JP2018018137A - Image processor, image processing method, program and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor which can create a synthetic image with which a plurality of users are satisfied by using a plurality of images which are acquired from a plurality of the users without making the user edit a layout, an imaging processing method, a program and a recording medium.SOLUTION: In an image processor, an imaging processing method, a program and a recording medium, a synthetic image evaluation value calculation part calculates a synthetic image analysis evaluation value. A synthetic image user evaluation value acquisition part acquires synthetic image user evaluation values from a plurality of users, and a synthetic image user evaluation value calculation part calculates a synthetic image user evaluation comprehensive value. When a difference between the synthetic image user evaluation comprehensive value and the synthetic image analysis evaluation value is larger than a threshold, a synthetic condition change part creates a change condition which is obtained by changing at least one of an individual image evaluation condition, an image extraction condition and a synthetic image creation condition. The synthetic image creation part re-creates the synthetic image on the basis of the change condition.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an image processing apparatus, an image processing method, a program, and a recording medium that create a composite image using a plurality of images acquired from a plurality of users.

  Currently, a service is provided for acquiring a plurality of images (image data) owned by a user via the Internet or the like and creating a photo album having a layout desired by the user using the plurality of images acquired from the user. Yes.

  In recent years, not only a photo album is created using only a plurality of images owned by a single user, but also a photo album is created using a plurality of images owned by a plurality of users (Patent Document 1 and 2) is known.

  In addition, based on registration information of an image service using a network, an image group of a plurality of users belonging to the same group is extracted and used as a composite image (Patent Document 3), and messages from a plurality of participants are used. Receives and creates a draft (Patent Document 4) and receives the evaluation of a specific user for the evaluation target, compares the evaluation of the specific user with the evaluation standard of the general user, identifies the evaluation standard of the specific user, and specifies A device for creating an evaluation object highly evaluated by a user (Patent Document 5) is known.

JP 2008-257308 A JP, 2015-066941, A JP, 2015-066941, A JP2015-133000A Japanese Patent No. 3177746

  However, when multiple users collaborate to create a composite image such as a photo album, if multiple users can edit the layout simultaneously, there is a risk that if multiple users edit the layout at the same time, they will not be collected There is.

  For example, as the number of users who edit the layout increases, the layout is frequently changed, so that it is difficult to edit the layout. Further, synchronization cannot be achieved when a plurality of users edit conflicting layouts. The final layout editing goal is difficult to understand, and there is a problem that the burden on the secretary user who becomes a coordinator when creating a composite image such as a photo album among a plurality of users is large.

  An object of the present invention is to provide an image processing apparatus, an image processing method, and an image processing method capable of creating a composite image that can be accepted by the majority of users using a plurality of images acquired from a plurality of users without editing the layout by the user. It is to provide a program and a recording medium.

In order to achieve the above object, the present invention provides an image acquisition unit that acquires a plurality of images from a plurality of users,
An individual image analysis unit for analyzing each of the plurality of images;
An individual image evaluation value calculation unit that calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on an analysis result of each image;
An image extraction unit that extracts one or more images corresponding to one or more image extraction conditions from a plurality of images based on the individual image evaluation value of each image;
A composite image creation unit that creates a composite image obtained by combining one or more images based on one or more composite image creation conditions;
A composite image analysis unit for analyzing the composite image;
A composite image evaluation value calculation unit that calculates each of one or more composite image analysis evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items based on the analysis result of the composite image;
A composite image user evaluation value acquisition unit that acquires each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items from a plurality of users;
For each composite image evaluation item, a composite image user evaluation value calculation unit that calculates a composite image user evaluation total value based on composite image user evaluation values from a plurality of users;
For each composite image evaluation item, when the difference between the composite image user evaluation total value and the composite image analysis evaluation value is larger than the threshold, the individual image is set so that the composite image analysis evaluation value approaches the composite image user evaluation total value. A synthesis condition change unit that creates a change condition in which at least one of the evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference;
The composite image creation unit provides an image processing apparatus that recreates a composite image based on the change condition created by the synthesis condition change unit.

