JP2007135501A - Nose characteristic information-producing apparatus and nose characteristic information-producing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus which can inexpensively, easily and quickly obtain information for discriminating a pet even by a general pet owner without damaging the pet and without a special device. <P>SOLUTION: A pet control apparatus 100 comprises an image input portion 111 for inputting a front face photograph of a pet, a nose outline image extraction portion 112 for extracting a nose outline image front face photograph along the outline of the nose, a correction determining portion 113 for determining whether a prescribed correction is necessary or not for the nose outline image, then subjecting the prescribed correction to the nose outline image, when the correction is necessary, or not subjecting the prescribed correction to the nose outline image, when the correction is not necessary, and then outputting the nose outline image, an element extraction portion 114 for extracting the outline of the nose and the outlines of the first nose nostril and the second nostril from the nose outline image outputted from the correction determining portion 113, an element shape-specifying portion 115 for specifying the outline shape of the nose and the outlines of a first nose nostril and a second nostril, and a nose characteristic information production portion for producing nose characteristic information including the outline shape of the nose and the outlines of the first nose nostril and the second nostril. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a nasal feature information generation apparatus and a nasal feature information generation program for generating nasal feature information used for individual identification of a dog, a cat, or the like from the shape of a nose of a mammal such as a dog or a cat and the shape of a nostril.

  To identify pets such as dogs and cats, pet characteristics are registered in a database (Patent Document 1), a microchip is embedded in the body of a pet (Patent Document 2), or attached to a collar (Patent Document 3). Or a method of analyzing DNA and registering the result in a database. In the case of a dog, there is a method such as attaching a signature tag recording a dog ID to the dog collar. In the case of dogs and cows, a method for identifying them by a nose pattern has been used (Patent Document 4).

  These identification methods include going to health centers, animal hospitals, pet shops, etc. with pets, creating IDs and creating identification tags, writing pet identification IDs and owner information on microchips, and pet DNA. It is a method of collecting and analyzing.

  The method of using a microchip is costly for the cost of the microchip itself, a device for writing pet information and owner information on the microchip, and a device for reading the information of the microchip, etc. .

  Moreover, embedding a microchip in the body of a pet is hurting and painful for the individual pet.

In addition, there are dogs without collars, cats don't wear collars very much, and collecting nose prints is what makes cows and dogs nose dirty. There is a need for an alternative to using microchips and tags for pets such as cats and cats, and collecting noseprints. Furthermore, these identifications have to be performed at health centers, animal hospitals, pet shops, etc. to search for a label, read a microchip, collect DNA, and search pet information and owner information. As described above, in the conventional method, in order to obtain information for identifying a pet, it is necessary to bring the animal to a health center, an animal hospital, a pet shop, or the like. Alternatively, some special device or surgery is required. Therefore, it is not easy for a general pet owner to acquire information for identifying a pet.
JP 2003-263488 A JP 2002-10720 A JP 2004-357589 A JP-A-9-265523

  INDUSTRIAL APPLICABILITY The present invention can acquire information for identifying pets such as dogs and cats quickly and easily without damaging dogs and cats and without using special devices. The purpose is to provide a device. Moreover, it aims at provision of the pet management system using this apparatus.

The nose feature information generating device of the present invention comprises:
An image input unit for inputting a nose photographed image in which the entire nose including two nostrils is photographed;
The nose contour image is identified by identifying the outline of the nose from the nose photographed image input by the image input unit, and cutting out a nose contour image that is a nose image including two nostrils from the nose photographed image along the identified nose contour. A nose contour image extraction unit to extract;
It is determined whether or not the nose contour image extracted by the nose contour image extraction unit needs to be corrected. When it is determined that correction is necessary, the corrected nose contour image is corrected by performing the predetermined correction. A correction determination unit that outputs a nose contour image as a correction determined image without performing correction when it is determined that correction is not necessary,
An element extraction unit that extracts the outline of the nose and the outlines of the first and second nostrils, which are the two nostrils, from the correction determined image output by the correction determination unit;
The nasal reference position calculated by calculating the nasal reference position, which is a reference position used for specifying the shape of the nose outline extracted by the element extraction unit, by a predetermined nasal reference position calculation method and the element extraction unit extract Identify the shape of the nose from the contour of the nose
A first nostril reference calculated by calculating a first nostril reference position, which is a reference position used for specifying the contour shape of the first nostril extracted by the element extraction unit, by a predetermined first nostril reference position calculation method. Identifying the contour shape of the first nostril from the position and the contour of the first nostril extracted by the element extraction unit;
A second nostril reference calculated by calculating a second nostril reference position, which is a reference position used for specifying the contour shape of the second nostril extracted by the element extraction unit, by a predetermined second nostril reference position calculation method. An element shape specifying unit for specifying the contour shape of the second nostril from the position and the contour of the second nostril extracted by the element extracting unit;
A nose feature information generating unit configured to generate information including nose contour shape specified by the element shape specifying unit, contour shape of the first nostril, and contour shape of the second nostril as nose feature information; Features.

The element shape specifying part is
As a predetermined nose reference position calculation method, the center of gravity position of the figure of the contour shape determined by the nose outline extracted by the element extraction unit is calculated as the nose reference position,
As a first nostril reference position calculation method determined in advance, a center of gravity position of a figure having a contour shape determined by the outline of the first nostril extracted by the element extraction unit is calculated as a first nostril reference position;
As a predetermined second nostril reference position calculation method, the center of gravity position of a figure having a contour shape determined by the contour of the second nostril extracted by the element extraction unit is calculated and used as the second nostril reference position.

The image input unit
Enter a nose photographed image of the entire nose, including two nostrils, a nostril corresponding to each of the two nostrils, and a nose bridge,
The element extraction unit includes:
As components of the nose, the outline of the nose, the outlines of the two nostrils, the first nostril and the second nostril, the nose bridge part, the first nostril corresponding to the first nostril, and the second nostril Extract the corresponding second nose wing,
The element shape specifying part is
Calculate the nose reference position based on the nose bridge extracted by the element extraction unit as a predetermined nose reference position calculation method,
Calculating a first nostril reference position based on the first nostril extracted by the element extraction unit as a predetermined first nostril reference position calculating method;
As a predetermined second nostril reference position calculation method, the second nostril reference position is calculated based on the second nose wing extracted by the element extraction unit.

The element extraction unit includes:
Extract the nose bridge as a band, and extract each of the first and second nose wings as line segments,
The element shape specifying part is
As a predetermined nose reference position calculation method, a predetermined two points are set on the nose bridge portion extracted as a belt-like shape, a line segment passing through the two points, and both ends thereof are positioned on the nose contour. The longest line segment is created as a second line segment that is a line segment that is orthogonal to the first line segment and whose both ends are located on the nose contour, and the first and second line segments are created. Calculate the intersection with the minute and set it as the nose reference position,
As a predetermined first nostril reference position calculation method, a third line segment is created by sharing the outline of the first nostril and two intersections by extending the first nose wing extracted as a line segment. The longest line segment among the line segments that are orthogonal to each other and whose both ends are located on the outline of the first nostril is created as the fourth line segment, and the intersection of the third and fourth line segments is calculated. To the first nostril reference position,
As a predetermined second nostril reference position calculation method, a fifth line segment is created by sharing the outline of the second nostril and two intersection points by extending the second nose wing extracted as a line segment. The longest line segment is created as the sixth line segment among the line segments that are orthogonal to each other and whose both ends are located on the outline of the second nostril, and the intersection of the fifth and sixth line segments is calculated. And the second nostril reference position.

The element shape specifying part is
Identify the shape of the nasal bridge extracted by the element extraction unit,
The nose feature information generation unit
The nasal feature information is generated by including the shape of the nasal bridge specified by the element shape specifying unit in the nasal feature information.

The element shape specifying part is
The nose reference position is calculated as any position in the inner region of the nose contour, the first nostril reference position is calculated as any position in the inner region of the first nostril contour, and the second nostril reference When the position is calculated as any position of the inner region of the outline of the second nostril,
Lengths and directions of a plurality of line segments set by setting a plurality of line segments that start from the nose reference position in a plurality of different directions and have end points on the nose outline Identify the contour shape of the nose from the
Each of a plurality of line segments that are set by setting a plurality of line segments that are in a plurality of different directions starting from the first nostril reference position and whose end points exist on the outline of the first nostril From the length and direction, specify the contour shape of the first nostril,
Each of a plurality of line segments that are set by setting a plurality of line segments that are in a plurality of different directions starting from the second nostril reference position and whose end points are present on the outline of the second nostril The contour shape of the second nostril is specified from the length and direction.

The element shape specifying part is
36 lines set by setting 36 line segments starting from the nose reference position and extending in 36 directions that are different from each other at approximately 10 degrees and whose end points are present on the outline of the nose. Identify the contour shape of the nose from the length and direction of each minute,
The first nostril reference position is set as a starting point, and 36 line segments are set at approximately 10 degrees in different 36 directions and the end point is on the outline of the first nostril. Identify the contour shape of the first nostril from the length and direction of each of the 36 line segments,
36 line segments are set by setting 36 line segments starting from the second nostril reference position and extending in different 36 directions at intervals of approximately 10 degrees, with the end point existing on the outline of the second nostril. The contour shape of the second nostril is specified from the length and direction of each of the 36 line segments.

