JP2015035070A - Image collation device, image collation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately and properly collate images.SOLUTION: A portable terminal 100 includes: a corresponding area determination part 7d which determines a corresponding area corresponding to a registered image in a pickup image based on pixel information on the registered image registered in an image registration part 6 and acquired pixel information on the pickup image; a collation object area setting part 7f which sets a collation object area in the determined corresponding area; and an image collation part 7g which performs collation based on the result of correspondence between feature information on the set collation object area and feature information on the area corresponding to the collation object area in a registered image.

Description

  The present invention relates to an image collation apparatus, an image collation method, and a program.

  Conventionally, a method for comparing feature amounts of feature points when matching images has been proposed (see, for example, Patent Document 1). Specifically, for example, when an image captured by a camera and an image registered in a database are collated, feature points are matched using local feature amounts to calculate a similarity degree.

JP 2010-250529 A

  By the way, if the image to be collated is composed of relatively simple textures, there are a plurality of similar textures in the image, so that miscorrespondence between feature points occurs and images are properly collated. There is a problem that can not be.

  The present invention has been made in view of such a problem, and an object of the present invention is to provide an image collation apparatus, an image collation method, and a program capable of accurately collating images with high accuracy. is there.

In order to solve the above-described problem, an image collation apparatus according to the present invention includes:
Registration means for registering at least one image in advance, acquisition means for acquiring a captured image obtained by capturing a specific subject, pixel information of a registered image registered in the registration means, and imaging acquired by the acquisition means Based on pixel information of the image, specifying means for specifying a corresponding area corresponding to the registered image in the captured image, setting means for setting a matching target area in the corresponding area specified by the specifying means, The feature information of the comparison target region set by the setting unit is associated with the feature information of the region corresponding to the comparison target region in the registered image, and based on the result of the association, It is characterized by comprising collation means for collating with a registered image.

In addition, an image matching method according to the present invention includes:
An image matching method using an image matching device, a process of registering at least one image in advance, a process of acquiring a captured image obtained by capturing a specific subject, and pixel information and acquisition of a registered image registered Based on the pixel information of the captured image, a process for specifying a corresponding area corresponding to the registered image in the captured image, a process for setting a matching target area in the specified corresponding area, and A process of associating the feature information of the collation target area with the feature information of the area corresponding to the collation target area in the registered image, and collating the collation target area with the registered image based on the result of the correlation It is characterized by including.

The program according to the present invention is
Registering means for registering at least one image in advance, computer for acquiring a captured image in which a specific subject is imaged, pixel information of registered image registered in said registration means, and said acquisition means Based on the pixel information of the captured image acquired by the above, a specifying unit that specifies a corresponding region corresponding to the registered image in the captured image, and a matching target region is set in the corresponding region specified by the specifying unit A setting unit that associates the feature information of the region to be collated set by the setting unit with the feature information of the region corresponding to the region to be collated in the registered image, and based on the result of the association, It is characterized by functioning as a collating means for collating the region with the registered image.

  According to the present invention, it is possible to accurately collate images with high accuracy.

It is a block diagram which shows schematic structure of the portable terminal of Embodiment 1 to which this invention is applied. 3 is a flowchart illustrating an example of an operation related to an image registration process by the mobile terminal of FIG. 1. It is a figure which shows typically an example of the image which concerns on the image registration process of FIG. 3 is a flowchart illustrating an example of an operation related to an image matching process performed by the mobile terminal in FIG. 1. It is a figure which shows typically an example of the image which concerns on the image collation process of FIG. It is a figure which shows typically an example of the image which concerns on the image collation process of FIG. It is a block diagram which shows schematic structure of the portable terminal of Embodiment 2 to which this invention is applied. It is a flowchart which shows an example of the operation | movement which concerns on the image registration process by the portable terminal of FIG. It is a figure which shows typically an example of the image which concerns on the image registration process of FIG. It is a flowchart which shows an example of the operation | movement which concerns on the image collation process by the portable terminal of FIG. It is a figure which shows typically an example of the image which concerns on the image collation process of FIG. It is a figure which shows typically an example of the image which concerns on the image collation process of FIG.

  Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. However, the scope of the invention is not limited to the illustrated examples.

[Embodiment 1]
FIG. 1 is a block diagram showing a schematic configuration of a mobile terminal 100 according to the first embodiment to which the present invention is applied.
As illustrated in FIG. 1, the mobile terminal 100 according to the present embodiment includes a central control unit 1, a memory 2, an imaging unit 3, an imaging control unit 4, an image data generation unit 5, an image registration unit 6, An image processing unit 7, a display unit 8, a display control unit 9, a transmission / reception unit 10, a communication control unit 11, an operation input unit 12, and the like are provided.
The central control unit 1, the memory 2, the imaging unit 3, the imaging control unit 4, the image data generation unit 5, the image registration unit 6, the image processing unit 7, the display control unit 9, the transmission / reception unit 10, and the communication control unit 11 Are connected via a bus line 13.

  Note that the mobile terminal 100 includes, for example, an imaging device, a mobile station used in a mobile communication network such as a mobile phone or a PHS (Personal Handy-phone System), a PDA (Personal Data Assistants), and the like.

  The central control unit 1 controls each unit of the mobile terminal 100. Specifically, the central control unit 1 includes a CPU (Central Processing Unit; not shown) that controls each part of the mobile terminal 100, and performs various control operations according to various processing programs (not shown) for the mobile terminal 100. Do.

  The memory 2 is composed of, for example, a DRAM (Dynamic Random Access Memory) or the like. The memory 2 includes a buffer memory that temporarily stores data processed by the central control unit 1 and the image processing unit 7, a working memory such as the central control unit 1, and various types of functions related to the function of the mobile terminal 100. A program memory (not shown) in which programs and data are stored is provided.

The imaging unit 3 includes a lens unit 3a and an electronic imaging unit 3b.
The lens unit 3a includes a plurality of lenses such as a zoom lens and a focus lens.
The electronic imaging unit 3b is composed of, for example, an image sensor such as a charge coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS), and converts an optical image that has passed through various lenses of the lens unit 3a into a two-dimensional image signal. To do.
In addition, although illustration is abbreviate | omitted, the imaging part 3 may be provided with the aperture_diaphragm | restriction which adjusts the quantity of the light which passes the lens part 3a.

The imaging unit 3 captures an image of a seal S (see FIG. 5A) of a seal stamped on the recording medium as a specific subject.
Here, the imprint S includes code information obtained by encoding predetermined information into a regular array of pixel sets in the imprint image (for example, an image schematically representing a train) for forming the imprint S. It may be.
The code information is obtained by encoding original predetermined information (for example, URL) according to a predetermined encoding format (for example, Reed-Solomon code, Golay code, etc.). For example, in the code information, a set of white pixels having a pixel value “1” and a set of black pixels having a pixel value “0” are regularly arranged with a predetermined number of dimensions.