  Here, when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value, the composite condition changing unit is configured so that the composite image analysis evaluation value decreases according to the composite image user evaluation total value. Or, when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the individual image evaluation condition, so that the composite image analysis evaluation value increases according to the composite image user evaluation total value, It is preferable to create a change condition in which at least one of the image extraction condition and the composite image creation condition is changed.

Further, the composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of images used in the composite image,
When the composite image user evaluation total value is greater than the threshold with respect to the composite image analysis evaluation value, the composite condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value, or Template used to create a composite image so that the composite image analysis evaluation value increases according to the composite image user evaluation total value when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value Change the number of arranged images in
It is preferable that the composite image creation unit recreates the composite image using the template in which the number of arranged images has been changed by the synthesis condition change unit.

The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of person images used in the composite image,
When the composite image user evaluation total value is greater than the threshold with respect to the composite image analysis evaluation value, the composite condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value, or When the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, one or more of the plurality of images are set so that the composite image analysis evaluation value increases according to the composite image user evaluation total value. Change the weight of the person image when extracting the image of
The composite image creating unit preferably recreates the composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the composition condition changing unit.

Further, the composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the atmosphere of the composite image,
When the composite image user evaluation total value is greater than the threshold with respect to the composite image analysis evaluation value, the composite condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value, or Template used to create a composite image so that the composite image analysis evaluation value increases according to the composite image user evaluation total value when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value Change the background image of
The composite image creating unit preferably recreates the composite image using the template whose background image has been changed by the composite condition changing unit.

  The composite image user evaluation value calculation unit preferably calculates the sum, average, or product of the composite image user evaluation values from a plurality of users as the composite image user evaluation total value.

In addition, a weight calculation unit for calculating the weight of each of a plurality of users is provided,
The composite image user evaluation value calculation unit preferably calculates the sum, average or product of the composite image user evaluation values from a plurality of users weighted according to the weight of each user as the composite image user evaluation total value. .

In addition, a personal information acquisition unit that acquires personal information of each user is provided,
The weight calculation unit preferably calculates the weight of each of the plurality of users based on the personal information of each user.

  Moreover, it is preferable that a weight calculation part calculates the weight of each of a some user so that the weight of the user from whom more images were acquired by the image acquisition part becomes heavier.

  In addition, the weight calculation unit may calculate the weight of each of the plurality of users so that the weight of the user whose composite image user evaluation value is acquired at an earlier timing by the composite image user evaluation value acquisition unit becomes heavier. preferable.

Further, a synthesis condition storage unit that stores a plurality of change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference;
A synthesis condition selection unit that selects one change condition corresponding to a composite image evaluation item having a difference larger than a threshold value from a plurality of change conditions stored in the synthesis condition storage unit;
The composite image creation unit preferably recreates the composite image based on the change condition selected by the synthesis condition selection unit.

In addition, a synthesis condition storage unit that stores a plurality of change conditions obtained by changing at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition according to the difference,
The composite image creating unit preferably recreates the composite image based on one change condition selected by each of a plurality of users from the plurality of change conditions stored in the composite condition storage unit.

The synthesis condition changing unit creates two or more change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference,
The composite image creation unit re-creates two or more composite images based on the two or more change conditions created by the composite condition change unit,
A composite image selection number acquisition unit that acquires the number of each of two or more composite images selected by a plurality of users for two or more composite images;
It is preferable to include a composite image information acquisition unit that acquires information of a selected composite image having the largest number selected by a plurality of users from two or more composite images.