The element shape specifying part is
72 lines set by setting 72 line segments starting from the nose reference position and starting in the direction of 72 different from each other at intervals of about 5 degrees and having the end points existing on the nose outline. Identify the contour shape of the nose from the length and direction of each minute,
72 line segments are set by setting 72 line segments that start in the first nostril reference position and move in different 72 directions at intervals of about 5 degrees and whose end points are on the outline of the first nostril. Identify the contour shape of the first nostril from the length and direction of each of the 72 line segments,
72 line segments that are in the direction of 72 different from each other at approximately every 5 degrees starting from the second nostril reference position are set and set as 72 line segments whose end points exist on the outline of the second nostril. The contour shape of the second nostril is specified from the length and direction of each of the 72 line segments.

The nose feature information generating device further includes:
A nasal feature information registration unit for registering nasal feature information of a predetermined type of mammal;
Communication that receives transmission source information including a nose photographed image of the predetermined type of mammal and transmission source data indicating the transmission source from the transmission source, and outputs the nose imaging information included in the received transmission source information to the image input unit With
The nose feature information generation unit
The communication unit generates nose feature information based on the nose photographed image output to the image input unit, and the generated nose feature information is associated with the transmission source data included in the transmission source information received by the communication unit. The attached nose feature information is registered in the nose feature information registration unit.

The nose feature information registration unit
For the predetermined type of mammal, a plurality of associated nose feature information is registered,
The communication unit is
Determination request information for requesting determination as to whether or not any of the associated nose feature information registered in the nose feature information registration unit is matched, and transmitting determination request information including a nose photographed image related to the determination request Receiving determination request information from the requesting device, outputting a nose photographed image related to the received determination request to the image input unit,
The nose feature information generation unit
Generating determination request nose feature information that is nose feature information based on a nose photographed image related to the determination request output by the communication unit to the image input unit;
The nose feature information generating device further includes:
The determination request nose feature information generated by the nose feature information generation unit is set in advance by comparing each of the plurality of associated nose feature information registered in the nose feature information registration unit with a predetermined comparison method. And a nose feature information registration unit that searches the associated nose feature information from the nose feature information registration unit and returns a search result to the determination requesting device via the communication unit.

The search unit
As a result of the search, when the associated nose feature information can be retrieved from the nose feature information registration unit, transmission source data corresponding to the retrieved associated nose feature information is returned as a search result. .

The transmission source information received by the communication unit is:
Including individual feature data indicating characteristics of the individual mammal of the predetermined type related to the nose photograph image included in the transmission source information,
The nose feature information generation unit
The associated nose feature information is further registered in the nose feature information registration unit in association with the individual feature data,
The search unit
As a result of the search, when the associated nose feature information can be retrieved from the nose feature information registration unit, the source data and the individual feature data corresponding to the associated nose feature information are returned as a search result. And

The individual feature data received by the communication unit is:
A whole photographed image obtained by photographing the whole mammal of the predetermined type is included.

The nose feature information generation program of the present invention is
A process of inputting a nose photograph image in which the entire nose including two nostrils is photographed, which is an image of a mammal's nose;
A nose outline is identified by identifying a nose outline from the input nose photographed image, and extracting a nose outline image that is a nose outline image including two nostrils from the nose photographed image along the identified nose outline. Processing to
It is determined whether or not the extracted nose contour image needs to be corrected. If it is determined that correction is necessary, the corrected nose contour image is corrected and determined as a corrected image. A process of outputting a nose contour image as a corrected determination image without performing correction when it is determined that there is no need for correction;
A process of extracting the outline of the nose and the outlines of the two nostrils, the first nostril and the second nostril, from the output corrected and determined image,
A nose based on the nose reference position calculated by calculating a nose reference position, which is a reference position used for specifying the shape of the extracted nose outline, by a predetermined nose reference position calculation method and the extracted nose outline Identify the contour shape of
The first nostril reference position calculated by calculating the first nostril reference position, which is a reference position used for specifying the extracted contour shape of the first nostril, by a predetermined first nostril reference position calculation method is extracted. Identifying the contour shape of the first nostril from the contour of the first nostril,
A second nostril reference position calculated by calculating a second nostril reference position, which is a reference position used for specifying the contour shape of the extracted second nostril, by a predetermined second nostril reference position calculation method is extracted. A process of specifying the contour shape of the second nostril from the contour of the second nostril;
It is characterized by causing a computer to execute processing for generating information including the identified nose contour shape, the first nostril contour shape, and the second nostril contour shape as nose feature information.

According to the present invention, information for identifying pets such as dogs and cats can be acquired easily and quickly by a general pet owner.

Embodiment 1 FIG.
The first embodiment will be described with reference to FIGS. The first embodiment relates to a nasal feature information generation apparatus that generates nasal feature information indicating the characteristics of the nose of a mammal individual. In the following embodiments, a pet management apparatus will be described as an example of a nose feature information generation apparatus. The pet management apparatus described below takes a dog as an example of a mammal, but this is an example. The nasal feature information generation device can be applied to mammals including humans.

  Prior to the description of the pet management apparatus, the names of the nose parts will be described with reference to FIG. FIG. 1 is a diagram showing names of respective parts of the nose. FIG. 1 corresponds to the nose of the photograph (nose photographed image) of FIG. 4 described later. In FIG. 1, there is an eye at the top of the nose tip and a mouth at the bottom of the nose bridge. Of the two nostrils, the nostril on the left side of the figure is called the first nostril, and the right nostril is called the second nostril. There is a nasal bridge between the first nostril and the second nostril. A line segment extending from the first nostril and the second nostril is referred to as a nose wing. The nose wing of the first nostril is referred to as the first nose wing, and the nose wing of the second nostril is referred to as the second nose wing.

  FIG. 2 is a configuration diagram of the pet management apparatus 100 (an example of the nose feature information generation apparatus) in the first embodiment. The pet management apparatus 100 includes a pet information identification unit 110, a pet information storage unit 120 (nose feature information generation unit), a pet information search unit 130 (search unit), a communication unit 140, and a pet management database (DB) 150. (Nose feature information registration unit), a pet owner database (DB) 160, and a display unit 170. The pet information identification unit 110 includes an image input unit 111, a nose contour image extraction unit 112, a correction determination unit 113, an element extraction unit 114, and an element shape identification unit 115.

  The pet information identification unit 110 inputs an image of the nose of a dog (an example of a mammal), and specifies the contour shape of the nose, the contour shape of each of the two nostrils, and the like from this image.

  The pet information storage unit 120 (nose feature information generation unit) generates nose feature information based on the contour shape of the nose specified by the pet information identification unit 110, and stores the generated nose feature information in the pet management DB 150. sign up. The nose feature information will be described later with reference to FIG.

  The pet information search unit 130 (search unit) searches predetermined nose feature information from the nose feature information registered in the pet management DB 150 based on a search request from another device such as a client terminal. A detailed description of the pet information search unit 130 will be given in the second embodiment.

  The communication unit 140 communicates with other devices via a network.

  The pet management DB 150 (nose feature information registration unit) registers and stores the nose feature information of each of a plurality of dogs.

  The pet owner management DB 160 stores pet owner data associated with the nose feature information.

  The operation | movement which the pet management apparatus 100 produces | generates the nose characteristic information of a dog is demonstrated using FIG. FIG. 3 is a flowchart showing image processing for generating nose feature information by the pet management apparatus 100.

  First, the image input unit 111 of the pet management apparatus 100 inputs a front photograph (nose photographed image) in which the entire nose of the pet (dog) is photographed (S101). FIG. 4 shows a front photograph 20 of a dog as an example. The front photograph 20 is input by the image input unit 111 from another device via the communication unit 140. Alternatively, the image input unit 111 inputs from the photo reader 200. As described above, the image input unit 111 is an image of the nose of a dog (an example of a mammal) and the entire nose including two nostrils (nose photography) as shown in the front photograph 20 of FIG. Image).

  Next, the image input unit 111 displays the input front photograph (nose photographed image) on the display unit 170 and prompts the operator of the pet management apparatus 100 to designate the nose region 1 including the entire nose (S102). . The operator designates the nose region 1 with respect to the front photograph 20 displayed on the display unit 170. The frame in FIG. 4 shows the nose region 1 designated by the operator. FIG. 5 is a diagram showing the nose region 1.

  Next, the nose contour image extraction unit 112 identifies the nose contour from the nose region 1 and cuts out the “nasal contour image 3” including two nostrils from the nose region 1 along the identified nose contour. The nose contour image 3 ”is extracted (S103). FIG. 6 shows the nose contour image 3 cut out by the nose contour image extraction unit 112.

