JP2004341992A - Image photographing device and image collation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To shorten processing time for collation and to improve the accuracy of collation in an image photographing device and an image collation device provided with an interface for easily collating a photographed image for collation with a registered image. <P>SOLUTION: Images photographed by the image photographing device provided with a photographing part 101 for photographing an object, a feature point detection part 102 for extracting feature points of the photographed image, an image synthesis part 103 for synthesizing the image with the feature points, and an image display part 104 for displaying the synthesized image and suited to collation are previously registered in an image database 304, a registered image of an object to be collated which is stored in the image database 304 is synthesized with the current image of the object and the synthesized image is displayed. Since an image of a position, size and posture which are almost the same as the registered image and suited to collation can be obtained in collation, the processing time of collation in a collation part 701 can be shortened and the accuracy of collation can be improved. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image capturing apparatus and an image collating apparatus including an interface for facilitating collation with a registered image when capturing an image for collation.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, as an image collating apparatus, for example, there has been an apparatus described in Patent Document 1. FIG. 15 shows a block configuration of a conventional image matching device described in Patent Document 1.
[0003]
In FIG. 15, an input unit 1501 is for inputting biological information, a display unit 1502 is for displaying the input biological information, and a frame display unit 1503 is for inputting a frame for specifying the size and position of the biological information. The authentication means 1504 authenticates a pre-registered user based on the input biometric information.
[0004]
As another image matching device, there is one described in Patent Document 2. FIG. 16 shows a block configuration of a conventional image matching device described in Patent Literature 2. As shown in FIG.
[0005]
In FIG. 16, an image input unit 1610 inputs an image, a face area detection unit 1620 detects an area of a face image from an input image, and a feature point detection unit 1630 detects an image based on the detected face image. The normalization unit 1640 moves the position of the detected feature point and deforms the image so that the image comes to a preset position. A person is identified and recognized using a normalized image and a database image of each registrant registered in advance.
[0006]
[Patent Document 1]
JP-A-2002-229955 (page 3-5, FIG. 1)
[Patent Document 2]
JP-A-2000-67237 (pages 2, 3; FIG. 1)
[0007]
[Problems to be solved by the invention]
However, in the configuration of Patent Document 1, since the same frame is displayed uniformly regardless of the individual difference of the living body to be compared, when photographing a living body that does not conform to the frame standard, the position and the posture may be adjusted. It was difficult. For example, if the position of the face is specified by an elliptical frame of a certain size, and if the face does not match the size of the ellipse, the face direction, position, and size are likely to differ between registration and verification. However, there is a possibility that the collation accuracy is reduced.
[0008]
Further, the configuration of Patent Literature 2 has a problem that a lot of load is applied to a process for moving the position of a detected feature point or deforming an image, and it takes a long time to perform a matching process. .
[0009]
An object of the present invention is to solve the above-mentioned conventional problems, to provide an interface for facilitating collation with a registered image, and to improve processing time for collation and precision of collation.
[0010]
[Means for Solving the Problems]
An image capturing apparatus according to the present invention that solves the above-described conventional problems includes a capturing unit that outputs image data of an image of an object, a feature point detecting unit that extracts a feature point that is a feature of image data of the captured object, , An image synthesizing unit that outputs synthesized image data obtained by synthesizing image data and feature points, and an image display unit that displays the synthesized image data.
[0011]
This allows the user to recognize whether or not the object currently being photographed is image data that can clearly detect feature points, so that the direction of the face or the like can be changed or the level of light can be adjusted accordingly. It becomes possible to do.
[0012]
Further, an image collating apparatus according to the present invention includes an image photographing apparatus according to the present invention, an image database storing image data of a collation target object photographed in advance in association with identification information, And a collating unit for collating the image data extracted by the registered data management unit with the image data extracted based on the identification information, and an output unit for outputting the collation result of the collating unit.