The composite image user evaluation value calculation unit further calculates a standard deviation of the composite image user evaluation values from a plurality of users,
The synthesis condition changing unit creates two or more change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the synthesized image creation condition is changed according to the difference and the standard deviation,
The composite image creation unit re-creates two or more composite images based on the two or more change conditions created by the composite condition change unit,
A composite image selection number acquisition unit that acquires the number of each of two or more composite images selected by a plurality of users for two or more composite images;
It is preferable to include a composite image information acquisition unit that acquires information of a selected composite image having the largest number selected by a plurality of users from two or more composite images.

The present invention also provides an image acquisition unit that acquires a plurality of images from a plurality of users;
An individual image analysis unit for analyzing each of the plurality of images;
An individual image evaluation value calculation unit that calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on an analysis result of each image;
An image extraction unit that extracts one or more images corresponding to one or more image extraction conditions from a plurality of images based on the individual image evaluation value of each image;
A composite image creation unit that creates a composite image obtained by combining one or more images based on one or more composite image creation conditions;
A composite image user evaluation value acquisition unit that acquires each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items from a plurality of users;
For each composite image evaluation item, a composite image user evaluation value calculation unit that calculates a standard deviation of composite image user evaluation values from a plurality of users,
For each composite image evaluation item, when the standard deviation is larger than the threshold, two or more change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the standard deviation A synthesis condition change part for creating
Acquisition of the number of selected composite images for acquiring the number of each of two or more composite images selected by a plurality of users for two or more composite images recreated based on two or more change conditions by the composite image creation unit And
Provided is an image processing apparatus including a composite image information acquisition unit that acquires information of a selected composite image having the largest number selected by a plurality of users from two or more composite images.

Here, the composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of images used in the composite image,
When the standard deviation is larger than the threshold, the synthesis condition changing unit changes the number of arranged images in the template used for creating the composite image according to the standard deviation.
It is preferable that the composite image creation unit recreates the composite image using the template in which the number of arranged images has been changed by the synthesis condition change unit.

The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of person images used in the composite image,
The composition condition changing unit changes the weight of the person image when extracting one or more images from the plurality of images according to the standard deviation when the standard deviation is larger than the threshold,
The composite image creating unit preferably recreates the composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the composition condition changing unit.

Further, the composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the atmosphere of the composite image,
When the standard deviation is larger than the threshold, the synthesis condition changing unit changes the template background image used for creating the composite image according to the standard deviation,
The composite image creating unit preferably recreates the composite image using the template whose background image has been changed by the composite condition changing unit.

Further, a synthesis condition storage unit that stores a plurality of change conditions obtained by changing at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition according to the standard deviation;
A synthesis condition selection unit that selects two or more change conditions corresponding to a composite image evaluation item having a standard deviation larger than a threshold value from a plurality of change conditions stored in the synthesis condition storage unit;
The composite image creation unit preferably recreates the composite image based on the change condition selected by the synthesis condition selection unit.

In addition, a synthesis condition storage unit that stores a plurality of change conditions obtained by changing at least one of the individual image evaluation conditions, the image extraction conditions, and the composite image creation conditions according to the standard deviation,
The composite image creation unit preferably recreates the composite image based on two or more change conditions selected by each of a plurality of users from the plurality of change conditions stored in the composite condition storage unit.

Further, based on the analysis result of each image, an object detection unit that detects an object other than a person shown in each image and creates a list of objects shown in a plurality of images;
An object number acquisition unit for acquiring the number of each object selected by a plurality of users from the list of objects;
The composition condition changing unit further includes each object when extracting one or more images from the plurality of images according to the number of each object selected by the plurality of users from the list of objects. Change the weight of the image
It is preferable that the composite image creation unit recreates the composite image using one or more images extracted from the plurality of images according to the weight of the object changed by the synthesis condition change unit.

In addition, a composite image evaluation item storage unit that stores one or more composite image evaluation items,
The composite image user evaluation value acquisition unit preferably acquires a composite image user evaluation value obtained by evaluating the composite image for each of the one or more composite image evaluation items stored in the composite image evaluation item storage unit.