  The correction determination unit 113 determines whether or not the nose contour image 3 extracted by the nose contour image extraction unit 112 needs to be corrected. If it is determined that correction is necessary, the correction determination unit 113 performs the predetermined correction. Then, the corrected nose contour image is output as a “correction determined image”, and when it is determined that correction is not necessary, the nose contour image is output as a “correction determined image” without performing correction. Here, the predetermined correction is correction for correcting the nose contour image 3 to a certain size and processing so as to be in front by an image processing technique. This is to make it possible to compare the nose feature information generated based on different front photographs (nose photographed images) by correcting the nose contour image 3 to a constant size and a state seen from the front. It is. The correction determination unit 113 determines that the correction is not necessary when the nose contour image 3 is already a certain size and is in a state viewed from the front (S104).

  The element extraction unit 114, as a nose component, from the correction-determined image output by the correction determination unit 113, includes the outline of the nose, the outlines of the first and second nostrils, which are the two nostrils, and the nose. The bridge portion, the first nose wing and the second nose wing are extracted (S105).

  FIG. 7 shows the state extracted by the element extraction unit 114. In the first embodiment, the element extraction unit 114 extracts the nose bridge portion as a band and extracts each of the first nose wing and the second nose wing as a line segment.

  The element shape specifying unit 115 extracts the outline of the nose from the outline of the nose, the outline of the first nostril, the outline of the second nostril, the nose bridge part, the first nose wing, and the second nose wing extracted by the element extraction unit 114. The shape of the outline of the first nostril, the shape of the outline of the second nostril, and the shape of the nasal bridge portion are specified (S106). The nose contour shape specified by the element shape specifying unit 115 is referred to as “nasal shape information”. In addition, the nostril contour shapes identified by the element shape identifying unit 115 are referred to as “first nostril shape information” and “second nostril shape information”, respectively. Further, as the shape of the nasal bridge portion, the element shape specifying unit 115 compares (a) the width of the band, (b) the length of the band with respect to the nasal bridge portion extracted as a band shape by the element extraction unit 114, (c) The nasal bridge portion is specified by four positions: a substantially central position of the width of the upper part (shown in FIG. 7), and (d) a substantially central position of the width of the lower part (shown in FIG. 7). (A) to (d) specified by the element shape specifying unit 115 are referred to as “nasal bridge part information”. The details of specifying the contour shape of the nose and the contour shape of the nostril will be described later.

  The pet information storage unit 120 displays information including the nose contour shape specified by the element shape specifying unit 115, the first nostril contour shape, the second nostril contour shape, and the nasal bridge portion shape as "nose characteristics". Information ”is generated, and this“ nose feature information ”is registered in the pet management DB 150 (S107).

  FIG. 8 is a flowchart for explaining a process in which the element shape specifying unit 115 specifies the contour shape of the nose. With reference to FIG. 8, a case where the element shape specifying unit 115 specifies the contour shape of the nose will be described.

  FIG. 9 shows the outline of the nose extracted by the element extraction unit 114.

  The element shape specifying unit 115 first calculates a “nose reference position” that is a reference position used for specifying the outline shape of the nose extracted by the element extraction unit 114 by the following method (nasal reference position calculation method). . As shown in FIG. 9, the element shape specifying unit 115 sets two predetermined points, for example, a point b and point b, on the nasal bridge extracted by the element extraction unit 114 as a band shape. Then, a line segment 8 (first line segment) that passes through these two points and whose both ends are located on the outline of the nose is created (S201).

  Next, as shown in FIG. 10, the element shape specifying unit 115 determines the longest line segment among the line segments orthogonal to the line segment 8 and having both ends located on the nose contour as the line segment 9 (first segment). 2 lines). Then, an intersection point 10 between the line segment 8 and the line segment 9 is calculated, and this intersection point 10 is set as a “nose reference position” (S202).

  As shown in FIG. 11, the element shape specifying unit 115 differs from each other by approximately 10 degrees with respect to a line segment whose start point is the intersection point 10 (nose reference position) and whose end point is on the nose outline, with the angle reference line as a reference. 36 are set in 36 directions. Then, the contour shape of the nose is specified from the length and direction of each of the set 36 line segments. Hereinafter, the length of the 36 line segments of the nose may be referred to as “length in 36 direction”. Note that the value of “36 lines approximately every 10 degrees” is an example. As long as the outline shape of the nose can be specified, “360 / n (book) every approximately n degrees” may be used. For example, “every 5 degrees, 360/5 = 72 (books)” may be set (S203).

  Next, a case where the element shape specifying unit 115 specifies the outline shape of the first nostril will be described with reference to FIG. FIG. 12 is a flowchart illustrating an operation in which the element shape specifying unit 115 specifies the contour shape of the first nostril. The specification of the contour shape of the second nostril is the same as in the case of the first nostril, and is therefore omitted.

  The element shape specifying unit 115 is a “first nostril reference position” which is a reference position used for specifying the contour shape of the first nostril extracted by the element extracting unit 114 by the following method (first nostril reference position calculating method). Is calculated. As shown in FIG. 13, the element shape specifying unit 115 is a portion in contact with the outline of the first nostril extracted by the element extracting unit 114, and the element extracting unit 114 extracts it as a line segment. The first nose wing 11 is identified (S301).

  As shown in FIG. 14, the element shape specifying unit 115 extends the first nose wing 11 to create a line segment 13 (third line segment) sharing the outline of the first nostril and the two intersections (S302). .

  Next, as shown in FIG. 15, the element shape specifying unit 115 determines the longest line segment among the line segments orthogonal to the line segment 13 and having both ends positioned on the outline of the first nostril. 14 (fourth line segment). And the intersection 15 of the line segment 13 and the line segment 14 is calculated, and this intersection 15 is made into the "1st nostril reference position" (S303). The element shape specifying unit 115 also applies the second nostril to the second nostril from the line segment corresponding to the line segment 13 (fifth line segment) and the line segment corresponding to the line segment 14 (sixth line segment). A second nostril reference position, which is a reference position used for specifying the contour shape of the second nostril, is calculated.

  Next, as shown in FIG. 16, the element shape specifying unit 115 includes 36 line segments that start at the intersection 15 (first nostril reference position) and go to different 36 directions at approximately 10 degrees. Thus, 36 line segments whose end points are present on the outline of the first nostril are set. Then, the contour shape of the first nostril is specified from the length and direction of each of the set 36 line segments. Hereinafter, the 36 lengths of the first nostril may be referred to as “the length of the first nostril in the 36 direction”. Similarly, the second nostril may also be referred to as “the length of the second nostril in the 36 direction”. Note that the value of “36 lines approximately every 10 degrees” is an example. As long as the contour shape of the first nostril can be specified, “360 / n (book) every approximately n degrees” may be used. For example, “every 5 degrees, 360/5 = 72 (books)” may be set (S304).

FIG. 17 is a diagram illustrating the contents of the nose feature information generated by the pet information storage unit 120 and registered in the pet management DB 150. The nose feature information includes nose shape information, first nostril shape information, second nostril shape information, and nasal bridge portion information regarding “nose”.
(1) The nose shape information is the length and angle (direction) of a line segment in 36 directions specified by the element shape specifying unit 115 with respect to the outline of the nose.
(2) The first nostril shape information is the 36-direction length of the first nostril identified by the element shape identifying unit 115 with respect to the outline of the first nostril, and the angle (direction) of the line segment.
(3) The second nostril shape information is the 36-direction length specified by the element shape specifying unit 115 and the angle (direction) of the outline of the second nostril.
(4) Nasal bridge information is the length, width, upper position, and lower position of the nose bridge extracted as a band.

  In S202 of FIG. 8 and S303 of FIG. 12, the element shape specifying unit 115 obtains the line segment 8 and the line segment 9, calculates the intersection point 10 as the nose reference position, and the line segment 13 and the line segment 9. The minutes 14 were calculated and the intersection 15 was calculated as the first nostril reference position. However, the calculation method of these reference positions is an example. In addition to these, the element shape specifying unit 115 calculates the barycentric position of the contour-shaped figure determined by the nose contour extracted by the element extraction unit 114 to obtain the nose reference position, and the first nostril extracted by the element extraction unit 114 The centroid position of the contour-shaped figure determined by the contour is calculated as the first nostril reference position, and the centroid position of the contour-shaped figure determined by the contour of the second nostril extracted by the element extraction unit 114 is calculated to calculate the second nostril It does not matter as the reference position.

  According to the pet management apparatus of the present embodiment, a pet owner such as a dog or cat can be identified quickly and easily without damaging dogs and cats and without using a special apparatus. Nose feature information can be acquired.

Embodiment 2. FIG.
Next, Embodiment 2 will be described with reference to FIGS. The second embodiment relates to a pet management system 1000 including the pet management apparatus 100 described in the first embodiment.

  FIG. 18 is a diagram illustrating a configuration of the pet management system 1000 according to the second embodiment. The pet management system 1000 includes a pet management apparatus 100 to which a photo reader 200 is connected, a mobile phone 401, a client terminal 402, a client terminal 403, and a client LAN 410 that are communicably connected to each other via the Internet 300. Yes. Client terminals 411 and 412 are connected to the client LAN 410, and these can communicate with the pet management apparatus 100 via the Internet 300.