[0013]
Accordingly, the present invention provides an interface that combines and displays image data of a registered object and image data of an object being photographed, and allows the object to be photographed in a similar size, position, and orientation as the registered image data. Thereby, the user can photograph the object in a state suitable for the collation. In addition, the position of the object at the time of collation is easily adjusted, and the processing speed and collation accuracy are improved.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
An image capturing apparatus according to the present invention includes a capturing unit that outputs image data of a captured object, a feature point detecting unit that extracts a feature point that is a feature of the captured object from the image data, An image combining unit that outputs combined image data obtained by combining points and an image display unit that displays the combined image data. Thus, when photographing, the photographing can be performed while confirming whether or not the feature points of the object have been clearly detected, so that it is easy to photograph the object in a state suitable for collation.
[0015]
Further, an image photographing apparatus according to the present invention includes a photographing unit that outputs image data of a photographed object, a feature point detecting unit that extracts a characteristic point that is a characteristic of the photographed object from the image data, An image synthesizing unit that outputs synthesized image data obtained by synthesizing the image data of the object to be checked and the characteristic points and the image of the object to be checked, and an image display unit that displays the synthesized image data.
[0016]
Further, the image photographing apparatus according to the present invention includes: an image database storing image data of a target object to be collated photographed in advance in association with identification information; an information input unit for inputting the identification information; A registration data management unit that extracts image data based on the identification information, wherein the image synthesis unit stores the image data of the currently captured object, the feature points of the image data, and the identification information from the information input unit. And outputting composite image data obtained by combining the extracted image data with the extracted image data.
[0017]
Thus, when shooting, the user can shoot while confirming whether or not the feature points of the subject have been clearly detected, and also provide information that can shoot the subject with the same size, position, and orientation as at the time of registration. Therefore, it is possible to capture an image that improves the accuracy of the matching.
[0018]
Further, the image photographing apparatus according to the present invention inputs a feature amount database that stores feature data, which is a feature of image data of an object to be checked photographed in advance, in association with identification information, and the identification information. An information input unit; and a registered data management unit that extracts feature data from the feature amount database based on the identification information, wherein the image combining unit includes a currently captured image, a feature point of the image, and an information input unit. And the feature data extracted on the basis of the identification information.
[0019]
This allows the user to perform shooting while checking whether or not feature points of the subject have been clearly detected when shooting, and to shoot the subject in the same size, position, and orientation as at the time of registration. Only provided. Therefore, it is possible to capture an image that improves the accuracy of the matching, and it is possible to obtain a similar effect by storing a small amount of data as compared with storing image data. Become.
[0020]
The characteristic data of the image photographing apparatus according to the present invention is at least one of the shape and position of the contour of the face, the position of the eyes, the position of the nose, and the position of the mouth. Thus, the same effect can be obtained only by storing a smaller amount of data.
[0021]
The image synthesizing unit of the image photographing apparatus according to the present invention includes a transmission processing unit that performs a transmission process on an image and a synthesis unit that performs a synthesis process. Thus, if the registered image is subjected to the transparency processing, comparison with the current captured image becomes easier.
[0022]
An image collating device according to the present invention includes an image photographing device according to the present invention, a collating unit for collating a currently photographed image with an image extracted based on identification information by a registration data managing unit, and a collating unit. And an output unit for outputting the collation result.
[0023]
Further, the image matching device according to the present invention is an image capturing device according to the present invention, an image database that stores a previously captured image to be matched in association with identification information, and a currently captured image, A collation unit for collating the image extracted based on the identification information by the registration data management unit is provided, and an output unit for outputting a collation result of the collation unit.
[0024]
As a result, since the matching can be performed using the image suitable for the matching, the processing time required for the position adjustment of the object at the time of the matching can be shortened, and the accuracy of the matching can be improved.
[0025]
The image synthesizing unit of the image matching device according to the present invention includes a transmission processing unit that performs a transmission process on an image and a synthesis unit that performs a synthesis process.
[0026]
In addition, the registration data management unit of the image matching device according to the present invention stores the currently captured image in the image database in association with the identification information from the information input unit. Thereby, a new image to be collated can be added to the database while collation is being performed, so that the number of collation targets is increased and the accuracy is improved.