In addition, a composite image evaluation item storage unit that stores one or more composite image evaluation items,
The composite image user evaluation value acquisition unit evaluates the composite image for a composite image evaluation item selected by each of a plurality of users from one or more composite image evaluation items stored in the composite image evaluation item storage unit. It is preferable to acquire a user evaluation value.

  Further, the composite image evaluation value calculation unit determines that the composite image is evaluated for a composite image evaluation item that exceeds the threshold when the composite image analysis evaluation value exceeds the threshold. It is preferable to recommend to other users.

In addition, a composite image evaluation item creation unit is provided that creates a composite image evaluation item for an object whose appearance count in the composite image is greater than a threshold from one or more objects based on the individual image evaluation value of each image. ,
The composite image user evaluation value acquisition unit preferably acquires a composite image user evaluation value obtained by evaluating the composite image for the composite image evaluation item created by the composite image evaluation item creation unit.

In addition, a composite image information storage unit that stores information of the selected composite image is provided,
The composition condition changing unit is used to create the next composite image so as to match the selection tendency of the composite image by a plurality of users based on the information of the selected composite image stored in the composite image information storage unit. It is preferable to change at least one of the image evaluation condition, the image extraction condition, and the composite image creation condition.

Further, for each composite image evaluation item, a composite image user evaluation value storage unit that stores an average of composite image user evaluation values from a plurality of users,
The composite image creation unit, for each composite image evaluation item, the individual image so that the composite image analysis evaluation value approaches the average of the composite image user evaluation values from a plurality of users stored in the composite image user evaluation value storage unit. It is preferable to create the next composite image under a condition in which at least one of the evaluation condition, the image extraction condition, and the composite image creation condition is determined.

In the present invention, the image acquisition unit acquires a plurality of images from a plurality of users;
An individual image analysis unit analyzing each of the plurality of images;
An individual image evaluation value calculating unit calculating an individual image evaluation value obtained by evaluating each image according to one or more individual image evaluation conditions based on an analysis result of each image;
An image extraction unit extracting one or more images corresponding to one or more image extraction conditions from a plurality of images based on the individual image evaluation values of each image;
A step of creating a composite image obtained by compositing one or more images based on one or more composite image creation conditions;
A step in which the composite image analysis unit analyzes the composite image;
A step of calculating one or more composite image analysis evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items based on the analysis result of the composite image;
A composite image user evaluation value acquisition unit acquiring, from a plurality of users, each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items;
A step of calculating a composite image user evaluation total value based on a composite image user evaluation value from a plurality of users for each composite image evaluation item;
When the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value is larger than the threshold value for each composite image evaluation item, the composite image analysis evaluation value is changed to the composite image user evaluation comprehensive value for the composite condition changing unit. Creating a change condition in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference so as to approach;
And a step of re-creating a composite image based on the change condition created by the step of creating the change condition.

In the present invention, the image acquisition unit acquires a plurality of images from a plurality of users;
An individual image analysis unit analyzing each of the plurality of images;
An individual image evaluation value calculating unit calculating an individual image evaluation value obtained by evaluating each image according to one or more individual image evaluation conditions based on an analysis result of each image;
An image extraction unit extracting one or more images corresponding to one or more image extraction conditions from a plurality of images based on the individual image evaluation values of each image;
A step of creating a composite image obtained by compositing one or more images based on one or more composite image creation conditions;
A composite image user evaluation value acquisition unit acquiring, from a plurality of users, each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items;
A step of calculating a standard deviation of a composite image user evaluation value from a plurality of users for each composite image evaluation item;
For each composite image evaluation item, when the standard deviation is larger than the threshold, the composition condition changing unit changes at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition according to the standard deviation. Creating two or more change conditions,
Each of the two or more composite images selected by a plurality of users for the two or more composite images regenerated based on the two or more change conditions by the composite image selection number acquisition unit in the step of generating the composite image Obtaining a number of
A composite image information acquisition unit including a step of acquiring information of a selected composite image having the largest number selected by a plurality of users from two or more composite images.