  Registration of later-described pet identification information including nose characteristic information by transmitting a front view photograph (nose photographed image) of the dog (an example of a predetermined mammal) from the client terminal or mobile phone to the pet management apparatus 100 Can be requested. In addition, by transmitting a nose shot image or the like from the client terminal or the mobile phone to the pet management apparatus 100, a search for whether the pet management apparatus 100 has pet identification information regarding the dog that is the subject may be requested. it can.

FIG. 19 is a diagram showing pet identification information. “Pet identification information” is information for identifying a pet. The pet identification information is registered in the pet management DB 150 of the pet management apparatus 100. The nose feature information is registered in the pet management DB 150 by being included in the pet identification information. As shown in FIG. 19, the pet identification information is created for each individual and managed for each type of pet. That is, FIG. 19 shows pet identification information regarding a dog. The pet identification information 30a is information regarding a certain dog, the pet identification information 30b is information regarding a certain other dog, and the pet identification information 30c is information regarding a certain other dog.
The pet identification information 30a is
(1) “NO (number)” for distinguishing pet identification information;
(2) Overall photo and front photo of the pet (nose shot image)
(3) Pet feature information indicating pet features such as type, size, gender, fur,
(4) Owner ID that identifies the owner,
(5) Registration applicant ID indicating the person who applied for registration of the pet identification information;
(6) including nose feature information. (2) The whole pet photograph and the front photograph (nose photographed image) and (3) pet characteristic information are individual characteristic data indicating the characteristics of the pet individual.

  FIG. 20 is a diagram illustrating an example of the pet owner information 40a. The pet owner information is created for each owner. The pet owner information 40a is stored in the pet owner management DB 160. The “owner ID” of the pet identification information 30a corresponds to the “owner ID” of the pet owner information 40a. Thereby, the pet identification information 30a is associated with the pet owner information 40a. As shown in FIG. 20, the pet owner information 40 a includes an owner ID, owner name, owner address, owner e-mail address, telephone number, and the like.

  FIG. 21 shows an example of the registration applicant information 50a. If the person who applied for (requests) registration of pet identification information from the pet management apparatus 100 is not the owner, the person is registered as a registration applicant. The registration applicant information is stored in the pet owner management DB 160. If the pet identification information registration applicant is a pet owner, it is registered in the pet owner information. The “registration applicant ID” of the pet identification information 30a corresponds to the “registration applicant ID” of the registration applicant information 50a. Thereby, the pet identification information 30a is associated with the registration applicant information 50a. As illustrated in FIG. 21, the registration applicant information 50 a includes a registration applicant ID, a name, an address of the registration applicant, an e-mail address, a telephone number, and the like as the contact information of the registration applicant. Registered as registration applicant information is assumed to be health centers, pet shops, animal hospitals, and the like.

  FIG. 22 is a sequence diagram showing exchanges between the “pet owner side terminal” of the pet owner, the pet management apparatus 100, and the “health center side terminal” of the health center. The “pet owner side terminal” is a registration requesting side that requests registration of pet identification information. The “health center terminal” is a search requester that requests a search. In FIG. 18, the “owner side terminal” is, for example, the mobile phone 401, and the “health center terminal” is the client terminal 411.

  The pet owner who is the registration request side of the pet identification information sends the pet (dog) owner side photo data (nose photographed image) from the mobile phone 401 via the Internet 300, and the name and contact information shown in FIG. Owner information (an example of transmission source data) and “registration request information” (an example of transmission source information) including an entire pet photo (entire captured image) and pet characteristic information are transmitted to the pet management apparatus 100 (S401). .

  The pet information identification unit 110 performs the operation described in the first embodiment, and the “owner side photo data” (nose photographed image) of the pet included in the “registration request information” transmitted from the mobile phone 401 via the Internet 300. Based on the above, the “nose contour shape”, “contour shape of the first nostril”, “contour shape of the second nostril”, “shape of the nasal bridge”, etc. are specified (S402).

As described in the first embodiment, the pet information storage unit 120 (nose feature information generation unit) has the “nose outline shape”, “first nostral outline shape”, “ Nose feature feature information including “nose nose contour shape”, “nasal bridge shape” and the like is generated, and this nose feature feature information is registered as a part of pet identification information (S403). When the nose feature feature information is registered, as shown in FIG. 19, the pet information storage unit 120 displays the nose feature feature information as (1) an entire photograph and a front photograph (nose photographed image),
(3) Pet characteristic information,
(4) Owner ID
At the same time, it is registered in the pet management DB 150 as “pet identification information”.

  When the registration requesting side is the pet owner in this way, the pet information storage unit 120 associates the “pet owner information” included in the registration request information with the “pet identification information” and registers it in the pet owner management DB 4 ( S404). Specifically, as shown in FIG. 19, the pet information storage unit 120 provides an item “owner ID” in the pet identification information 30a and sets the owner ID. In the pet identification information 30a, the “owner ID” is “1001”. This “owner ID” is associated with the “owner ID” of the pet owner information 40a shown in FIG. In this way, “pet owner information” is associated with “pet identification information” by “owner ID”. Therefore, one of the “pet identification information” and the “pet owner information” can be specified. Therefore, since “pet identification information” always includes nose feature information, nose feature information (associated nose feature information) is associated with “pet owner information”.

  In addition, health centers, animal hospitals, pet shops, etc. other than pet owners may be registration requesters. In this case, the pet information storage unit 120 records the health center, the animal hospital, the pet shop, etc. with the “registration applicant ID” in FIG. 19 as in the case of the “pet owner”. “Pet identification information” is associated with the registration applicant information of FIG. The pet information storage unit 120 registers the registration applicant information in the pet owner management DB 4 (S404). The above S401 to S404 are operations in the process in which the pet management apparatus 100 registers pet identification information in response to a request from the registration request side.

  Next, the search request process will be described with reference to S405 to S407.

(Search request process)
For example, for a dog, the pet management apparatus 100 registers and stores a plurality of “pet identification information” in the pet management DB 150 as shown in FIG. The health center transmits search request information (determination request information) from the client terminal 411 (determination request device) (S405). The search request information includes health center photo data (nose photographed image), health center information including the name, address, and contact information of the health center, pet feature information indicating the characteristics of the dog, and individual feature data that is a photograph. The health center side makes a search request as a search request side in the following cases. In the health center, when a dog (an example) is brought in, in order to find the owner of the dog, a picture of the dog's nose (nose photographed image) is taken, and the photograph taken is taken from the client terminal 411. The search request information is included in the search request information and transmitted to the pet management apparatus 100 via the Internet 300.

  When the pet management apparatus 100 receives the search request information, the pet information identification unit 110 is transmitted from the client terminal 411 via the Internet 300 by the operation described in the first embodiment, as in the case of S402. Based on the health center photo data (nose image) included in the search request information, the dog's “nose outline shape”, “first nostril outline shape”, and “second nostril outline” related to the search request The pet information storage unit 120 identifies “shape”, “shape of the nose bridge”, etc., and the pet information storage unit 120 performs “nose contour shape”, “contour shape of the first nostril”, “second nostril” for the dog related to the search request. Nose feature information (determination request nose feature information) including “contour shape”, “shape of nasal bridge portion”, and the like.

  The pet information retrieval unit 130 (retrieval unit) obtains nasal feature information (determination request nose feature information) related to the retrieval request generated by the pet information storage unit 120, and a plurality of nasal feature information (corresponding to the correspondence). Each of the artificial nose characteristic information) is compared with a predetermined predetermined comparison method. By this comparison, the pet information search unit 130 searches the pet management DB 150 for nose feature information based on a preset search criterion. The pet information search unit 130 searches the pet management DB 150 and extracts pet identification information whose “pet identification information” (nose feature information) matches or approximates within a certain range (S406). This search process will be described later with reference to FIGS.

  The pet information search unit 130 returns the search result to the client terminal 411 of the health center via the communication unit 140 (S407). As a search result, the pet information search unit 130 causes the client terminal 411 to display a front view of the pet (nose photographed image), a whole pet photo, pet feature information, If the owner information exists, the pet owner information is extracted from the pet owner management DB 160 and returned.

  If the health center looking for the owner of the dog brought in determines that it is a registered pet (dog) from the returned information, it will contact the owner of the associated pet and come to pick it up. That's fine.

  As described above, in the case of FIG. 22, the pet owner on the registration request side uses the Internet 300 from his / her personal computer or mobile phone to send a photo of the pet to the pet management apparatus 100, and the pet management apparatus A case has been described in which “nose feature information” of a pet is extracted at 100 and registered in the pet management DB 150 in advance as “pet identification information”. Further, the case where “pet owner information” is associated with “pet identification information” and registered in the pet owner management DB 160 has been described. When a pet is brought into a health center (others may be an animal hospital, pet shop, etc.), the health center (or animal hospital, pet shop, etc.) takes a picture of the pet's nose and requests a search. As a side, a photograph taken from its own client terminal is transmitted to the pet management apparatus 100 using the Internet 300. The pet management apparatus 100 extracts the “nose feature information” of the photo according to the search request, and searches the pet management DB 150 using the extracted “nose feature information” to explain that the pet owner can be found. did. In this case, if the owner is not found, “pet identification information” related to the search request is registered in the pet management DB 150.