[0027]
Hereinafter, embodiments of the present invention will be described with reference to the drawings. The same components are denoted by the same reference numerals.
[0028]
(Embodiment 1)
FIG. 1 is a block diagram showing the configuration of an image photographing apparatus according to Embodiment 1 of the present invention.
[0029]
In FIG. 1, a photographing unit 101 is a camera for photographing an object such as a human face to be collated, and a feature point detecting unit 102 detects a feature point which is a feature of the image data captured by the photographing unit 101. The image synthesizing unit 103 synthesizes an index indicating a feature point on an image, and the image display unit 104 is a display for displaying an image.
[0030]
The feature point detection unit 102 previously stores templates of the eyes, nose, and mouth. When an input image is provided, the entire image is scanned, and based on a predetermined similarity, the most similar template is obtained. The feature points are detected by finding the local area position.
[0031]
The image synthesizing unit 103 synthesizes the input image and an index indicating the position of the detected feature point by performing an AND process on an internal memory.
[0032]
The operation and operation of the image capturing apparatus configured as described above will be described below. FIG. 2 is a flowchart showing the operation of the image photographing apparatus.
[0033]
First, the photographing unit 101 photographs a subject (step S201). FIG. 11A is the photographed image.
[0034]
Next, the feature point detection unit 102 detects feature points from the image captured by the imaging unit 101 (step S202).
[0035]
Next, the image synthesizing unit 103 synthesizes the image captured by the photographing unit 101 and the index indicating the feature point detected by the feature point detection unit 102, and outputs the synthesized image to the image display unit 104 (step S203). FIG. 11B is a composite image in which indices indicating characteristic points are displayed with crosses and superimposed on the captured image.
[0036]
Next, the image display unit 104 receives this and displays the composite image on the display (step S204).
[0037]
As described above, in the present embodiment, the captured image is combined with the detected feature points of the image and displayed. This allows the user to recognize the current state of the image being captured, and by changing the direction of the face or the like or adjusting the amount of light in accordance with that, an image that can clearly detect feature points can be obtained. It becomes possible to shoot. That is, originally, in FIG. 11B, a feature point must appear at the mouth, but it cannot be detected in this shooting state. In order to improve this, the subject can change the direction of the face, adjust the lighting condition, make the feature points of the mouth appear, and take an image with all the feature points displayed It will be easier.
[0038]
The display of the composite image is preferably performed in real time with the shooting, but it is also possible to detect the feature points and perform the composite display only when the user gives a shooting instruction. In this case, the load on the image capturing device can be reduced.
[0039]
(Embodiment 2)
FIG. 3 shows a block diagram of an image photographing apparatus according to Embodiment 2 of the present invention.
[0040]
In FIG. 3, an image synthesis unit 301 includes a synthesis unit 302 and a transmission processing unit 303. The transmission processing unit 303 performs a transmission process on an input image. The synthesis unit 302 ANDs a plurality of images on an internal memory. Processing and synthesis.
[0041]
The image database 304 is a hard disk in which an image of an object to be collated taken in advance and identification information for identifying the image are stored in association with each other, and the information input unit 305 specifies the identification information by the user. It is for doing.
[0042]
Then, the registered data management unit 306 extracts a corresponding image from the image database 304 based on the input identification information.
[0043]
The other imaging unit 101, feature point detection unit 102, and image display unit 104 are the same as those in the first embodiment.
[0044]
The operation and operation of the image capturing apparatus configured as described above will be described below. FIG. 4 is a flowchart showing the operation of the image photographing apparatus. FIGS. 12A to 12D show a current captured image, a registered image, and a composite image thereof in the present embodiment.
[0045]
First, the user inputs identification information from the information input unit 305 (step S401).
[0046]
Next, the registered data management unit 306 extracts an image from the image database 304 based on the identification information (Step S402). FIG. 12A shows the extracted registered image.