The imaging control unit 4 controls the imaging of the subject by the imaging unit 3. That is, the imaging control unit 4 includes a timing generator, a driver, and the like, although not illustrated. Then, the imaging control unit 4 scans and drives the electronic imaging unit 3b with a timing generator and a driver, converts the optical image into a two-dimensional image signal with the electronic imaging unit 3b at a predetermined period, and the electronic imaging unit 3b. The frame image is read out from the imaging area for each screen and is output to the image data generation unit 5.
In addition, the imaging control unit 4 performs adjustment control of imaging conditions of the subject such as AF (automatic focusing process), AE (automatic exposure process), AWB (automatic white balance), and the like.

  The image data generation unit 5 appropriately adjusts the gain for each color component of RGB with respect to the analog value signal of the frame image transferred from the electronic image pickup unit 3b, and then samples and holds it by a sample hold circuit (not shown). The digital signal is converted into digital data by a / D converter (not shown), color processing including pixel interpolation processing and γ correction processing is performed by a color process circuit (not shown), and then a digital luminance signal Y and color difference signal Cb , Cr (YUV data).

  Then, the image data generation unit 5 sequentially outputs the generated YUV data of each frame image to the memory 2 and stores it in the memory 2.

The image registration unit 6 includes a database 6a for registering at least one registered image Ia (see FIG. 3).
The database (registration means) 6a records each registered image Ia in association with the feature information. That is, the database 6a stores the coordinate positions and feature amounts of feature points (eg, SIFT (Scale-Invariant Features Transform) feature amounts, etc.) as feature information of each registered image Ia extracted by feature extraction processing (described later). Recorded for each registered image Ia. Here, the coordinate positions and feature amounts (feature information) of the feature points of each registered image Ia may be recorded as a table for each of the plurality of small areas A1,... Divided by the image dividing unit 7e.
Further, the database 6a may record each registered image Ia in association with subject information indicating the position of a predetermined subject existing in the registered image Ia and the size of the registered image Ia. Here, the subject information includes, for example, the average value (center of gravity) of the coordinate position of each pixel constituting the registered image Ia calculated by subject information calculation processing (described later) and the eigenvalue of the covariance matrix of the registered image Ia. It is expressed as the position of the subject and the size of the registered image Ia.

Examples of the registered image Ia include an imprint model image for forming a predetermined mark on the seal surface of a seal, an imprint image obtained by capturing an imprint S imprinted on a recording medium, and the like. However, the present invention is not limited to this, and can be arbitrarily changed as appropriate.
The registered image Ia may include code information obtained by encoding predetermined information into a regular array of pixel sets.

The image processing unit 7 includes an image acquisition unit 7a, a subject information calculation unit 7b, a feature extraction unit 7c, a corresponding region specification unit 7d, an image division unit 7e, a verification target region setting unit 7f, and an image verification unit 7g. It is equipped with.
Note that each unit of the image processing unit 7 is configured by, for example, a predetermined logic circuit, but the configuration is an example and is not limited thereto.

The image acquisition unit 7a acquires the captured image Ib (see FIG. 5A).
That is, the image acquisition unit (acquisition unit) 7a acquires a captured image Ib in which a specific subject (for example, an imprint S imprinted on a recording medium) is captured. Specifically, the image acquisition unit 7a stores a copy of image data (for example, luminance data) of a predetermined resolution of the captured image Ib generated by the image data generation unit 5 after the impression S is captured by the imaging unit 3. Get from 2.

The subject information calculation unit 7b performs subject information calculation processing on the captured image Ib and the registered image Ia.
In other words, the subject information calculation unit (calculation unit) 7b calculates subject information representing the position and size of a specific subject in the captured image Ib acquired by the image acquisition unit 7a. Specifically, for example, the subject information calculation unit 7b performs subject information calculation processing on the image data of the captured image Ib, and calculates the average value (center of gravity) of the coordinate position of each pixel constituting the captured image Ib. The eigenvalues of the covariance matrix of the captured image Ib are calculated as subject information representing the position of a specific subject (for example, the seal impression S) and the size of the captured image Ib.
In addition, the subject information calculation unit 7b performs subject information calculation processing on the image data of the registered image Ia registered in the database 6a of the image registration unit 6, and the position and registration of a predetermined subject in the registered image Ia. Subject information representing the size of the image Ia (for example, the center of gravity of the registered image Ia and the eigenvalue of the covariance matrix) is calculated. Here, the subject information calculation process may be performed on an image already registered in the database 6a (registered image), or may be performed on an image to be registered (registered image). Also good.
Then, the subject information calculation unit 7 b outputs the calculated subject information to the memory 2 and temporarily stores it in the memory 2.

The feature extraction unit 7c performs feature extraction processing on the captured image Ib and the registered image Ia.
Specifically, for example, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the captured image Ib acquired by the image acquisition unit 7a to extract a plurality of feature points. A feature amount (for example, a SIFT feature amount) is calculated.
In addition, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the registered image Ia registered in the database 6a of the image registration unit 6, for example, to extract a plurality of feature points. The feature amount is calculated. Here, the feature extraction processing may be performed on an image (registered image) already registered in the database 6a, or may be performed on an image to be registered (registered image). good.
Then, the feature extraction unit 7 c outputs the calculated coordinate position and feature amount (feature information) of the feature point to the memory 2 and temporarily stores them in the memory 2.
The feature information may be represented by using, for example, a SURF (Speeded Up Robust Features) feature amount, or a brightness value (0 to 255) expressing the brightness level of the feature point in 8 bits (256 levels). Alternatively, RGB values (0 to 255) expressing each color component of RGB in 8 bits (256 levels) may be used.

The corresponding area specifying unit 7d specifies the corresponding area B (see FIG. 5A) in the captured image Ib.
That is, the corresponding region specifying unit (specifying unit) 7d is configured to detect the captured image based on the pixel information of the registered image Ia registered in the image registering unit 6 and the pixel information of the captured image Ib acquired by the image acquiring unit 7a. A corresponding area B corresponding to the registered image Ia is specified in Ib. Specifically, the corresponding area specifying unit 7d specifies the corresponding area B in the captured image Ib based on the subject information of each of the captured image Ib and the registered image Ia calculated by the subject information calculating unit 7b.
For example, the corresponding region specifying unit (subject information acquisition unit) 7d acquires the center of gravity of the captured image Ib calculated by the subject information calculation unit 7b and the eigenvalue of the covariance matrix. Further, the corresponding area specifying unit 7d acquires the center of gravity of the registered image Ia to be collated and the eigenvalue of the covariance matrix among the registered images Ia registered in the image registering unit 6. Then, the corresponding region specifying unit 7d compares the eigenvalues of the center of gravity and the covariance matrix of the captured image Ib with the eigenvalues of the registered image Ia to be verified and the eigenvalues of the covariance matrix, respectively, A corresponding area B corresponding to the image Ia is specified.
As a result, the approximate position and size of the region (corresponding region B) to be collated with the registered image Ia are specified in the captured image Ib. That is, a corresponding area B having an outline substantially similar to the outer shape (for example, a square) of the registered image Ia is specified, and the relative position and size of a specific subject (for example, an imprint S) in the corresponding area B is specified. Is substantially equal to the relative position and size of a predetermined subject in the registered image Ia.