  In some cases, the pet owner is the search requester. This is the case. When the owner determines that his / her pet has been lost or stolen, the search request side uses the Internet 300 from his / her personal computer or mobile phone to take a picture of the pet (nose shot image). ), Pet characteristic information and the like are transmitted to the pet management apparatus 100, and the pet management apparatus 100 generates "nose characteristic information" of the pet and searches the pet management DB 150 based on the nose characteristic information. As a result, it is possible to search for a pet that is lost or stolen. If the search requesting side is a pet owner and it is determined from the reply from the pet management apparatus 100 that the pet is his / her pet, it can be stored in a linked storage location (such as a health center, animal hospital, or pet shop). Just go to pick up. On the other hand, if no hit is found as a result of the search, the pet management apparatus 100 registers “pet identification information” related to the search request used in the search in the pet management DB 150, and if it exists, The attached pet owner is registered in the pet owner management DB 160.

  Next, the contents of the search process in S406 of FIG. 22 will be described with reference to FIGS. FIG. 23 is a diagram showing search processing by the pet information search unit 130. The nose feature information on the search request side in FIG. 23 is based on the front photograph (nose photographed image) of the pet related to the search request (determination request), the image input unit 111, the nose contour image extraction unit 112, and the correction determination unit 113. The pet information identifying unit 110 including the element extracting unit 114 and the element shape identifying unit 115 identifies the contour shape of the nose, the contour shape of the first nostril, and the contour shape of the second nostril. The information is generated by the pet information storage unit 120 based on the outline shape of the nose. When there is a search request, the pet information search unit 130 stores the search request side nose feature information (determination request nose feature information) in advance with each of the nose feature information of the plurality of pet identification information registered in the pet management DB 150. Comparison is made by a predetermined predetermined comparison method. Then, the nose feature information is searched from the pet management DB 150 based on the preset search criteria, and the search result is returned to the search requesting terminal via the communication unit 140. That is, in FIG. 23, the pet information search unit 130 compares the search request side nose feature information with each of the nose feature information 1, the nose feature information 2,. If the comparison result indicates that there is a piece that matches the search criteria among the nose feature information 1 to the nose feature information n, the pet information search unit 130 selects pet identification information including the nose feature information. In this way, the pet information search unit 130 searches for pet identification information using the nose feature information as a key.

24-36 is a figure explaining the comparison method by the pet information search part 130. FIG. The pet information search unit 130 performs comparison using a predetermined comparison method set in advance. The following four items (1) to (4) will be described as examples of the comparison method. The pet information search unit 130 can search by using any of these four.
(1) Comparison method for comparing sizes,
(2) A comparison method using a linear model by the method of least squares,
(3) A comparison method using a polynomial model by the method of least squares,
(4) A comparison method using a Fourier series.

  These comparison methods for searching can be performed independently, but in order to improve the search accuracy, the search accuracy can be improved by combining the method (1) with another method. . These comparison methods compare and determine the shape of the nose of the pet (mammal) and the shape of the nostril. To search for individual pets, add a pet type, fur, color, size, pet owner information, etc., and adopt a method that considers that individual pets match when each item matches. The search accuracy can be improved.

(1. Comparison method for comparing sizes)
(1.1) The first comparison method among the “comparison methods for comparing sizes” will be described with reference to FIG. The first comparison method compares and matches all of the “nose contour shape”, “first nostril contour shape”, and “second nostril contour shape” (length from each center: 36 × 3). This is a method of checking and determining whether or not. That is,
36-direction length of the "nose contour shape"
36-direction length of “contour shape of first nostril”,
This is a method of comparing 108 sets of corresponding lengths of “36 × 3 = 108 (books)” in the 36-direction length of the “second nostril shape”. For example, in FIG. 23, a case is assumed where the search request side nose feature information is compared with the nose feature information 1 included in the pet identification information 1. In this case, as shown in FIG. 24, in this comparison method, the search request side nose feature information and the nose feature information 1 have 108 sets of 36 × 3 = 108 corresponding “lengths in 36 directions”. Ratios are obtained for all, and if the ratios are within a certain range (for example, within 95%), they are regarded as matching. Further, for all 108 sets, if a certain number of sets or more (for example, 100 sets or more: allowable number) match, the noses of the animals being compared are the same nose (an example of search criteria) and search The pet identification information 1 is selected on the assumption that the animals are the same individual.
(1.2) Next, the second comparison method among the “comparison methods for comparing sizes” will be described with reference to FIG. In the second comparison method, “the size of the nose”, “the size of the first nostril”, and “the size of the second nostril” are individually compared, and all of these three sizes are the same. Is determined to be the same individual. That is, in FIG. 23, assuming that the search request side nose feature information is compared with the nose feature information 1 included in the pet identification information 1,
In search request side nose feature information and nose feature information 1,
Compare the 36 lengths corresponding to each other's "nose contour shape",
Compare the 36-direction lengths corresponding to each other's "contour shape of the first nostril",
The 36-direction lengths corresponding to each other's “contour shape of the second nostril” are compared.
As for each “nose outline shape”, a reference for determining that each case matches is provided in the same manner as in the first method. FIG. 25 shows a case of “nose outline shape”.
The pet information search unit 130 determines that they match.
"Nose contour shape matches"&"First nostril contour shape matches"&"Second nostril contour shape matches"
And when all three are determined to be coincident. In this case, the pet information search unit 130
Pet identification information 1 including nose feature information 1 is selected.

(2. Comparison method using linear model by least square method)
Next, a comparison method for obtaining “identification information” of the size of the nose of the pet and the size of the nostril by regression analysis using the least square method will be described with reference to FIGS. The dog nose in FIG. 26 will be described as an example. In FIG. 26, when each of the lengths in the 36 directions from the center of the nose (nasal reference position) is plotted as a point, a model (graph) shown in FIG. 27 is obtained. That is, the horizontal axis represents the “angle” of the length in each 36 direction, and the vertical axis represents the “length” of each length in the 36 direction. Let us consider expressing this model (graph) by a mathematical expression. The angle is an explanatory variable (x), and the length value is an objective variable (y). A predicted value of y with respect to x is obtained. The prediction formula of y is a straight line,
y = ax + b
And In this case, the regression coefficients a and b that minimize the square sum of the difference between the actual measurement value yi and the predicted value yj (= axi + b) may be obtained. There are three types of models (graphs): “nose shape”, “first nostril shape”, and “second nostril shape”. (Nose), b (nose), etc., and “a, b” etc. for “the shape of the first nostril” are a (first nostril), b (first nostril), etc., and “the shape of the second nostril” ", A, b, etc." are represented as a (second nostril), b (second nostril), and the like. As shown in FIG. 28, the number N of actually measured values, the actually measured value of the i-th x as xi, the actually measured value of y with respect to xi as yi, the slope of the straight line as a, and the intercept of the straight line as b. In this case, a and b can be obtained by the equation shown in FIG. This means that the coefficient of the predicted value is calculated such that a and b minimize the square of the difference between the predicted value and the actually measured value. The inclination a (nose) and the segment b (nose) obtained by the equation of FIG. 29 are set as “identification information” for identifying the “nose size” of the pet (dog). For the first nostril and the second nostril, the pet information search unit 130 obtains a (first nostril), b (first nostril) and a (second nostril), b (second nostril). If the slopes a and intercepts b of all three types of nose, first nostril, and second nostril are all within a certain error (eg, within 3 pixels, within 0.3 mm, etc .: allowable value). Assume that the animals being searched are the same individual. That is, in FIG. 23, assuming that the search request side nose feature information is compared with the nose feature information 1 included in the pet identification information 1, a (nose) and b obtained from the search request side nose feature information. A (nose), a (first nostril), b (first nostril), a (second nostril), b (second nostril) and a (nose), b (nose), a obtained from nose feature information 1 When the corresponding inclination a and the corresponding intercepts b are within a certain error with respect to (first nostril), b (first nostril), a (second nostril), b (second nostril), pet information The search unit 130 selects the pet identification information 1 including the nose feature information 1.