[0047]
Next, the transmission processing unit 303 performs transmission processing on the extracted image, and writes the image in the internal memory (step S403). FIG. 12B shows the registered image subjected to the transmission processing.
[0048]
Next, a subject is fetched from the photographing unit 101 and stored in the internal memory (step S404). FIG. 12C shows this photographed image.
[0049]
Next, the feature point detection unit 102 detects feature points from the captured image (step S405).
[0050]
Next, the image synthesizing unit 301 synthesizes the transmission image stored in the internal memory, the image fetched from the imaging unit 101, and the characteristic points, and outputs the synthesized image to the image display unit 104 (step S406). FIG. 12D shows this composite image.
[0051]
Next, the image display unit 104 receives this and displays the composite image on the display (step S407).
[0052]
As described above, in the present embodiment, the image of the subject and the detected feature points of the detected image are combined with the image of the collation target registered in advance by the user and displayed. This makes it possible to recognize the current situation of the image being photographed and to make it easier for the user to adjust the face, which is the subject, to a specified position. Or by adjusting the amount of light, it becomes possible to capture an image at a specified position where a feature point can be clearly detected.
[0053]
Since this alignment displays the image of the subject itself acquired in the past, the contour of the face and the position of each part such as eyes, nose, mouth, etc. are displayed with the conventional uniform frame and index. It will be much easier than to adapt to it.
[0054]
Further, it is more preferable that the registered image is a monochrome transparent image so that the registered image and the image of the subject can be easily distinguished.
[0055]
The display of the composite image is preferably performed in real time with the shooting, but it is also possible to detect the feature points and perform the composite display only when the user gives a shooting instruction. In this case, the load on the image capturing device can be reduced.
[0056]
Further, in the present embodiment, the transmission processing unit 303 transmits and synthesizes the registered image. However, the present invention is not limited to this, and the registered image and the currently captured image can be alternately displayed. In this case, the transmission processing unit 303 becomes unnecessary.
[0057]
(Embodiment 3)
FIG. 5 shows a block diagram of an image photographing apparatus according to Embodiment 3 of the present invention.
[0058]
In FIG. 5, a feature amount database 501 is a hard disk in which feature data detected in advance from an image of an object to be collated and identification information for identifying the image are stored in association with each other. This feature data is information on the shape and position of the contour of the face, and the position of each part such as eyes, nose, and mouth.
[0059]
The other components of the photographing unit 101, the feature point detecting unit 102, the image synthesizing unit 103, the image display unit 104, the information input unit 305, and the registered data management unit 306 are the same as those in the first and second embodiments. .
[0060]
The operation and operation of the image capturing apparatus configured as described above will be described below. FIG. 6 is a flowchart showing the operation of the image photographing apparatus. FIGS. 13A to 13D show a registered image, a captured image, and a composite image thereof in the present embodiment.
[0061]
First, the user inputs identification information from the information input unit 305 (step S601).
[0062]
Next, the registration data management unit 306 extracts feature data from the feature amount database 501 based on the identification information (step S602). In the present embodiment, the shape and position of the contour of the face and the feature points are used as the feature data. FIG. 13A shows a registered image, and FIG. 13B shows registered outlines and feature points (x marks) of the image.
[0063]
Next, a subject is fetched from the photographing unit 101 and stored in the internal memory (step S603). FIG. 13C shows a current captured image.
[0064]
Next, the feature point detection unit 102 detects feature points from the captured image (step S604).
[0065]
Next, the image synthesizing unit 103 synthesizes the index indicating the position of the feature point and the feature data extracted from the feature amount database 501 on the image stored in the internal memory, and outputs the synthesized image to the image display unit 104. (Step S605). At this time, it is more preferable to change each color, shape, size, and the like in order to distinguish the index indicating the feature data from the index indicating the feature point of the currently captured image. FIG. 13D shows a composite image, in which characteristic points detected from the current photographed image are indicated by circles.
[0066]
Next, the image display unit 104 receives this and displays the composite image on the display (step S606).