  For example, the corresponding area specifying unit 7d may specify the entire captured image Ib as the corresponding area B when it is determined from the content of the subject information of the captured image Ib that the specific subject exists in the entire composition. That is, the corresponding area specifying unit 7d may specify the corresponding area B in the captured image Ib based on the subject information of the captured image Ib calculated by the subject information calculating unit 7b.

The image dividing unit 7e divides the corresponding area B of the captured image Ib into a plurality of small areas B1,.
That is, the image dividing unit (dividing unit) 7e divides the corresponding area B specified by the corresponding area specifying unit 7d into a plurality of small areas B1,.
Specifically, the image dividing unit 7e reduces the corresponding region B to a plurality of small areas according to the division mode (for example, bisection or quadrant) of the registered image Ia registered in the database 6a of the image registration unit 6. The area is divided into areas B1,. That is, the image dividing unit 7e encloses the registered image Ia registered in the database 6a of the image registration unit 6 in advance with a plurality of rectangular small regions (two-dot chain lines) in a predetermined division mode (for example, quadrant). The area is divided into A1,... Here, the division of the registered image Ia may be performed on an image already registered in the database 6a (registered image), or may be performed on an image to be registered (registered image). May be.
Then, the image dividing unit 7e sets the corresponding region B to a small region (indicated by a one-dot chain line) having substantially the same shape and size so as to have the same number of divisions as the number (for example, four) of the registered image Ia divided into the small regions A1. The enclosed area; see FIG. 5B) is divided into B1.
At this time, the image dividing unit 7e may divide the corresponding region B into a plurality of small regions B1,... So as to overlap end portions of the adjacent small regions B1. For example, in the case of the upper left small region B1 of the corresponding region B shown in FIG. 5B, the image dividing unit 7e overlaps the right end of the upper left small region B1 with the left end of the upper right small region A1, and The lower end of the upper left small area B1 is overlapped with the upper end of the lower left small area B1, and the lower right end of the upper left small area B1 is overlapped with the upper left end of the lower right small area B1 To divide. The other small regions B1 are substantially the same, and detailed description thereof is omitted.
Note that the number of divisions and the division mode of the small areas A1 and B1 described above are examples and are not limited thereto, and can be arbitrarily changed as appropriate. For example, the outer shapes of the small regions A1 and B1 may be different from each other, and the sizes of the regions where the adjacent small regions B1 overlap with the end portions may be different from each other.

The verification target area setting unit 7f sets a verification target area C (see FIG. 6A) in the captured image Ib.
That is, the collation target area setting unit (setting unit) 7f sets the collation target area C in the corresponding area B specified by the corresponding area specifying unit 7d. Specifically, the collation target area setting unit 7f sets a predetermined number of small areas B1 among the plurality of small areas B1,... Obtained by dividing the corresponding area B by the image dividing unit 7e as collation target areas C. That is, the matching target area setting unit 7 f sets a matching target area C smaller than the corresponding area B in the corresponding area B.
Here, the collation target area setting unit 7f may change the position and number of the small areas B1 set as the collation target area C in consideration of the positions of the feature points extracted by the feature extraction process, for example. . That is, when feature points are extracted from substantially the entire corresponding area B, the collation target area setting unit 7f individually sets all of the plurality of small areas B1,. May be set. When feature points are extracted from a part of the corresponding region B, the collation target region setting unit 7f includes a predetermined region including a region from which the feature points are extracted among the plurality of small regions B1,. A number of small regions B1 may be set as the verification target region C.
In FIG. 6A, the corresponding region B other than the verification target region C (upper left small region B1) is schematically shown with dots.

The image collating unit 7g collates the collation target area C of the captured image Ib with the registered image Ia.
That is, the image collation unit (collation unit) 7g includes the feature information of the collation target region C set by the collation target region setting unit 7f and the region corresponding to the collation target region C in the registered image Ia (small region for collation). D) feature information is associated with each other, and the collation target area C and the registered image Ia are collated based on the result of the association.

Specifically, the image matching unit 7g stores the coordinate positions and feature amounts (feature information) of feature points in the matching target area C (for example, the upper left small area A1) extracted from the feature extraction process from the memory 2. get. Further, the image collating unit 7g identifies a collation small area D (see FIG. 6A) corresponding to the collation target area C in the registered image Ia, and features in the identified collation small area D are identified. The coordinate position and the feature amount (feature information) of the point are acquired from the database 6 a of the image registration unit 6. In FIG. 6A, regions other than the collation small region D in the registered image Ia are schematically shown with dots.
The image matching unit 7g then associates (matches) the acquired feature points in the matching target area C with the feature points in the matching small area D. At this time, when a plurality of collation target areas C are set, the image collation unit 7g specifies a corresponding collation small area D for each collation target area C and associates the feature points with each other. Do.
In addition, the image matching unit 7g registers the shape of a specific subject in the matching target area C based on the result of the association between the feature points in the matching target area C and the feature points in the matching small area D. A coordinate transformation matrix (example of homography line) for transforming into a shape substantially equal to the shape of the subject in the image Ia is calculated. Then, the image matching unit 7g performs a coordinate conversion process on the matching target region C according to the calculated coordinate conversion formula, and generates a coordinate-converted image Ic (see FIG. 6B).
Further, the image matching unit 7g performs a predetermined calculation based on the pixel value of the generated coordinate-converted image Ic and the pixel value of the registered image Ia, and determines the similarity between the matching target region C and the registered image Ia. Calculate (similarity calculation processing). As the similarity calculation processing, for example, adaboost (adaboost) output calculation, SSD (Sum of Squared Differences), etc. can be used. However, this is an example and is not limited to this, and can be arbitrarily changed as appropriate. Is possible.
Note that the projective transformation process and the similarity calculation process described above are known techniques, and thus detailed description thereof is omitted here.

  The display unit 8 is composed of, for example, a liquid crystal display panel, and displays an image (for example, a live view image) captured by the imaging unit 3 based on a video signal from the display control unit 9 on a display screen.

The display control unit 9 performs control to read out display image data temporarily stored in the memory 2 and display it on the display unit 8.
Specifically, the display control unit 9 includes a VRAM (Video Random Access Memory), a VRAM controller, a digital video encoder, and the like. The digital video encoder reads the luminance signal Y and the color difference signals Cb and Cr read from the memory 2 and stored in the VRAM (not shown) under the control of the central control unit 1 through the VRAM controller. Are read out at a predetermined playback frame rate (for example, 30 fps), and a video signal is generated based on these data and output to the display unit 8.
For example, the display control unit 9 performs live updating on the display unit 8 while sequentially updating a plurality of frame images captured by the imaging unit 3 and the imaging control unit 4 and generated by the image data generation unit 5 at a predetermined display frame rate. Display the view.

The transmitter / receiver unit 10 performs a call with an external user of an external device connected via the communication network N.
Specifically, the transmission / reception unit 10 includes a microphone 10a, a speaker 10b, a data conversion unit 10c, and the like. The transmission / reception unit 10 performs A / D conversion processing on the user's transmission voice input from the microphone 10a by the data conversion unit 10c and outputs the transmission voice data to the central control unit 1. Under the control, voice data such as received voice data outputted and inputted from the communication control unit 11 is D / A converted by the data conversion unit 10c and outputted from the speaker 10b.