(3. Comparison method using polynomial model by least square method)
Next, a comparison method for obtaining “identification information” of the pet nose size and nostril size by approximation of a polynomial model by the least square method will be described with reference to FIGS. As in the case of FIG. 27, plotting each of the lengths in the 36 directions from the center (nose reference position) of the pet's nose as a point, the model (graph) shown in FIG. 30 is obtained. Consider a polynomial that approximates the model (graph) of FIG. For example, if approximated by a quadratic equation, (Expression 1) to (Expression 3) shown in FIG. 31 are obtained. When partial differentiation is performed in (Equation 3) in FIG. 31, a, b, and c can be obtained from the equation shown in FIG. The a (nose), b (nose), and c (nose) obtained from the equation shown in FIG. 32 are used as identification information of the “nose size” of the pet. The “size of the first nostril” and the “size of the second nostril” are also the same as described in “2. Comparison method using the first-order model by the least square method”. A total of three types a, b, and c of “the size of the nose”, “the size of the first nostril”, and “the size of the second nostril” are all constant errors (eg, within 3 pixels or 0. 0). If it is within 3 mm, etc. (referred to as an allowable value), the searched animals are assumed to be the same individual. The number of dimensions may be determined as appropriate. In the case of the following five-dimensional expression, a, b, c, d, e, and f may be obtained in the same manner.
y = ax ⁵ + bx ⁴ + cX ³ + dx ² + ex + f

(4. Comparison method using Fourier series)
Next, a comparison method for obtaining “identification information” of the size of the pet's nose and the size of the nostril based on the Fourier series model of the function of the period 2π will be described with reference to FIGS. In the example of the nose of the pet (dog) in FIG. 33, the length from the center of the nose of the pet (nose reference position) to the outline of the nose (the length in the 36 direction) is represented by a point taking the X axis and the Y axis. A model (graph) shown in FIG. 34 is obtained. It can be seen that the model (graph) in FIG. 34 is a function of period 2π. This periodic function f (x) can be expressed as (Equation 4) shown in FIG. 35, when the coefficients a1, b2,..., B1,... Are obtained by using the orthogonal relationship, (Expression 5) to (Expression 9) in FIG. In the case of an even function, it can be obtained from (Expression 5) to (Expression 7). In the case of an odd function, it can be obtained from (Expression 8) to (Expression 9). The a ₀ , a _x , and b _x obtained for each of “the size of the nose” and “the size of the nostril” are used as pet identification information. If the obtained “identification information” is within a certain error, it is assumed that the size of the nose of the pet to be compared, the size of the first nostril, and the size of the second nostril are the same. Furthermore, when the pet information search unit 130 determines that the size of the nose, the size of the first nostril, and the size of the second nostril match, the pets to be compared are determined to be the same individual. .

  The pet management system of the second embodiment is a system that manages pets and their owners in a DB (database) based on identification information of noses such as dogs and cats. Since the pet's nose characteristic information is extracted from a digitized photograph, a photograph from a digital camera or a camera attached to a mobile phone may be input. Therefore, if you can use the Internet from your owner's mobile phone or personal computer, or a personal computer such as a health center, animal hospital, or pet shop, you can register and search for pets, and dogs that are lost or stolen. It is very convenient for managing pets such as cats and cats.

  In addition, any animal having a nose can be managed. Therefore, it can also be used for pet management at pet shops and online sales systems. Furthermore, the present invention can be applied to the management of animals in zoos and individual management of animals that naturally inhabit.

  This nose authentication method can also be used for human personal authentication.

  The pet management system of the second embodiment is a system that manages pets and their owners by registering pet nose identification information in a DB. Therefore, if you can handle digital photos and can use the Internet, you can register pet information in the DB and search for lost or stolen pets at low cost without using special devices or equipment. Find owners quickly. In addition, when the owner is searching for a lost pet or a stolen pet, the place where the lost pet is protected can be found. Therefore, when a pet is brought into a health center, an animal hospital, a pet shop, or the like due to a lost child, theft, a traffic accident, or the like, the owner can be quickly searched for and delivered.

Embodiment 3 FIG.
The third embodiment will be described with reference to FIGS. The third embodiment is an embodiment in which the operation of the pet management apparatus 100 according to the first embodiment is implemented by a program and a recording medium on which the program is recorded.

  In the first embodiment, the operations of the constituent elements shown as “to” in the pet management apparatus 100 are related to each other, and are replaced with a series of processes to be performed by the computer by considering the relationship of the operations. Can do. By replacing the operation of each component with a series of processes, the embodiment of the nose feature information generation program can be obtained. Further, by recording this nose feature information generation program on a computer-readable recording medium, an embodiment of a computer-readable recording medium on which the program is recorded can be obtained.

FIG. 37 shows (1) the operation of the image input unit 111 of the pet management apparatus 100 shown in FIG.
(2) operation of the nose contour image extraction unit 112;
(3) Operation of the correction determination unit 113,
(4) operation of the element extraction unit 114;
(5) Operation of the element shape specifying unit 115,
(6) Replacing a series of operations called operations of the pet information storage unit 116 (nose feature information generation unit) with a series of processes for causing a computer to execute,
It is the flowchart made into embodiment of a nose characteristic information generation program.

  S501 is a process of inputting a nose photograph image in which the entire nose including two nostrils is photographed, which is an image of a mammal's nose.

  S502 identifies a nose outline from the input nose photographed image, and cuts out a nose outline image that is a nose image including two nostrils from the nose photographed image along the identified nose outline. This is a process of extracting an image.

  In step S503, it is determined whether or not the extracted nose contour image needs to be corrected. If it is determined that correction is necessary, the corrected nose contour image is corrected and determined. This is a process of outputting a nose contour image as a corrected and determined image without performing correction when it is determined that there is no need for correction.

  S504 is a process of extracting the outline of the nose and the outlines of the two nostrils, the first nostril and the second nostril, from the output correction-determined image.

  In step S505, the nasal reference position calculated by calculating a nasal reference position, which is a reference position used to specify the shape of the extracted nasal contour, by a predetermined nasal reference position calculation method and the extracted nasal contour The nose contour shape is identified from the first nostril reference position, which is a reference position used to identify the extracted first nostril contour shape, by a predetermined first nostril reference position calculation method. The first nostril reference position and the extracted outline of the first nostril are used to specify the outline shape of the first nostril, and the second nostril is a reference position used for specifying the extracted outline shape of the second nostril. This is a process for specifying the outline shape of the second nostril from the second nostril reference position calculated by calculating the reference position using a predetermined second nostril reference position calculation method and the extracted outline of the second nostril.

  S506 is a process of generating information including the identified nose contour shape, the first nostril contour shape, and the second nostril contour shape as the nose feature information.

  The embodiment of the program and the embodiment of the computer-readable recording medium on which the program is recorded can be configured by a program operable by a computer.

  FIG. 38 shows the external appearance of a computer system 800 that implements the functions of the pet management apparatus 100 of the first embodiment shown in FIG. 38, a computer system 800 includes a system unit 830, a liquid crystal display 813, a keyboard (K / B) 814, a mouse 815, a photo reader 818, and a printer 819, which are connected by a cable. The computer system 800 is connected to a mobile phone 401, a client terminal 402, and the like.

  FIG. 39 is a hardware configuration diagram of the computer system 800 of FIG. In FIG. 39, the computer system 800 includes a CPU (Central Processing Unit) 810 that executes a program. The CPU 810 is connected to the ROM 811, RAM 812, liquid crystal display 813, K / B 814, mouse 815, communication board 816, FDD 817, photo reader 818, printer 819, magnetic disk device 820, compact disk drive (CDD) 826 via the bus 825. It is connected.

  The magnetic disk device 820 stores an operating system (OS) 821, a window system 822, a program group 823, and a file group 824. The program group 823 is executed by the CPU 810, the OS 821, and the window system 822.

  The program group 823 stores a program for executing the function described as “˜unit” in the description of the first embodiment. The program is read and executed by the CPU 810.

  In the file group 824, the information of the pet management DB 150 and the pet owner management DB 160 described in FIG. 2 is stored as a file.

  2 described in FIG. 2 of the first embodiment may be realized by firmware stored in the ROM 811. Alternatively, it may be implemented by software alone, hardware alone, a combination of software and hardware, or a combination of firmware.

  Each processing in the embodiment of the program and the embodiment of the computer-readable recording medium on which the program is recorded is executed by the program, and this program is recorded in the program group 823 as described above. Then, the program is read from the program group 823 into the CPU 810, and each process of the program is executed by the CPU 810.

  The software or program may be executed by firmware stored in the ROM 811. Alternatively, the program may be executed by a combination of software, firmware, and hardware.

  In the above embodiment, the nose is extracted from the front photo of the pet, the contour of the pet's nose is identified, the groove in the middle of the pet's nose is identified from the identified contour, and the center of the pet's nose is identified from this groove. Determine the length and angle of each nose from the center in 36 directions, then identify the left and right nostrils of the pet, identify the left and right nostrils, The center is obtained, the length from the obtained center to the outline of the nostril and the respective angles in 36 directions, and the obtained nose size information (the length in the 36 directions from the center of the nose to the outline and the respective angles) And nose groove information (groove length, groove width, groove upper position, groove lower position), and right and left nostril size information (nose hole center to nose hole) Pet's nose characteristics with 36 directions length to each contour and each angle) It described the information generating apparatus.

  In the above embodiment, the pet nose characteristic information generating apparatus having the method for identifying the outline of the pet nose from the front photo of the pet has been described.