[0067]
As described above, in the present embodiment, the contours and feature points of the face, which are the feature data to be collated, which are specified by the user, the subject image, and the feature points of the detected subject image are synthesized. Display, the user can know the deviation between the position of the contour of the registered image and the subject, and the degree of detection of the characteristic points of the captured image. Therefore, it is easy for the user to move the position of the face or the like so as to match the frame of the outline, change the direction, or adjust the degree of light. Can be taken.
[0068]
Since this alignment displays the image of the subject itself acquired in the past, the contour of the face and the position of each part such as eyes, nose, mouth, etc. are displayed with the conventional uniform frame and index. There is a feature that it can be done much easier than it has been.
[0069]
The display of the composite image is preferably performed in real time with the shooting, but it is also possible to detect the feature points and perform the composite display only when the user gives a shooting instruction. In this case, the load on the image capturing device can be reduced.
[0070]
Further, in the present embodiment, both the contour and the feature point are used as the feature data, but the present invention is not limited to this, and it is possible to use only one of them.
[0071]
(Embodiment 4)
FIG. 7 shows a block configuration diagram of an image matching device according to Embodiment 4 of the present invention.
[0072]
7, a matching unit 701 matches an image input from the photographing unit 101 with a registered image input from the registered data management unit 704, and an output unit 702 outputs the matching result of the matching unit 701 to a display or a speaker. The collation instructing unit 703 instructs collation with the image currently displayed by the user, and the registration data management unit 704 registers the input image in the image database 304 in association with the identification information. And extracting a registered image based on the identification information. The other imaging unit 101, feature point detection unit 102, image display unit 104, image synthesis unit 301, image database 304, and information input unit 305 are the same as those in the first and second embodiments.
[0073]
The operation of the image collating device configured as described above will be described below. FIG. 8A is a flowchart illustrating an image registration operation according to the present embodiment. FIG. 12 is an example of an image displayed when a collation image is captured.
[0074]
Before the collation, the target image is stored as a registered image. Specifically, first, the registration data management unit 306 acquires an image captured by the imaging unit 101 by the processing described in the first embodiment (step S801).
[0075]
Next, the registration data management unit 306 acquires identification information such as the name of the subject input by the user from the information input unit 305 (step S802).
[0076]
Next, the registered data management unit 306 stores the image input in step S801 as a registered image in the image database 304 together with the identification information input in step S802 (step S803).
[0077]
The above processing is performed for all objects to be collated. It is preferable that a plurality of registered images be registered for one collation target object. This is because, in the collation operation described below, the more registered images to be collated, the more accurate the collation can be performed.
[0078]
Next, a collation method of a collation target will be specifically described below.
[0079]
FIG. 8B is a flowchart illustrating the collation operation according to the present embodiment.
[0080]
First, the user inputs the name of the person to be verified from the information input unit 305 (step S804).
[0081]
Next, the registered data management unit 704 extracts a registered image corresponding to the name from the image database 304 and outputs the registered image to the image combining unit 301. At the same time, the photographing unit 101 starts photographing and sends the current image of the subject to the image synthesizing unit 301 (step S805). FIG. 12A shows an extracted registered image, and FIG. 12C shows a current captured image.
[0082]
Next, the image synthesizing unit 301 performs the transmission processing on the registered image by the processing described in Embodiment 2, and then synthesizes the registered image with the captured image. Then, the synthesized image is output to the image display unit 104 and displayed (step S806). FIG. 12B is an image obtained by performing a transmission process on the registered image, and FIG. 12D is a composite image of the image subjected to the transmission process and the captured image.
[0083]
Next, the user compares the captured image displayed on the screen of the image display unit 104 with the registered image, and detects all of the predetermined feature points. The image has the same position and the same size as the registered image. When the photographed image is displayed, a collation instruction is issued to the collation instruction unit 703 (step S807). The matching unit 701 matches the matching face image designated by the user with the registered image (step S808), and outputs to the output unit 702 whether the matching face image is the same person as the registered image face (step S808). S809).
[0084]
Here, the matching process performed by the matching unit 701 will be described.
