of the eye, or the lip end point, the position of the center of gravity is obtained from each position. Then, the positions of the feature amounts of either one of the images are arranged in parallel so that the calculated positions of the center of gravity coincide between the near-infrared face image as the input face image and the estimated face image or the registered face image. Moving. And the position of each feature-value is compared and each positional offset amount is calculated | required. An average value of positional deviation amounts for each feature amount is calculated as a similarity. Further, the obtained degree of displacement may be weighted by a part such as an eye or a mouth and a weighted average may be used as the similarity.

  Next, in step S408, based on the similarity obtained in step S407, it is determined whether the near-infrared face image that is the input face image and the estimated face image or the registered face image are face images of the same person. To do. That is, if the degree of similarity is equal to or higher than a predetermined threshold that can be determined to be a person determined by experiment / experience, the person is a collation OK, and if the degree of similarity is less than the threshold, the person is not a person, so a collation NG is determined. Here, in the case of collation NG, the fact that the person cannot be authenticated is displayed on the operation display unit 101, and the process ends. If the registered face image is not registered in the registered image storage unit 105 in step S402, that is, if the collation with the estimated face image is NG, the operation display unit 101 indicates that the collation has failed and / or the face Display that the image cannot be registered.

  If verification is OK in step S408, it is checked whether or not a registered face image corresponding to the ID number of the person to be verified is registered in the registered image storage unit 105. It registers with the registered image storage unit 105 in association with the ID number, and the process proceeds to step S411. On the other hand, if it has already been registered, the process proceeds to step S411 without registering the face image. If it is desired to sequentially update the registered face image, the input near-infrared face image may be updated and registered instead of the estimated face image regardless of whether or not it has already been registered.

  In step S411, an unlock signal is output to the electric lock 4 connected from the output unit 106, and this process ends. Based on the unlock signal, the electric lock 4 is unlocked for a certain period of time, and the person to be verified can enter the room through the entrance.

  As described above, in the first embodiment of the present invention, the estimated face image is stored in the registered image storage unit 105 in step S410, but the input face image input from the near infrared face image input unit 103 is used. May be registered. In such a case, since the face image acquired in the actual matching environment can be used as the registered face image rather than the estimated face image estimated from the color face image, more accurate face image matching can be performed in the matching after registration. It can be carried out.

  In the first embodiment, in the similarity determination in step S407, which is a collation process, the registered face image stored in the registered image storage unit 105 and the near-infrared face image acquired in step S406 are collated. Although the same threshold value is used in the case of collating the estimated face image configured in step S405 and the near-infrared face image acquired in step S406, different threshold values may be used for each.

  Furthermore, in the first embodiment, the face image matching device using the near-infrared face image has been described. However, the present invention is not limited to this. For example, a monochrome face image is created from a color face image and is converted into a monochrome face image. You may make it collate. In this case, the estimation method of the image estimation unit 104a estimates a monochrome face image. Such an estimation method may be appropriately employed according to the embodiment of the invention.

  Next, a second embodiment using the IC card 5 will be described with reference to FIGS. FIG. 3 is a block diagram of the face image matching device 11 in the second embodiment using the IC card 5. Since the basic configuration is the same as that of the first embodiment, blocks having the same functions as those of the first embodiment are denoted by the same reference numerals.

  Here, only differences from the first embodiment will be described. That is, the card R / W 112, which is a color face image acquisition unit, includes an antenna, a wireless transmission / reception interface, a driver, and the like, communicates with the IC card in a non-contact manner, and stores the ID number and color face stored in the IC card. Read the image and write the registered face image to the IC card. In the control unit 104, the face image registration unit 104c in the first embodiment registers the face image in the registered image storage unit 105. However, in the second embodiment, the face image registration unit 104d performs the card R. The registered face image is stored in the registered face image area 533 of the memory area 53 of the IC card 5 to be described later from / W112.

  FIG. 6 is a functional block diagram of the IC card 5. The IC card 5 includes a CPU, a ROM, a RAM, and the like, and includes a control unit 51 that controls various processes, a communication unit 52 that includes an antenna, a communication interface, and the like, and communicates with the face image matching device 11, and an EEPROM. The area includes a memory 53 having at least an ID number area 531, a color face image area 532, and a registered face image area 533.

Next, differences from the description of FIG. 4 in the first embodiment will be described.
First, when starting the processing of FIG. 4, at least an ID number is stored in the ID number area 531 and a color face image is stored in the color face image area 532 in the memory area of the IC card 5. It is assumed that the card is inserted in the card R / W 112 of the face image matching device 11.

  In step S 401, when an ID number is input on the operation display unit 101, the ID number is transmitted from the card R / W 112 to the IC card 5. The IC card 5 refers to the memory area 53 to determine whether or not the ID numbers match. If the ID numbers match, it determines whether or not the registered face image is stored in the registered face image area 533. . The IC card 5 returns the result from the communication unit 52 to the face image matching device 11. The face image matching device 11 determines whether or not the face image has been registered in step S402 based on the reply from the card R / W 112.

In step S409, it is determined whether or not the facial image has been registered by storing the reply signal acquired in step S401.
In step S410, the face image registration unit 104d stores the registered face image from the card R / W 112 in the registered face image area 533 of the memory area 53 of the IC card 5.

Since these points are the same as those in the first embodiment, description thereof is omitted.
In the face image collation device according to the third embodiment of the present invention, when a color face image is stored in advance in the IC card, the registered face image is registered not in the IC card but in the face image collation device.