  In the above embodiment, the pet nose characteristic information generating apparatus having the method for identifying the groove in the middle of the pet nose from the identified contour has been described.

  In the above embodiment, the pet nose feature information generating apparatus having the method of obtaining the center of the pet's nose from the groove in the middle of the identified nose has been described.

  In the above embodiment, the nose characteristic information generation apparatus for a pet having the method for obtaining the length from the identified nose center to the nose contour in 36 directions has been described.

  In the above embodiment, the pet nose characteristic information generation apparatus having the method for identifying the right and left nostrils of the pet has been described.

  In the above embodiment, the pet nose feature information generation apparatus having the method of obtaining the center of the right and left nostrils of the pet from the identified pet nose contour has been described.

  In the above embodiment, the nasal feature information generation apparatus having the method for obtaining the length from the center of the obtained pet nose to the outline of the nostril in 36 directions has been described.

  In the above embodiment, the line from the position below the identified groove to the point where it intersects the contour of the pet's nose is obtained. From the top to the bottom, the line is extended to the right and left at right angles to the extension game of the groove, and the length to the point where it intersects with the outline of the nose is obtained. A description has been given of a nasal feature information generating device in which the point where the longest line among the obtained lines intersects with the line extending from the bottom of the nose to the top intersects with the center of the pet's nose.

  In the above embodiment, the nose feature information generation device is obtained in which the length from the center of the pet's nose to the point where it intersects with the contour of the pet's nose is obtained in 36 directions and the obtained length and angle are the size of the pet's nose. explained.

  In the above-described embodiment, for the identified left and right nostrils, the portion in contact with the identified pet nose contour is obtained. A groove-like portion (nose wing) is obtained from the portion in contact with the determined nose contour toward the nostril. Find the longest line in the nostril from the contact point of the nose. Find the line that intersects the line at right angles on the calculated line in the nostril. A description has been given of a nasal feature information generation device in which the intersection of the longest line and the groove obtained above is the center of the nostril.

  In the above embodiment, the length from the center of the determined nostril to the point where it intersects the contour of the pet's nostril in 36 directions is determined, and the nose with the determined length and angle as the size of the pet's nostril A feature information generation apparatus has been described.

  In the above embodiment, the description has been given of the nasal feature information generation apparatus using the obtained pet nose size information, nasal groove information, and left and right nostril size information as pet nose identification information.

FIG. 4 is a diagram for describing names of respective nose portions in the first embodiment. The structure of the pet management apparatus 100 in Embodiment 1 is shown. 3 is a flowchart showing image processing by the pet management apparatus 100 according to the first embodiment. The front photograph of the pet in Embodiment 1 is shown. The nose area | region cut out from the front photograph in Embodiment 1 is shown. 3 shows a nose outline image in the first embodiment. The element which the nose outline image extraction part 112 in Embodiment 1 extracted is shown. 5 is a flowchart of a process of specifying a nose contour shape by the element shape specifying unit 115 according to the first embodiment. 6 is a diagram for describing identification of a nose contour shape by an element shape identifying unit 115 according to Embodiment 1. FIG. 6 is a diagram for describing identification of a nose contour shape by an element shape identifying unit 115 according to Embodiment 1. FIG. 6 is a diagram for describing identification of a nose contour shape by an element shape identifying unit 115 according to Embodiment 1. FIG. 5 is a flowchart of a process for specifying the contour shape of the first nostril by the element shape specifying unit 115 according to the first embodiment. FIG. 10 is a diagram for describing identification of the contour shape of the first nostril by the element shape identification unit 115 in the first embodiment. FIG. 10 is a diagram for describing identification of the contour shape of the first nostril by the element shape identification unit 115 in the first embodiment. FIG. 10 is a diagram for describing identification of the contour shape of the first nostril by the element shape identification unit 115 in the first embodiment. FIG. 10 is a diagram for describing identification of the contour shape of the first nostril by the element shape identification unit 115 in the first embodiment. FIG. 6 is a diagram showing the contents of nose feature information in the first embodiment. The system configuration | structure of the pet management system 1000 in Embodiment 2 is shown. The content of the pet identification information in Embodiment 2 is shown. The content of the pet owner information in Embodiment 2 is shown. The contents of health center information in Embodiment 2 are shown. It is a sequence diagram of the exchange between the pet owner side terminal, the pet management device 100, and the health center side terminal in the second embodiment. It is a figure which shows the case where the search request side nose feature information in Embodiment 2 is compared with nose feature information 1 grade | etc.,. It is a figure which shows the "comparison method which compares a magnitude | size" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method which compares a magnitude | size" by the pet information search part 130 in Embodiment 2. FIG. FIG. 10 is a diagram illustrating a “comparison method using a linear model using a least square method” performed by a pet information search unit 130 according to Embodiment 2. FIG. 10 is a diagram illustrating a “comparison method using a linear model using a least square method” performed by a pet information search unit 130 according to Embodiment 2. FIG. 10 is a diagram illustrating a “comparison method using a linear model using a least square method” performed by a pet information search unit 130 according to Embodiment 2. FIG. 10 is a diagram illustrating a “comparison method using a linear model using a least square method” performed by a pet information search unit 130 according to Embodiment 2. It is a figure which shows the "comparison method using the polynomial model by the least square method" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using the polynomial model by the least square method" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using the polynomial model by the least square method" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using a Fourier series" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using a Fourier series" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using a Fourier series" by the pet information search part 130 in Embodiment 2. FIG. It is a figure which shows the "comparison method using a Fourier series" by the pet information search part 130 in Embodiment 2. FIG. 14 is a flowchart illustrating a process of a nose feature information generation program according to the third embodiment. 14 shows an external appearance of a computer system according to a third embodiment. 10 illustrates a hardware configuration of a computer system according to a third embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Nose area, 3 nose outline image, 10 intersection, 11 1st nose wing, 15 intersection, 20 front photograph, 30a, 30b, 30c pet identification information, 40a pet owner information, 50a registration applicant information, 100 pet management apparatus, 110 Pet information identification unit, 111 image input unit, 112 nose contour image extraction unit, 113 correction determination unit, 114 element extraction unit, 115 element shape identification unit, 120 pet information storage unit, 130 pet information search unit, 140 communication unit, 150 Pet management DB, 160 Pet owner management DB, 170 Display unit, 200 Photo reader, 300 Internet, 401 Mobile phone, 402,403 Client terminal, 410 Client LAN, 411, 412 Client terminal, 413 Modem / router, 800 Computer system, 8 0 CPU, 811 ROM, 812 RAM, 813 DSP, 814 K / B, 815 mouse, 816 communication board, 817 FDD, 818 photo reader, 819 printer, 820 magnetic disk unit, 821 OS, 822 window system, 823 program group, 824 files, 825 bus, 826 CDD, 830 system unit, 1000 pet management system.

Claims (14)