[0085]
FIG. 14A shows a current captured image, and FIG. 14B shows a registered image.
[0086]
.., A21,..., A21, a22,..., And the shade values of each pixel of the registered image are b11, b12, b13,. , ..., the distance d between the captured image and the registered image is represented by (Equation 1).
[0087]
d = (a11−b11) × (a11−b11) + (a12−b12) × (a12−b12) + (1)
The smaller the value of the distance d, the more similar the two images are.
[0088]
The collation unit 701 performs the calculation of the distance for all the registered images registered in the image database 304 and corresponding to the identification information. If the value of the smallest distance is less than or equal to the predetermined value, a success is output in step S807.
[0089]
Another matching processing method is a template matching method using normalized correlation.
[0090]
The average of the gray value of each pixel in the captured image is m1, the average of the gray value of each pixel in the registered image is m2, and the vector amount A and the vector amount B are obtained from (Expression 2) and (Expression 3).
[0091]
Vector A = (a11-m1, a12-m1, a13-m1,...) (2)
Vector B = (b11-m2, b12-m2, b13-m2,...) (3)
At this time, the normalized correlation C between the captured image and the registered image is represented by (Equation 4).
[0092]
C = A · B ÷ (| A || B |) (4)
Here, AB represents the inner product of A and B, and | A | and | B | represent the magnitude of the vector.
[0093]
The larger the value of the normalized correlation C, the more similar the captured image and the registered image are.
[0094]
The collation unit 701 performs the calculation of the normalized correlation C for all the registered images registered in the image database 304 and corresponding to the identification information. When the largest value of the normalized correlation C is equal to or larger than a predetermined value, a success is output in step S809.
[0095]
As described above, in the present embodiment, the registered image subjected to the transmission processing, the image input from the imaging unit 101, and the index indicating the feature point detected for the image are synthesized by the image synthesis unit 301, By displaying it, it becomes easy for the user to take a picture with almost the same size, position, and orientation as at the time of registration. This makes it possible to obtain an image suitable for collation.
[0096]
In addition, since the captured image requires only a small amount of alignment for collation, the processing time for adjusting the position and size can be reduced, and the accuracy of collation can be improved, and the time required for collation can be shortened. Becomes possible.
[0097]
(Embodiment 5)
FIG. 9 shows a block diagram of an image matching apparatus according to Embodiment 5 of the present invention.
[0098]
9, a registered data management unit 901 stores a captured image as a registered image in an image database 304 together with identification information input from an information input unit 305, and identifies coordinates of feature points extracted from the registered image. The information is stored in the feature database 501 together with the information, or conversely, the coordinates of the registered image and the feature points are extracted from the image database 304 and the feature database 501 based on the identification information. In addition, this image matching apparatus differs from the fourth embodiment in that the image synthesizing unit 103 does not include the transmission processing unit 303 and that the image synthesizing unit 103 includes the feature amount database 501.
[0099]
The operation of the image collating apparatus configured as described above will be described below. FIG. 10A is a flowchart illustrating an image registration operation according to the present embodiment. FIGS. 13A to 13D show a registered image, an index indicating a feature point extracted from the registered image, an image captured at the time of capturing the collation image, and a composite image obtained by superimposing the captured image and the feature point of the registered image. is there.
[0100]
As in the fourth embodiment, before matching, the image to be matched is stored as a registered image. Specifically, first, the image captured by the capturing unit 101 by the processing described in the first embodiment is transmitted to the registration data management unit 901 and the feature point detection unit 102 (step S1001).
[0101]
Next, the feature point detection unit 102 detects feature points and contours of the image and sends them to the registered data management unit 901 (step S1002).
[0102]
Next, the registration data management unit 901 receives identification information such as the name of the subject input by the user from the information input unit 305 (step S1003).
[0103]
Next, the registration data management unit 306 stores the image received from the imaging unit 101 as a registration image in the image database 304 together with the identification information from the information input unit 305 (step S1004).