The communication control unit 11 transmits and receives data via the communication network N and the communication antenna 11a.
That is, the communication antenna 11a is a predetermined communication method (for example, W-CDMA (Wideband Code Division Multiple Access) method, GSM (Global System)) used by the mobile terminal 100 for communication with a radio base station (not shown). for Mobile Communications (registered trademark) system, etc.). The communication control unit 11 transmits / receives data to / from the radio base station via the communication antenna 11a through a communication channel set in the communication method according to a communication protocol corresponding to a predetermined communication method. That is, the communication control unit 11 transmits / receives voice during a call with an external user of the external device to the external device of the communication partner based on the instruction signal output from the central control unit 1 and input, Send and receive e-mail data.
Note that the configuration of the communication control unit 11 is an example and is not limited thereto, and can be arbitrarily changed as appropriate. For example, although not shown, a wireless LAN module is mounted and an access point (Access Point) is provided. It is good also as a structure which can access the communication network N via this.

The communication network N is a communication network that connects the mobile terminal 100 to an external device via a wireless base station, a gateway server (not shown), or the like, for example.
The communication network N is a communication network constructed using, for example, a dedicated line or an existing general public line, and various line forms such as a LAN (Local Area Network) and a WAN (Wide Area Network) are applied. Is possible. The communication network N includes, for example, various communication network networks such as a telephone line network, ISDN line network, dedicated line, mobile communication network, communication satellite line, CATV line network, IP network, VoIP (Voice over Internet Protocol). ) Gateways, Internet service providers, etc. are included.

The operation input unit 12 is for inputting various instructions to the terminal body.
Specifically, the operation input unit 12 executes a shutter button related to a subject photographing instruction, up / down / left / right cursor buttons and a determination button related to a selection instruction of a mode, a function, etc., making / receiving a call, sending / receiving an e-mail, etc. Various buttons (not shown) such as communication-related buttons related to instructions and numeric buttons and symbol buttons related to text input instructions are provided.
When various buttons are operated by the user, the operation input unit 12 outputs an operation instruction corresponding to the operated button to the central control unit 1. The central control unit 1 causes each unit to execute a predetermined operation (for example, imaging of a subject, incoming / outgoing calls, transmission / reception of an e-mail, etc.) according to an operation instruction output from the operation input unit 12 and input.

  The operation input unit 12 may include a touch panel provided integrally with the display unit 8, and an operation instruction corresponding to the predetermined operation is given to the central control unit based on a predetermined operation of the touch panel by the user. 1 may be output.

<Image registration process>
Next, image registration processing by the mobile terminal 100 will be described with reference to FIGS.
FIG. 2 is a flowchart illustrating an example of an operation related to the image registration process.

  As shown in FIG. 2, the image acquisition unit 7a of the image processing unit 7 acquires image data (YUV data) of an image to be registered (registration target image) registered in the database 6a of the image registration unit 6 ( Step S1; see FIG. For example, the image acquisition unit 7a acquires, as a registration target image, an imprint model image for forming a predetermined mark on a seal face from a not-shown external device or recording medium.

  Next, the subject information calculation unit 7b performs subject information calculation processing on the image data of the registration target image, and calculates the average value (center of gravity) of the coordinate positions of each pixel constituting the registration target image and the registration target image. The eigenvalues of the covariance matrix are calculated as subject information (step S2). Subsequently, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the registration target image to extract a plurality of feature points, and calculates a feature amount (for example, a SIFT feature amount) at an effective feature point. (Step S3).

Then, the image dividing unit 7e divides the registration target image into a plurality of small regions A1,... In a predetermined division mode (for example, equally divided into four) (step S4).
Thereafter, the image registration unit 6 associates the coordinate positions and feature amounts of the feature points for each of the plurality of small regions A1,... Of the registration target image, and also includes the center of gravity of the registration target image and the eigenvalues of the covariance matrix (subject information). Are associated and recorded in the database 6a (step S5).

  Then, the CPU of the central control unit 1 determines whether there are other images to be registered (step S6). Here, if it is determined that there is another image to be registered (step S6; YES), the CPU of the central control unit 1 returns the process to step S1, and uses the other registration target image as a processing target. The processing unit 7 and the image registration unit 6 are caused to execute each subsequent process.

  On the other hand, when it is determined in step S6 that there is no other image to be registered (step S6; NO), the CPU of the central control unit 1 ends the image registration process.

<Image matching process>
Next, image collation processing by the mobile terminal 100 will be described with reference to FIGS.
FIG. 4 is a flowchart illustrating an example of an operation related to the image matching process.
It is assumed that the seal S imaged in the following image collation process is stamped at a predetermined position on a recording medium such as a postcard.

As shown in FIG. 4, first, when an imaging instruction is input based on a predetermined operation of the operation input unit 12 by the user, the imaging control unit 4 causes the imaging unit 3 to image the seal S, and the image data generation unit 5. Generates image data of the captured image Ib transferred from the electronic imaging unit 3b (step S11). At this time, for example, the imprint S is imaged in a state in which the lens unit 3a of the image capturing unit 3 is arranged obliquely with respect to the imprint S so as to have a predetermined angle.
Then, the image data generation unit 5 outputs the generated YUV data of the captured image Ib to the memory 2 and stores it in the memory 2.

  In addition, in order to make it easy to image the seal impression S, the display control unit 9 may cause the display unit 8 to display a guide display corresponding to the outer shape of the seal impression S.

Then, the image acquisition unit 7a of the image processing unit 7 acquires image data (for example, luminance data) of a predetermined resolution of the captured image Ib generated by the image data generation unit 5 from the memory 2 (step S12).
Next, the subject information calculation unit 7b performs subject information calculation processing on the image data of the captured image Ib, and calculates the average value (center of gravity) of the coordinate position of each pixel constituting the captured image Ib and the captured image Ib. The eigenvalue of the covariance matrix is calculated as subject information (step S13). Subsequently, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the captured image Ib to extract a plurality of feature points, and calculates a feature amount (for example, a SIFT feature amount) at an effective feature point. (Step S14).

  Thereafter, the corresponding area specifying unit 7d reads out and acquires the image data of any one registered image Ia to be collated, the centroid of the registered image Ia, and the eigenvalue of the covariance matrix from the database 6a of the image registration unit 6 ( Step S15). Subsequently, the corresponding area specifying unit 7d acquires the centroid and eigenvalue of the covariance matrix of the captured image Ib calculated by the subject information calculation unit 7b, and the centroid of the captured image Ib and the eigenvalue of the covariance matrix and the matching target The center of gravity of the registered image Ia and the eigenvalue of the covariance matrix are respectively compared, and the corresponding region B corresponding to the registered image Ia to be collated is specified in the captured image Ib (step S16; see FIG. 5A).