  The present invention also provides a program for causing a computer to execute each step of the image processing method described above.

  The present invention also provides a computer-readable recording medium on which a program for causing a computer to execute each step of the image processing method described above is recorded.

  In the present invention, a plurality of users do not directly edit a layout of a composite image created using a plurality of images acquired from a plurality of users, but calculate a composite image analysis evaluation value, and each user The composite image user evaluation value is acquired from the image, the composite image user evaluation total value is calculated, and the composite image is recreated based on the difference between the composite image user evaluation total value and the composite image analysis evaluation value.

  As described above, according to the present invention, since a plurality of users do not directly edit the composite image layout, even if the number of users increases, the composite image is recreated, that is, the composite image layout is edited. can do. Further, since the image processing apparatus, not the user, recreates the composite image, it is possible to create a composite image that can be averaged by the majority of users without imposing a heavy burden on the secretary user.

  In the present invention, a composite image user evaluation value is acquired from each user, a standard deviation of composite image user evaluation total values from a plurality of users is calculated, and two or more composite images are created based on the standard deviation. Try again. Then, information on a selected composite image that is a composite image having the largest number selected by a plurality of users from two or more composite images is acquired, and a selected composite image is adopted from the two or more composite images.

  Thus, by recreating the composite image based on the standard deviation, according to the present invention, even if the distribution of the composite image user evaluation values from a plurality of users varies, a large number of users Can be synthesized on average.

It is a block diagram of one embodiment showing composition of an image processing device concerning the present invention. It is a block diagram of one Embodiment showing the structure of the server shown in FIG. It is a block diagram of one Embodiment showing the structure of the user's terminal device shown in FIG. 2 is a flowchart of an embodiment illustrating an operation of the image processing apparatus illustrated in FIG. 1. It is a conceptual diagram of one Embodiment showing the screen which evaluates a synthetic | combination image about the number of images currently used in a synthetic | combination image. It is a conceptual diagram of one Embodiment showing the screen which evaluates a synthesized image about the atmosphere of a synthesized image. 6 is a flowchart of another embodiment showing the operation of the image processing apparatus shown in FIG. 1.

  Hereinafter, an image processing apparatus, an image processing method, a program, and a recording medium of the present invention will be described in detail based on preferred embodiments shown in the accompanying drawings.

  FIG. 1 is a block diagram of an embodiment showing a configuration of an image processing apparatus according to the present invention. An image processing apparatus 10 illustrated in FIG. 1 includes a server 12 and a plurality of user terminal devices (clients) 14 connected to the server 12 via a network 16. The image processing apparatus 10 creates a composite image such as a photo album using a plurality of images of a plurality of users acquired from the terminal devices 14 of the plurality of users via the network 16.

  The server 12 performs various data processing for creating a composite image based on an instruction from the terminal device 14, and is configured by, for example, a desktop PC (Personal Computer) or a workstation. .

  The terminal device 14 gives various instructions to the server 12 to perform various data processing, and is configured by, for example, a smartphone, a tablet PC, or a notebook PC.

  The network 16 is, for example, a telephone line or an internet line, and the server 12 and the terminal device 14 are connected to each other by wire or wireless, and bidirectional communication is possible.

  FIG. 2 is a block diagram of an embodiment showing the configuration of the server shown in FIG. The server 12 shown in the figure includes an image acquisition unit 18, an individual image analysis unit 20, an individual image evaluation value calculation unit 22, an image extraction unit 24, a composite image creation unit 26, a composite image analysis unit 28, A composite image evaluation value calculation unit 30, a composite image user evaluation value acquisition unit 32, a composite image user evaluation value calculation unit 34, a combination condition change unit 36, a personal information acquisition unit 38, a weight calculation unit 40, a combination A condition storage unit 42, a synthesis condition selection unit 44, a composite image selection number acquisition unit 46, a composite image information acquisition unit 48, a composite image information storage unit 50, an object detection unit 52, and an object number acquisition unit 54 A composite image evaluation item storage unit 56, a composite image evaluation item creation unit 58, and a composite image user evaluation value storage unit 59.