[0104]
Next, the registration data management unit 901 stores the feature points and contours received from the feature point detection unit 102 in the feature amount database 501 together with the identification information (step S1005).
[0105]
The above processing is performed for all objects to be collated. It is preferable that a plurality of registered images be registered for one collation target object. This is because, in the collation operation described below, the more registered images to be collated, the more accurate the collation can be performed.
[0106]
Next, a collation method of a collation target will be specifically described below.
[0107]
FIG. 10B is a flowchart illustrating the matching process according to the present embodiment.
[0108]
First, the user inputs the name of the person to be verified from the information input unit 305 (step S1006).
[0109]
Next, the registration data management unit 901 extracts the coordinates of the feature point of the registered image corresponding to the name from the feature amount database 501 and outputs the coordinates to the image synthesis unit 103. At the same time, the photographing unit 101 starts photographing and sends the current image of the subject to the image synthesizing unit 103 (step S1007).
[0110]
Next, as in the third embodiment, the image combining unit 103 outputs the coordinates of the feature points of the registered image input from the registered data management unit 901 to the image display unit 104 by superimposing the coordinates on the image input from the image capturing unit 101. (Step S1008).
[0111]
Next, the user compares the position of the feature point in the registered image with the eye, nose, and mouth in the captured image while comparing the position of the feature point with the captured image displayed on the screen of the image display unit 104. Then, the photographed image is designated as a collation image (step S1009).
[0112]
The next process (step S1010) of the collation unit 701 and the process of the collation result (step S1011) are the same as steps S806 and S807 described in the fourth embodiment.
[0113]
In the present embodiment, both the feature points and the outline of the face are used, but only one of the feature points and the outline may be used. In addition, it is preferable to change the color, shape, and size of the indicator indicating the feature point in order to distinguish between the registered image and the current image.
[0114]
Thus, in the present embodiment, the image input from the imaging unit 101, the feature points detected from the image, and the feature points extracted from the registered image are displayed by being superimposed on the image display unit 104. It becomes easy for the user to take a picture with substantially the same size, position, and orientation as at the time of registration. This makes it possible to obtain an image suitable for collation.
[0115]
In addition, since the captured image requires only a small amount of alignment for collation, the processing time for adjusting the position and size can be reduced, and the accuracy of collation can be improved, and the time required for collation can be shortened. Becomes possible.
[0116]
【The invention's effect】
Advantageous Effects of Invention According to the present invention, it is possible to capture an image suitable for comparison in which feature points are clearly shown. Also, at the time of collation, it is possible to acquire an image suitable for collation with substantially the same position, size, and orientation as at the time of registration. As a result, the processing time for collation can be reduced, and the accuracy of collation can be improved.
[Brief description of the drawings]
FIG. 1 is a block diagram of an image photographing apparatus according to a first embodiment of the present invention;
FIG. 2 is a flowchart showing processing of the image photographing apparatus according to the first embodiment of the present invention;
FIG. 3 is a block diagram of an image photographing apparatus according to a second embodiment of the present invention;
FIG. 4 is a flowchart showing processing of the image photographing apparatus according to the second embodiment of the present invention;
FIG. 5 is a block diagram of an image photographing apparatus according to a third embodiment of the present invention.
FIG. 6 is a flowchart showing processing of the image photographing apparatus according to the third embodiment of the present invention;
FIG. 7 is a block diagram of an image matching device according to a fourth embodiment of the present invention.
FIG. 8A is a flowchart showing an image registration operation according to the fourth embodiment of the present invention;
(B) Flowchart showing operation of image collation in Embodiment 4 of the present invention
FIG. 9 is a block diagram of an image matching device according to a fifth embodiment of the present invention.
10A is a flowchart illustrating an image registration operation according to the fifth embodiment of the present invention. FIG. 10B is a flowchart illustrating an image collating operation according to the fifth embodiment of the present invention.
FIG. 11A is a diagram showing a captured image according to the first embodiment of the present invention;
(B) The figure which shows the synthetic | combination image in Embodiment 1 of this invention.