Next, the image dividing unit 7e divides the corresponding area B into a plurality of (for example, four) small areas B1 according to the division mode (for example, quadrant) of the registered image Ia (step S17; FIG. 5). (See (b)). Specifically, the image dividing unit 7e divides the corresponding region B so that end portions of the adjacent small regions B1 overlap each other among the plurality of small regions B1,.
Then, the collation target area setting unit 7f sets a predetermined number of small areas B1 (for example, the upper left small area B1) among the plurality of small areas B1,. S18).

Next, the image matching unit 7g acquires the coordinate position and feature amount (feature information) of the feature point in the matching target area C from the memory 2 and uses the matching point corresponding to the matching target area C in the registered image Ia. The coordinate position and feature amount (feature information) of the feature points in the small area D are acquired from the database 6a, and the feature points in the collation target area C and the feature points in the collation small area D are associated (matching). (Step S19; see FIG. 6A). Subsequently, the image matching unit 7g calculates a coordinate transformation matrix (homography row example) based on the result of the correspondence between the feature points in the matching target region C and the feature points in the matching small region D. Then, a coordinate conversion process is performed on the verification target area C according to the calculated coordinate conversion formula to generate a coordinate-converted image Ic (step S20; see FIG. 6B).
Then, the image matching unit 7g performs a predetermined calculation based on the pixel value of the generated coordinate-converted image Ic and the pixel value of the registered image Ia, and determines the similarity between the matching target region C and the registered image Ia. Calculate (step S21). The image matching unit 7g outputs the calculated similarity to the memory 2, and the similarity is temporarily stored in association with the identification information of the registered image Ia.
Note that the coordinate conversion process for the verification target area C is not necessarily performed, and the process may be skipped.

Thereafter, the CPU of the central control unit 1 determines whether there is another registered image Ia to be collated (step S22). Here, if it is determined that there is another image to be collated (step S22; YES), the CPU of the central control unit 1 returns the process to step S15 and selects another registered image Ia from the database 6a as a collation target. Then, the image processing unit 7 is caused to execute each subsequent process.
On the other hand, when it is determined in step S22 that there is no other registered image Ia to be collated (step S22; NO), the image processing unit 7 acquires the similarity corresponding to each registered image Ia from the memory 2. Then, the registered image Ia having the highest similarity is selected (step S23), and the image registration process is terminated.

Thereafter, the information reading unit (not shown) of the image processing unit 7 generates a binarized image of the selected registered image Ia, for example, and collects a set of white pixels having a pixel value “1” and a pixel value “0”. The array of black pixels is subjected to a decoding process according to the encoding method of code information, and original predetermined information (for example, URL) represented by the code information is read.
The CPU of the central control unit 1 controls each unit of the portable terminal 100 according to predetermined information (for example, URL) read by the information reading unit, and performs predetermined operations (for example, accessing the Internet to specify specific information). Sound, image playback, etc.).
The predetermined information reading process and the corresponding predetermined operation execution process are not necessarily performed.

As described above, according to the mobile terminal 100 of the first embodiment, the pixel information and the captured image of the registered image Ia registered in the image registration unit 6 (for example, an image obtained by capturing the imprint S imprinted on the recording medium) ) Based on the pixel information of Ib, the corresponding region B corresponding to the registered image Ia is specified in the captured image Ib, the matching target region C is set in the corresponding region B, and the set matching target region C The feature information (for example, the position and feature amount of the feature point) is associated with the feature information of the region (for example, the small region D for matching) corresponding to the matching target region C in the registered image Ia, and the association Based on the result, the verification target area C and the registered image Ia can be verified. At this time, as a rough alignment, a corresponding area B corresponding to the registered image Ia is specified in the captured image Ib in advance, and a collation target area C smaller than the corresponding area B is set and used in the corresponding area B. Thus, even if the image is composed of relatively simple textures and has a plurality of similar textures (for example, a pantograph, a window, a wheel, etc. in a train image), similar textures are present in the comparison target region C. It is difficult for a plurality to exist, and it is difficult to cause an incorrect correspondence in the correspondence between the feature information in the verification target region C and the feature information of the registered image Ia.
That is, for example, in the corresponding region B shown in FIG. 6A, there is a portion corresponding to the pantograph on the left side of the train in the verification target region C, and the region other than the verification target region C (region with dots). There is a part corresponding to the pantograph on the right side. On the other hand, only a portion corresponding to the left pantograph exists in the small area D for verification of the registered image Ia. Therefore, for example, it is possible to prevent the feature information of the portion corresponding to the pantograph on the left side of the captured image Ib from being erroneously associated with the feature information of the portion corresponding to the pantograph on the right side of the registered image Ia.
As a result, the comparison target region C of the captured image Ib and the registered image Ia can be properly compared, and the images can be accurately verified with high accuracy.

In addition, since a predetermined number of small regions B1 among the plurality of small regions B1,... Obtained by dividing the corresponding region B are set as the verification target region C, the verification target region C smaller than the corresponding region B in the captured image Ib. Setting can be simplified. In particular, by dividing the corresponding area B into a plurality of small areas B1,... So that the ends of the adjacent small areas B1 overlap each other, it is possible to appropriately deal with the feature information on the boundary between the adjacent small areas B1. Thus, it is possible to make it more difficult to cause erroneous correspondence in the association between the feature information in the verification target region C and the feature information of the registered image Ia.
Further, the registered image Ia is divided into a plurality of small areas A1,... And the corresponding area B is divided into a plurality of small areas B1,... According to the division mode of the registered image Ia. The region corresponding to the region C can be specified appropriately, and the feature information in the collation target region C and the feature information of the region corresponding to the collation target region C in the registered image Ia are more appropriately associated. Can do.

  Further, since the corresponding region B is specified in the captured image Ib based on the position of the specific subject in the captured image Ib and the subject information indicating the size of the captured image Ib, the position of the specific subject in the captured image Ib The corresponding region B can be appropriately specified in the captured image Ib in consideration of the size of the captured image Ib and the like. In particular, by specifying the corresponding region B in the captured image Ib based on the subject information of the registered image Ia in addition to the subject information of the captured image Ib, the position of the specific subject in the captured image Ib and the captured image Ib Considering not only the size but also the position of a predetermined subject in the registered image Ia, the size of the registered image Ia, and the like, the corresponding region B can be more appropriately specified in the captured image Ib.

[Embodiment 2]
Below, the portable terminal 200 of Embodiment 2 is demonstrated with reference to FIGS.
The portable terminal 200 according to the second embodiment has substantially the same configuration as the portable terminal 100 according to the first embodiment except for the details described below, and detailed description thereof is omitted.

FIG. 7 is a block diagram showing a schematic configuration of the mobile terminal 200 according to the second embodiment to which the present invention is applied.
As shown in FIG. 7, the image processing unit 207 of the mobile terminal 200 according to the second embodiment includes an image acquisition unit 7a, a face detection unit 7h, a feature extraction unit 7c, a corresponding area specifying unit 7d, and an image dividing unit 7e. A collation target area setting unit 7f and an image collation unit 7g.
Note that the components constituting the image processing unit 207 have substantially the same configurations and functions as those provided in the portable terminal 100 of the first embodiment except for those described below, and detailed descriptions thereof are omitted.