An image input unit for inputting a nose photographed image in which the entire nose including two nostrils is photographed;
The nose contour image is identified by identifying the outline of the nose from the nose photographed image input by the image input unit, and cutting out a nose contour image that is a nose image including two nostrils from the nose photographed image along the identified nose contour. A nose contour image extraction unit to extract;
It is determined whether or not the nose contour image extracted by the nose contour image extraction unit needs to be corrected. When it is determined that correction is necessary, the corrected nose contour image is corrected by performing the predetermined correction. A correction determination unit that outputs a nose contour image as a correction determined image without performing correction when it is determined that correction is not necessary,
An element extraction unit that extracts the outline of the nose and the outlines of the first and second nostrils, which are the two nostrils, from the correction determined image output by the correction determination unit;
The nasal reference position calculated by calculating the nasal reference position, which is a reference position used for specifying the shape of the nose outline extracted by the element extraction unit, by a predetermined nasal reference position calculation method and the element extraction unit extract Identify the shape of the nose from the contour of the nose
A first nostril reference calculated by calculating a first nostril reference position, which is a reference position used for specifying the contour shape of the first nostril extracted by the element extraction unit, by a predetermined first nostril reference position calculation method. Identifying the contour shape of the first nostril from the position and the contour of the first nostril extracted by the element extraction unit;
A second nostril reference calculated by calculating a second nostril reference position, which is a reference position used for specifying the contour shape of the second nostril extracted by the element extraction unit, by a predetermined second nostril reference position calculation method. An element shape specifying unit for specifying the contour shape of the second nostril from the position and the contour of the second nostril extracted by the element extracting unit;
A nose feature information generating unit configured to generate information including nose contour shape specified by the element shape specifying unit, contour shape of the first nostril, and contour shape of the second nostril as nose feature information; A nose feature information generating device. The element shape specifying part is
As a predetermined nose reference position calculation method, the center of gravity position of the figure of the contour shape determined by the nose outline extracted by the element extraction unit is calculated as the nose reference position,
As a first nostril reference position calculation method determined in advance, a center of gravity position of a figure having a contour shape determined by the outline of the first nostril extracted by the element extraction unit is calculated as a first nostril reference position;
A predetermined second nostril reference position calculation method is characterized in that a center of gravity position of a figure having a contour shape determined by the contour of the second nostril extracted by the element extraction unit is calculated and used as a second nostril reference position. Item 1. The nasal feature information generation device according to item 1. The image input unit
Enter a nose photographed image of the entire nose, including two nostrils, a nostril corresponding to each of the two nostrils, and a nose bridge,
The element extraction unit includes:
As components of the nose, the outline of the nose, the outlines of the two nostrils, the first nostril and the second nostril, the nose bridge part, the first nostril corresponding to the first nostril, and the second nostril Extract the corresponding second nose wing,
The element shape specifying part is
Calculate the nose reference position based on the nose bridge extracted by the element extraction unit as a predetermined nose reference position calculation method,
Calculating a first nostril reference position based on the first nostril extracted by the element extraction unit as a predetermined first nostril reference position calculating method;
The nose feature information generation apparatus according to claim 1, wherein the second nostril reference position is calculated based on the second nostril extracted by the element extraction unit as a predetermined second nostril reference position calculation method. The element extraction unit includes:
Extract the nose bridge as a band, and extract each of the first and second nose wings as line segments,
The element shape specifying part is
As a predetermined nose reference position calculation method, a predetermined two points are set on the nose bridge portion extracted as a belt-like shape, a line segment passing through the two points, and both ends thereof are positioned on the nose contour. The longest line segment is created as a second line segment that is a line segment that is orthogonal to the first line segment and whose both ends are located on the nose contour, and the first and second line segments are created. Calculate the intersection with the minute and set it as the nose reference position,
As a predetermined first nostril reference position calculation method, a third line segment is created by sharing the outline of the first nostril and two intersections by extending the first nose wing extracted as a line segment. The longest line segment among the line segments that are orthogonal to each other and whose both ends are located on the outline of the first nostril is created as the fourth line segment, and the intersection of the third and fourth line segments is calculated. To the first nostril reference position,
As a predetermined second nostril reference position calculation method, a fifth line segment is created by sharing the outline of the second nostril and two intersection points by extending the second nose wing extracted as a line segment. The longest line segment is created as the sixth line segment among the line segments that are orthogonal to each other and whose both ends are located on the outline of the second nostril, and the intersection of the fifth and sixth line segments is calculated. The nose feature information generating apparatus according to claim 3, wherein the second nostril reference position is used. The element shape specifying part is
Identify the shape of the nasal bridge extracted by the element extraction unit,
The nose feature information generation unit
5. The nasal feature information generation apparatus according to claim 3, wherein the nasal feature information is generated by including the shape of the nasal bridge specified by the element shape specifying unit in the nasal feature information. The element shape specifying part is
The nose reference position is calculated as any position in the inner region of the nose contour, the first nostril reference position is calculated as any position in the inner region of the first nostril contour, and the second nostril reference When the position is calculated as any position of the inner region of the outline of the second nostril,
Lengths and directions of a plurality of line segments set by setting a plurality of line segments that start from the nose reference position in a plurality of different directions and have end points on the nose outline Identify the contour shape of the nose from the
Each of a plurality of line segments that are set by setting a plurality of line segments that are in a plurality of different directions starting from the first nostril reference position and whose end points exist on the outline of the first nostril From the length and direction, specify the contour shape of the first nostril,
Each of a plurality of line segments that are set by setting a plurality of line segments that are in a plurality of different directions starting from the second nostril reference position and whose end points are present on the outline of the second nostril The nose feature information generation device according to any one of claims 1 to 5, wherein the contour shape of the second nostril is specified from the length and direction. The element shape specifying part is
36 lines set by setting 36 line segments starting from the nose reference position and extending in 36 directions that are different from each other at approximately 10 degrees and whose end points are present on the outline of the nose. Identify the contour shape of the nose from the length and direction of each minute,
The first nostril reference position is set as a starting point, and 36 line segments are set at approximately 10 degrees in different 36 directions and the end point is on the outline of the first nostril. Identify the contour shape of the first nostril from the length and direction of each of the 36 line segments,
36 line segments are set by setting 36 line segments starting from the second nostril reference position and extending in different 36 directions at intervals of approximately 10 degrees, with the end point existing on the outline of the second nostril. The nose feature information generation device according to claim 6, wherein the contour shape of the second nostril is specified from the length and direction of each of the 36 line segments. The element shape specifying part is
72 lines set by setting 72 line segments starting from the nose reference position and starting in the direction of 72 different from each other at intervals of about 5 degrees and having the end points existing on the nose outline. Identify the contour shape of the nose from the length and direction of each minute,
72 line segments are set by setting 72 line segments that start in the first nostril reference position and move in different 72 directions at intervals of about 5 degrees and whose end points are on the outline of the first nostril. Identify the contour shape of the first nostril from the length and direction of each of the 72 line segments,
72 line segments that are in the direction of 72 different from each other at approximately every 5 degrees starting from the second nostril reference position are set and set as 72 line segments whose end points exist on the outline of the second nostril. The nose feature information generation device according to claim 6, wherein the contour shape of the second nostril is specified from the length and direction of each of the 72 line segments. The nose feature information generating device further includes:
A nasal feature information registration unit for registering nasal feature information of a predetermined type of mammal;
Communication that receives transmission source information including a nose photographed image of the predetermined type of mammal and transmission source data indicating the transmission source from the transmission source, and outputs the nose imaging information included in the received transmission source information to the image input unit With
The nose feature information generation unit
The communication unit generates nose feature information based on the nose photographed image output to the image input unit, and associates the generated nose feature information with the transmission source data included in the transmission source information received by the communication unit. The nasal feature information generation apparatus according to claim 1, wherein the nasal feature information is registered in the nose feature information registration unit as the associated nose feature information. The nose feature information registration unit
For the predetermined type of mammal, a plurality of associated nose feature information is registered,
The communication unit is
Determination request information for requesting determination as to whether or not any of the associated nose feature information registered in the nose feature information registration unit is matched, and transmitting determination request information including a nose photographed image related to the determination request Receiving determination request information from the requesting device, outputting a nose photographed image related to the received determination request to the image input unit,
The nose feature information generation unit
Generating determination request nose feature information that is nose feature information based on a nose photographed image related to the determination request output by the communication unit to the image input unit;
The nose feature information generating device further includes:
The determination request nose feature information generated by the nose feature information generation unit is set in advance by comparing each of the plurality of associated nose feature information registered in the nose feature information registration unit with a predetermined comparison method. And a search unit configured to search the associated nose feature information from the nose feature information registration unit based on the search criteria and return a search result to the determination requesting device via the communication unit. Item 9. The nasal feature information generation device according to Item 9. The search unit
As a result of the search, when the associated nose feature information can be retrieved from the nose feature information registration unit, transmission source data corresponding to the retrieved associated nose feature information is returned as a search result. The nose characteristic information generation device according to claim 10. The transmission source information received by the communication unit is:
Including individual feature data indicating characteristics of the individual mammal of the predetermined type related to the nose photograph image included in the transmission source information,
The nose feature information generation unit
The associated nose feature information is further registered in the nose feature information registration unit in association with the individual feature data,
The search unit
As a result of the search, when the associated nose feature information can be retrieved from the nose feature information registration unit, the source data and the individual feature data corresponding to the associated nose feature information are returned as a search result. The nose feature information generation device according to claim 10. The individual feature data received by the communication unit is:
13. The nasal feature information generation apparatus according to claim 12, further comprising a whole photographed image obtained by photographing the whole mammal of the predetermined type. A process of inputting a nose photograph image in which the entire nose including two nostrils is photographed, which is an image of a mammal's nose;
A nose outline is identified by identifying a nose outline from the input nose photographed image, and extracting a nose outline image that is a nose outline image including two nostrils from the nose photographed image along the identified nose outline. Processing to
It is determined whether or not the extracted nose contour image needs to be corrected. If it is determined that correction is necessary, the corrected nose contour image is corrected and determined as a corrected image. A process of outputting a nose contour image as a corrected determination image without performing correction when it is determined that there is no need for correction;
A process of extracting the outline of the nose and the outlines of the two nostrils, the first nostril and the second nostril, from the output corrected and determined image,
A nose based on the nose reference position calculated by calculating a nose reference position, which is a reference position used for specifying the shape of the extracted nose outline, by a predetermined nose reference position calculation method and the extracted nose outline Identify the contour shape of
The first nostril reference position calculated by calculating the first nostril reference position, which is a reference position used for specifying the extracted contour shape of the first nostril, by a predetermined first nostril reference position calculation method is extracted. Identifying the contour shape of the first nostril from the contour of the first nostril,
A second nostril reference position calculated by calculating a second nostril reference position, which is a reference position used for specifying the contour shape of the extracted second nostril, by a predetermined second nostril reference position calculation method is extracted. A process of specifying the contour shape of the second nostril from the contour of the second nostril;
Nose feature information generation characterized by causing a computer to execute processing for generating information including the identified nose contour shape, the first nostril contour shape, and the second nostril contour shape as nose feature information program.