  The face image matching device according to the third embodiment of the present invention is obtained by adding a registered image storage unit 105 in the face image matching device 1 of the first embodiment to the face image matching device 11 of the second embodiment, This is realized by replacing the image registration unit 104d with the face image registration unit 104c of the first embodiment. In this case, the card R / W 112 only needs to have at least a reading function, and the IC card may be stored so that a color face image can be read.

The face image collation apparatus according to the fourth embodiment of the present invention is a form in which a color face image is stored and managed by the central management apparatus 2 and the registered face image is registered in the IC card.
In the fourth embodiment, the communication unit 102 in the first embodiment is added to the face image collating apparatus 11 in the second embodiment, and the processes in steps S401, S402, and S409 in FIG. It can be realized by using. In this case, as long as the card R / W 112 has a writing function, it is sufficient that the registered face image can be written to the IC card.

  As described above, various modifications can be made within the scope of the present invention according to the embodiment to be implemented.

It is a whole system block diagram of the example which used the face image collation apparatus to which this invention is applied for the electric lock control of the entrance / exit. It is a functional block diagram of the face image collation device concerning a first embodiment of the present invention. It is a functional block diagram of the face image collation device concerning a second embodiment of the present invention using an IC card. It is an operation | movement flowchart of the face image collation apparatus which concerns on 1st embodiment of this invention. It is a figure which shows the content memorize | stored in the database of the centralized management apparatus used in the example which used the face image collation apparatus to which this invention is applied for the electric lock control of an entrance / exit. It is a functional block diagram of the IC card which concerns on 2nd embodiment of this invention. It is a graph which shows the reflective characteristic of a human skin. It is an operation | movement flowchart of an estimated image structure process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,11 Face image collation apparatus 101 Operation display part 102 Communication part 103 Near-infrared face image input part 104 Control part 104a Image estimation part 104b Collation part 104c, 104d Face image registration part 105 Registered image memory | storage part 106 Output part 112 Card R / W
2 Centralized Management Device 21 Database 3 Network 4 Electric Lock 5 IC Card 51 Control Unit 52 Communication Unit 53 Memory Area 531 ID Number Area 532 Color Face Image Area 533 Registered Face Image Area

Claims (7)

A face image matching device for matching a face image inputted by a person to be verified with a registered face image,
A face image input unit for photographing a person to be verified under a predetermined environment and inputting a face image;
A color face image acquisition unit for acquiring a color face image of the person to be verified;
An estimation unit that calculates an estimated face image that can be collated with a face image captured under the predetermined environment based on the color face image;
A collation unit that collates the input face image acquired by the face image input unit and the estimated face image to determine whether both face images are of the same person;
A registered image storage unit for storing the registered face image;
If the collation unit determines that the person is the same, the estimated face image is registered as a registered face image in the registered image storage unit, and if the collation unit does not determine that the person is the same, a face image registration unit that is not registered;
A face image collating apparatus comprising: A face image matching device for matching a face image inputted by a person to be verified with a registered face image,
A face image input unit for photographing a person to be verified under a predetermined environment and inputting a face image;
A color face image acquisition unit for acquiring a color face image of the person to be verified;
An estimation unit that calculates an estimated face image that can be collated with a face image captured under the predetermined environment based on the color face image;
A collation unit that collates the input face image acquired by the face image input unit and the estimated face image to determine whether both face images are of the same person;
A registered image storage unit for storing the registered face image;
A face image registration unit that registers the input face image as a registered face image in the registered image storage unit when the collation unit determines that the person is the same person, and does not register unless the collation unit determines the same person;
A face image collating apparatus comprising: The registered image storage unit is provided in a portable information storage medium,
If the face image registration unit determines that the matching person is the same person, the face image registration unit registers the estimated face image or the input face image as a registered face image in the information storage medium, and the matching unit does not determine the same person. The face image collation apparatus according to claim 1 or 2, wherein registration is not performed.   The face image collation device according to any one of claims 1 to 3, wherein the color face image acquisition unit further includes a network communication unit, and acquires a color face image of the person to be verified through the network communication unit.   The face image according to any one of claims 1 to 3, wherein the color face image acquisition unit further includes data reading means from an information storage medium, and acquires a color face image of the person to be verified from the information storage medium. Verification device. The estimation unit includes
Pixel value extraction means for extracting, for each pixel, color gradation information possessed by pixel values of the pixels of the color face image;
For each pixel, based on the correlation between the color gradation information based on the skin color reflection characteristics and the pixel value in the near infrared region, and the color gradation information extracted by the pixel value extraction unit, the estimation in the wavelength of the near infrared region An estimation means for calculating a pixel value;
The face image collation apparatus according to claim 1, further comprising an estimated image creating unit that configures the estimated face image from pixels having estimated pixel values estimated by the estimating unit. A color face image acquisition step of acquiring a color face image of the person to be verified;
An estimation step of calculating an estimated face image that can be collated with a face image shot under a predetermined environment based on the color face image;
A face image input step of photographing a person to be verified under a predetermined environment and inputting a face image;
A collation step of collating the input face image acquired in the face image input step with the estimated face image to determine whether both face images are of the same person;
Registering the estimated face image as a registered face image when determined as the same person in the matching step, and a face image registration step not registering unless determined as the same person;
A face image registration method for collation characterized by comprising:

Images