The face detection unit 7h performs face detection processing on the captured image Ib and the registered image Ia.
That is, the face detection unit 7h performs a predetermined face detection process on the image data of the captured image Ib acquired by the image acquisition unit 7a, and the person (human) that is a specific subject in the captured image Ib. A face area Eb including the face F is detected. At this time, the face detection unit 7h may detect each component such as eyes, nose, and mouth constituting the face F, for example.
Further, the face detection unit 7h performs a predetermined face detection process on the image data of the registered image Ia (see FIG. 9A, etc.) registered in the database 6a of the image registration unit 6, and the registered image Ia An inner face area Ea is detected. Here, the face detection process may be performed on an image (registered image) already registered in the database 6a, or may be performed on an image to be registered (registered image). good.
Further, since the face detection process described above is a known technique, detailed description thereof is omitted here.

The corresponding area specifying unit 7d specifies the face area Eb detected by the face detecting unit 7h in the captured image Ib as the corresponding area B (see FIG. 11A).
That is, the corresponding area specifying unit 7d uses the face area Eb detected by the face detecting unit 7h in the captured image Ib as a corresponding area B corresponding to the face area Ea of the registered image Ia registered in the image registration unit 6. Identify.

The image dividing unit 7e divides the face area Eb specified as the corresponding area B of the captured image Ib into a plurality of small areas B1,.
Specifically, the image division unit 7e divides the face area Eb of the captured image Ib into a plurality of small areas B1,... According to the division mode of the registered image Ia registered in the database 6a of the image registration unit 6. In other words, the image dividing unit 7e uses the registered image Ia registered in advance in the database 6a of the image registration unit 6 as the face area Ea detected by the face detecting unit 7h (in particular, the eyes, nose, mouth in the face area Ea). Are divided into a plurality of small rectangular regions (regions surrounded by a two-dot chain line) A1,. For example, the image dividing unit 7e divides the face area Ea into left and right halves (see FIG. 9A), divides only the upper side of the face area Ea into halves (see FIG. 9B), face The area Ea is divided into a plurality of small areas A1,... In a manner of equally dividing the area Ea vertically and horizontally (see FIG. 9C). Here, the division of the registered image Ia may be performed on an image already registered in the database 6a (registered image), or may be performed on an image to be registered (registered image). May be.
Then, the image dividing unit 7e reduces the face region Eb (corresponding region B) to a small size having a shape and size substantially equal to the number of divisions of the registered image Ia into the small regions A1 (for example, two). The area is divided into areas (areas surrounded by alternate long and short dash lines; see FIG. 11B) B1.
At this time, the image dividing unit 7e may divide the face region Eb into a plurality of small regions B1,... So that the ends of the adjacent small regions B1 overlap each other. For example, the image dividing unit 7e overlaps the right end of the left small region B1 of the face region Eb with the left end of the right small region B1, and the left end of the right small region B1 to the right of the left small region B1. Split to overlap the edges.
Note that the number of divisions and the division mode of the small areas A1 and B1 described above are examples and are not limited thereto, and can be arbitrarily changed as appropriate. For example, the size of the region where the adjacent small region B1 overlaps the end portions may be different from each other.

<Image registration process>
Next, image registration processing by the mobile terminal 200 will be described with reference to FIGS.
FIG. 8 is a flowchart illustrating an example of an operation related to the image registration process.

  As shown in FIG. 2, the image acquisition unit 7 a of the image processing unit 207 acquires image data (YUV data) of an image to be registered (image to be registered) registered in the database 6 a of the image registration unit 6 ( Step S31; see FIG. For example, the image acquisition unit 7a acquires an image obtained by capturing a specific subject in advance as a registration target image from an external device or a recording medium (not shown).

Next, the face detection unit 7h performs a predetermined face detection process on the image data of the registration target image to detect a face area Ea in the registration image Ia (step S32). When the image of the face area of the substantially entire area is set as the registered image Ia, the face detection unit 7h does not necessarily perform the face detection process on the registered image Ia.
Subsequently, in substantially the same manner as the image registration process of the first embodiment, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the registration target image to extract a plurality of feature points, and an effective feature point. The feature amount (for example, SIFT feature amount) is calculated (step S3).

Then, the image dividing unit 7e includes a plurality of registration target images in a predetermined division mode (for example, a mode in which the face region Ea is equally divided into left and right) based on the face region Ea detected by the face detection unit 7h. It divides | segments into small area | region A1, ... (step S34).
Thereafter, the image registration unit 6 records the coordinate position and the feature amount of each feature point in the database 6a for each of the plurality of small regions A1,... Of the registration target image (step S35).

  Then, in substantially the same manner as the image registration process of the first embodiment, the CPU of the central control unit 1 determines whether there is another image to be registered (step S6). Here, if it is determined that there is another image to be registered (step S6; YES), the CPU of the central control unit 1 returns the process to step S1, and uses the other registration target image as a processing target. The processing unit 207 and the image registration unit 6 are caused to execute each subsequent process.

  On the other hand, when it is determined in step S6 that there is no other image to be registered (step S6; NO), the CPU of the central control unit 1 ends the image registration process.

<Image verification processing>
Next, image collation processing by the mobile terminal 200 will be described with reference to FIGS.
FIG. 10 is a flowchart illustrating an example of an operation related to the image matching process.

As illustrated in FIG. 10, first, when an imaging instruction is input based on a predetermined operation of the operation input unit 12 by the user, the imaging control unit 4 causes the imaging unit 3 to capture an image of a person's face F as a subject. The data generation unit 5 generates image data of the captured image Ib transferred from the electronic imaging unit 3b (step S41). At this time, for example, the human face F is assumed to be captured obliquely with respect to the lens unit 3a of the imaging unit 3 or rotated at a predetermined angle about the optical axis.
Then, the image data generation unit 5 outputs the generated YUV data of the captured image Ib to the memory 2 and stores it in the memory 2.

The image acquisition unit 7a of the image processing unit 207 has image data with a predetermined resolution (for example, the captured image Ib generated by the image data generation unit 5 from the memory 2), as in the image collation process of the first embodiment. , Luminance data) is acquired (step S12).
Next, the face detection unit 7h performs a predetermined face detection process on the image data of the captured image Ib to detect a face region Eb in the captured image Ib (step S43). Subsequently, in substantially the same manner as in the image matching process in the first embodiment, the feature extraction unit 7c performs a predetermined feature extraction process on the image data of the captured image Ib to extract a plurality of feature points, and an effective feature point. The feature amount (for example, SIFT feature amount) is calculated (step S14).

  Thereafter, the corresponding area specifying unit 7d reads out and acquires the image data of any one registered image Ia to be collated from the database 6a of the image registration unit 6 (step S45). Subsequently, the corresponding area specifying unit 7d specifies the face area Eb detected by the face detecting unit 7h in the captured image Ib as the corresponding area B corresponding to the face area Ea of the registered image Ia to be collated (step S46). ; See FIG. 11 (a)).