FIG. 12A shows a registered image according to the second embodiment of the present invention.
(B) A diagram showing an image after the transmission processing according to the second embodiment of the present invention.
(C) The figure which shows the captured image in Embodiment 2 of this invention.
(D) The figure which shows the synthetic | combination image in Embodiment 2 of this invention.
FIG. 13A shows a registered image according to the third embodiment of the present invention.
(B) A diagram showing characteristic points according to the second embodiment of the present invention.
(C) The figure which shows the captured image in Embodiment 2 of this invention.
(D) The figure which shows the synthetic | combination image in Embodiment 2 of this invention.
FIG. 14 (a) is a diagram showing a captured image according to Embodiment 4 of the present invention.
(B) The figure which shows the registration image in Embodiment 4 of this invention.
FIG. 15 is a block diagram showing a conventional image matching apparatus.
FIG. 16 is a block diagram showing a conventional image matching apparatus.
[Explanation of symbols]
101 shooting unit
102 Feature point detection unit
103, 301 Image synthesis unit
104 Image display unit
302 synthesis unit
303 Transmission unit
304 Image Database
305 Information input section
306, 704, 901 Registration data management unit
501 Feature database
701 Collation unit
702 output unit
703 Collation instruction section
1501 Input means
1502 Display means
1503 Frame display means
1504 Authentication means
1505 Biological information confirmation means
1610 Image input unit
1620 face area detection unit
1630 feature point detection unit
1640 Normalization unit
1650 Authentication Department

Claims (10)

A photographing unit for outputting image data of an object,
A feature point detection unit that extracts feature points that are features of the captured object from the image data,
An image synthesis unit that outputs synthesized image data obtained by synthesizing the image data and the feature points,
An image capturing apparatus, comprising: an image display unit that displays the composite image data. A photographing unit for outputting image data of an object,
A feature point detection unit that extracts feature points that are features of the captured object from the image data,
An image synthesizing unit that outputs synthesized image data obtained by synthesizing the image data, the feature points, and image data of the object to be collated that has been shot in advance;
An image capturing apparatus, comprising: an image display unit that displays the composite image data. An image database that stores image data of a target object to be collated taken in advance in association with identification information,
An information input unit for inputting the identification information;
A registration data management unit for extracting image data from the image database based on the identification information,
The image synthesizing unit outputs synthesized image data obtained by synthesizing image data of a currently captured object, feature points of the image data, and image data extracted based on the identification information from the information input unit. The image photographing device according to claim 1. A feature amount database that stores feature data, which is a feature of the image data of the object to be matched photographed in advance, in association with the identification information,
An information input unit for inputting the identification information;
A registration data management unit that extracts the feature data from the feature amount database based on the identification information,
2. The image capturing apparatus according to claim 1, wherein the image combining unit combines the currently captured image, a feature point of the image, and the feature data extracted based on the identification information from the information input unit. apparatus. The image photographing apparatus according to claim 4, wherein the feature data is at least one of a shape and position of a face contour, an eye position, a nose position, and a mouth position. The image capturing apparatus according to claim 2, wherein the image combining unit includes a transmission processing unit that performs a transmission process on an image and a combination unit that performs a combination process. An image photographing apparatus according to claim 3,
An image matching apparatus, comprising: a matching unit that matches the currently captured image with an image extracted based on the identification information by the registration data management unit; and an output unit that outputs a matching result of the matching unit. An image photographing apparatus according to claim 4 or 5,
An image database that stores the image of the object to be collated taken in advance in association with the identification information,
A collation unit that collates the currently captured image with an image extracted based on the identification information by the registration data management unit,
An image collating apparatus comprising: an output unit that outputs a collation result of the collation unit. The image matching device according to claim 7, wherein the image combining unit includes a transmission processing unit that performs a transmission process on the image and a combining unit that performs the combining process. 9. The image matching device according to claim 7, wherein the registration data management unit stores the currently captured image in the image database in association with identification information from the information input unit.

Images