Next, the image dividing unit 7e has a plurality of (for example, two) face areas Eb of the captured image Ib in accordance with a division mode of the registered image Ia (for example, a mode in which the face area Ea is divided into two equal parts). It divides | segments into small area | region B1, ... (step S47; refer FIG.11 (b)).
Then, the collation target area setting unit 7f sets a predetermined number of small areas B1 as a collation target area C (for example, the left small area B1) among the plurality of small areas B1,. S48).

Next, in substantially the same manner as the image matching process of the first embodiment, the image matching unit 7g acquires the coordinate position and feature amount (feature information) of the feature point in the matching target area C from the memory 2 and also registers the registered image. The coordinate position and feature amount (feature information) of the feature point in the small region D for matching corresponding to the matching target region C in Ia are acquired from the database 6a, and the feature point in the matching target region C and the feature point for matching are acquired. Matching (matching) with feature points in the small region D is performed (step S19; see FIG. 12A). Subsequently, the image matching unit 7g calculates a coordinate transformation matrix (homography row example) based on the result of the correspondence between the feature points in the matching target region C and the feature points in the matching small region D. Then, a coordinate conversion process is performed on the verification target region C according to the calculated coordinate conversion formula to generate a coordinate-converted image Ic (step S20; see FIG. 12B).
Then, the image matching unit 7g performs a predetermined calculation based on the pixel value of the generated coordinate-converted image Ic and the pixel value of the registered image Ia, and determines the similarity between the matching target region C and the registered image Ia. Calculate (step S21). The image matching unit 7g outputs the calculated similarity to the memory 2, and the similarity is temporarily stored in association with the identification information of the registered image Ia.

Thereafter, in substantially the same manner as the image collating process of the first embodiment, the CPU of the central control unit 1 determines whether there is another registered image Ia to be collated (step S22). Here, if it is determined that there is another image to be collated (step S22; YES), the CPU of the central control unit 1 returns the process to step S45, and sets the other registered image Ia as a collation target. The processing unit 207 is caused to execute each subsequent process.
On the other hand, when it is determined in step S22 that there is no other registered image Ia to be collated (step S22; NO), the image processing unit 207 obtains the similarity corresponding to each registered image Ia from the memory 2. Then, the registered image Ia having the highest similarity is selected (step S23), and the image registration process is terminated.

  As described above, according to the mobile terminal 200 of the second embodiment, as in the first embodiment, the comparison target region C of the captured image Ib and the registered image Ia can be appropriately compared. Thus, the images can be properly verified with high accuracy. Furthermore, by specifying the face area Eb including the face F of the person who is a specific subject as the corresponding area B in the captured image Ib, it is possible to more easily specify the corresponding area B in the captured image Ib. can do. That is, unlike the first embodiment, it is not necessary to calculate the subject information of the captured image Ib or the registered image Ia, and the corresponding region B is obtained by using the result of the face detection process generally performed in the human image capturing process. Can be specified more appropriately and more easily.

The present invention is not limited to the first and second embodiments, and various improvements and design changes may be made without departing from the spirit of the present invention.
For example, in the first and second embodiments, the captured image Ib captured by the imaging unit 3 is acquired. However, it is not always necessary to capture an image, and whether or not the imaging unit 3 is provided is arbitrarily determined as appropriate. It can be changed. For example, a captured image captured by an external device may be acquired.

  In the first and second embodiments, the verification target area C is selected from the plurality of small areas B1,... Obtained by dividing the corresponding area B. However, this is an example, and the present invention is not limited to this. It is not always necessary to divide the corresponding region B, and whether or not the image dividing unit 7e is provided can be arbitrarily changed as appropriate. For example, the collation target area setting unit 7f sets the collation target area C so that the area from which the feature points are extracted is included in the corresponding area B in consideration of the position of the feature point extracted by the feature extraction process. Also good.

  Further, in the first and second embodiments, the mobile terminals 100 and 200 are illustrated as the image matching devices. However, the portable terminals 100 and 200 are only examples, and are not limited thereto. It can be arbitrarily changed as appropriate.

In addition, in the first and second embodiments, functions as a registration unit, an acquisition unit, a specifying unit, a setting unit, and a collation unit are controlled by the image registration unit under the control of the central control unit 1 of the mobile terminal 100. 6, the image acquisition unit 7a, the corresponding region specifying unit 7d, the collation target region setting unit 7f, and the image collation unit 7g are configured to be driven. However, the present invention is not limited to this. A configuration may be realized in which a predetermined program or the like is executed by the CPU.
That is, a program including a registration processing routine, an acquisition processing routine, a specific processing routine, a setting processing routine, and a collation processing routine is stored in a program memory that stores the program. Then, the CPU of the central control unit 1 may function as means for registering at least one image in advance by a registration processing routine. Further, the CPU of the central control unit 1 may function as a means for acquiring a captured image Ib in which a specific subject is captured by an acquisition processing routine. Further, the CPU of the central control unit 1 corresponds to the registered image Ia in the captured image Ib based on the pixel information of the registered image Ia registered in advance and the acquired pixel information of the captured image Ib by the specific processing routine. It may be made to function as a means for specifying the corresponding area B to be performed. Further, the CPU of the central control unit 1 may function as a means for setting the collation target area C in the specified corresponding area B by the setting process routine. Further, the CPU of the central control unit 1 associates the set feature information of the verification target region C with the feature information of the region corresponding to the verification target region C in the registered image Ia by the verification processing routine. On the basis of the result of the above, it may be made to function as a means for comparing the verification target area C and the registered image Ia.

  Similarly, the subject information acquisition unit, the division unit, and the calculation unit may be realized by executing a predetermined program or the like by the CPU of the central control unit 1.

  Furthermore, as a computer-readable medium storing a program for executing each of the above processes, a non-volatile memory such as a flash memory or a portable recording medium such as a CD-ROM is applied in addition to a ROM or a hard disk. Is also possible. A carrier wave is also used as a medium for providing program data via a predetermined communication line.

[Appendix]
Although several embodiments of the present invention have been described, the scope of the present invention is not limited to the above-described embodiments, but includes the scope of the invention described in the claims and equivalents thereof.
The invention described in the scope of claims attached to the application of this application will be added below. The item numbers of the claims described in the appendix are as set forth in the claims attached to the application of this application.
<Claim 1>
Registration means for registering at least one image in advance;
Acquisition means for acquiring a captured image in which a specific subject is captured;
A specifying unit for specifying a corresponding area corresponding to the registered image in the captured image based on the pixel information of the registered image registered in the registration unit and the pixel information of the captured image acquired by the acquiring unit;
Setting means for setting a collation target area within the corresponding area specified by the specifying means;
The feature information of the comparison target region set by the setting unit is associated with the feature information of the region corresponding to the comparison target region in the registered image, and based on the result of the association, Collation means for collating registered images;
An image collating apparatus comprising:
<Claim 2>
The registration means registers the center of gravity of the registered image and the eigenvalue of the covariance matrix in association with each registered image,
Subject information acquisition means for acquiring the center of gravity of the captured image and eigenvalues of the covariance matrix,
The specifying unit compares the center of gravity of the captured image acquired by the subject information acquisition unit and the eigenvalue of the covariance matrix with the center of gravity of the registered image registered in the registration unit and the eigenvalue of the covariance matrix, respectively. The image matching apparatus according to claim 1, wherein the corresponding area is specified.
<Claim 3>
Further comprising a dividing means for dividing the corresponding area specified by the specifying means into a plurality of areas;
The image collating apparatus according to claim 1, wherein the setting unit sets a predetermined number of regions as the collation target region among the plurality of regions divided by the dividing unit.
<Claim 4>
The dividing unit further divides the image registered in the registration unit into a plurality of regions,
The registration means includes
The image collating apparatus according to claim 3, wherein an image is registered in association with the feature information for each of a plurality of areas divided by the dividing unit.
<Claim 5>
The image collating apparatus according to claim 4, wherein the dividing unit further divides the corresponding area into the plurality of areas in accordance with a division mode of an image registered in the registration unit.
<Claim 6>
The said division means further divides | segments the said corresponding | compatible area | region into these several small area | regions so that the edge parts of the mutually adjacent area | region may overlap. Image matching device.
<Claim 7>
A calculation unit that calculates subject information indicating the position of the specific subject and the size of the captured image in the captured image acquired by the acquisition unit;
The specifying means is:
The image matching apparatus according to claim 1, wherein the corresponding area is specified in the captured image based on the subject information calculated by the calculation unit.
<Claim 8>
The calculation means further calculates subject information representing a position of a predetermined subject in the registration image registered in the registration means and a size of the registration image,
The specifying means is:
The image collating apparatus according to claim 7, wherein the corresponding region is specified in the captured image based on subject information of each of the captured image and the registered image calculated by the calculating unit.
<Claim 9>
The image collating apparatus according to claim 1, wherein the specifying unit specifies an image area including an imprint that is the specific subject as the corresponding area.
<Claim 11>
The image collating apparatus according to claim 1, wherein the specifying unit specifies a face area including a face of a person who is the specific subject as the corresponding area.
<Claim 12>
An image matching method using an image matching device,
Processing to register at least one image in advance;
A process of acquiring a captured image in which a specific subject is captured;
Based on the pixel information of the registered image registered and the pixel information of the acquired captured image, a process of identifying a corresponding area corresponding to the registered image in the captured image;
A process of setting a matching target area within the identified corresponding area;
The set feature information of the matching target area is associated with the feature information of the area corresponding to the matching target area in the registered image, and the matching target area and the registered image are matched based on the result of the matching. Processing to match,
The image collation method characterized by including this.
<Claim 13>
The computer of the image verification device
Registration means for registering at least one image in advance;
Acquisition means for acquiring a captured image in which a specific subject is captured;
A specifying unit for specifying a corresponding region corresponding to the registered image in the captured image based on the pixel information of the registered image registered in the registration unit and the pixel information of the captured image acquired by the acquiring unit;
Setting means for setting a collation target area within the corresponding area specified by the specifying means;
The feature information of the comparison target region set by the setting unit is associated with the feature information of the region corresponding to the comparison target region in the registered image, and based on the result of the association, Verification means for verifying registered images,
A program characterized by functioning as

100, 200 Mobile terminal 1 Central control unit 6 Image registration unit 6a Database 7, 207 Image processing unit 7a Image acquisition unit 7b Subject information calculation unit 7d Corresponding region specifying unit 7e Image dividing unit 7f Collation target region setting unit 7g Image collating unit 7h Face detector

Claims (12)

Registration means for registering at least one image in advance;
Acquisition means for acquiring a captured image in which a specific subject is captured;
A specifying unit for specifying a corresponding area corresponding to the registered image in the captured image based on the pixel information of the registered image registered in the registration unit and the pixel information of the captured image acquired by the acquiring unit;
Setting means for setting a collation target area within the corresponding area specified by the specifying means;
The feature information of the comparison target region set by the setting unit is associated with the feature information of the region corresponding to the comparison target region in the registered image, and based on the result of the association, Collation means for collating registered images;
An image collating apparatus comprising: The registration means registers the center of gravity of the registered image and the eigenvalue of the covariance matrix in association with each registered image,
Subject information acquisition means for acquiring the center of gravity of the captured image and eigenvalues of the covariance matrix,
The specifying unit compares the center of gravity of the captured image acquired by the subject information acquisition unit and the eigenvalue of the covariance matrix with the center of gravity of the registered image registered in the registration unit and the eigenvalue of the covariance matrix, respectively. The image matching apparatus according to claim 1, wherein the corresponding area is specified. Further comprising a dividing means for dividing the corresponding area specified by the specifying means into a plurality of areas;
The image collating apparatus according to claim 1, wherein the setting unit sets a predetermined number of regions as the collation target region among the plurality of regions divided by the dividing unit. The dividing unit further divides the image registered in the registration unit into a plurality of regions,
The registration means includes
The image collating apparatus according to claim 3, wherein an image is registered in association with the feature information for each of a plurality of areas divided by the dividing unit.   The image collating apparatus according to claim 4, wherein the dividing unit further divides the corresponding area into the plurality of areas in accordance with a division mode of an image registered in the registration unit.   The said division means further divides | segments the said corresponding | compatible area | region into these several small area | regions so that the edge parts of the mutually adjacent area | region may overlap. Image matching device. A calculation unit that calculates subject information indicating the position of the specific subject and the size of the captured image in the captured image acquired by the acquisition unit;
The specifying means is:
The image matching apparatus according to claim 1, wherein the corresponding area is specified in the captured image based on the subject information calculated by the calculation unit. The calculation means further calculates subject information representing a position of a predetermined subject in the registration image registered in the registration means and a size of the registration image,
The specifying means is:
The image collating apparatus according to claim 7, wherein the corresponding region is specified in the captured image based on subject information of each of the captured image and the registered image calculated by the calculating unit.   The image collating apparatus according to claim 1, wherein the specifying unit specifies an image area including an imprint that is the specific subject as the corresponding area.   The image collating apparatus according to claim 1, wherein the specifying unit specifies a face area including a face of a person who is the specific subject as the corresponding area. An image matching method using an image matching device,
Processing to register at least one image in advance;
A process of acquiring a captured image in which a specific subject is captured;
Based on the pixel information of the registered image registered and the pixel information of the acquired captured image, a process of identifying a corresponding area corresponding to the registered image in the captured image;
A process of setting a matching target area within the identified corresponding area;
The set feature information of the matching target area is associated with the feature information of the area corresponding to the matching target area in the registered image, and the matching target area and the registered image are matched based on the result of the matching. Processing to match,
The image collation method characterized by including this. The computer of the image verification device
Registration means for registering at least one image in advance;
Acquisition means for acquiring a captured image in which a specific subject is captured;
A specifying unit for specifying a corresponding region corresponding to the registered image in the captured image based on the pixel information of the registered image registered in the registration unit and the pixel information of the captured image acquired by the acquiring unit;
Setting means for setting a collation target area within the corresponding area specified by the specifying means;
The feature information of the comparison target region set by the setting unit is associated with the feature information of the region corresponding to the comparison target region in the registered image, and based on the result of the association, Verification means for verifying registered images,
A program characterized by functioning as

Images