  In the server 12, the image acquisition unit 18 acquires a plurality of images from the terminal devices 14 of a plurality of users via the network 16.

  Subsequently, the individual image analysis unit 20 analyzes each of the plurality of images acquired by the image acquisition unit 18.

  The individual image analysis unit 20 is, for example, Exif (Exchangeable Image File Format: Exchangeable Image File Format) in addition to image quality analysis such as image brightness, hue, degree of blur and blur, and image size analysis. Format) and other information included in the image information, such as shooting date and time, shooting location, etc., analysis of people and non-person objects in the image, facial recognition of the person in the image, protagonist and important person identification Analyze image scenes.

  Subsequently, the individual image evaluation value calculation unit 22 calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on the analysis result of each image by the individual image analysis unit 20.

  The individual image evaluation conditions are conditions for evaluating each image, and are the image quality of the image, that is, the brightness of the image, the hue, the degree of blur and blur, and the like. For example, the individual image evaluation value calculation unit 22 calculates an individual image evaluation value obtained by evaluating the synthesized image with respect to the degree of blur and blur of the image based on the analysis result of the degree of blur and blur of the image. For example, the smaller the degree of blurring and blurring of the image, the higher the individual image evaluation value for the degree of blurring and blurring of the image is calculated.

  Subsequently, the image extraction unit 24, based on the individual image evaluation value of each image calculated by the individual image evaluation value calculation unit 22, one or more corresponding to one or more image extraction conditions from the plurality of images. Extract images. Note that extracting all of the plurality of images is also included in the image extraction.

  The image extraction condition is a condition for extracting one or more images from a plurality of images, and is the image quality, that is, the brightness of the image, the hue, the degree of blur, and the blur. The image extraction unit 24, for example, based on the individual image evaluation value of the degree of blurring and blurring of the image, one or more images having an individual image evaluation value of the degree of blurring and blurring that is higher than the threshold value. To extract.

Subsequently, the composite image creation unit 26 creates a composite image by combining the one or more images extracted by the image extraction unit 24 based on one or more composite image creation conditions.
The composite image creation unit 26 recreates a composite image based on at least one of a change condition created by a composition condition change unit 36 described later and a change condition selected by the composition condition selection unit 44. .

  The composite image creation condition is a condition for creating a composite image. For example, the number of images used in the composite image, the number of person images used in the composite image, and an object other than a person shown in the image used in the composite image , And a template used in the composite image.

  In the template, the number of images used in the composite image, the arrangement position of the images, the background image, and the like are determined in advance. For example, when a composite image is created using five images, the composite image can be created by arranging the five images at predetermined positions on the template. Further, the atmosphere of the composite image can be changed by changing the background image of the template.

  The composite image creation unit 26 creates, for example, a composite image obtained by arranging one or more images on a template and combining them, a postcard with an image obtained by combining one image, and the like. In addition, the composite image creation unit 26 can create a composite image in which not only an image but also a stamp and a user comment are combined.

  Subsequently, the composite image analysis unit 28 analyzes the composite image created by the composite image creation unit 26.

The composite image analysis unit 28, for example, the number of images used in the composite image, the number of person images used in the composite image, an object other than the person shown in the image used in the composite image, and the composite Analyze the atmosphere of the image.
As the atmosphere of the composite image, various atmospheres can be assumed, for example, whether the atmosphere is bright or dark, or the atmosphere is pop or chic.

  Subsequently, the composite image evaluation value calculation unit 30 evaluates the composite image for each of the one or more composite image evaluation items based on the analysis result of the composite image by the composite image analysis unit 28. Is calculated.

The composite image evaluation items include, for example, the number of images used in the composite image, the number of person images used in the composite image, the atmosphere of the composite image, and a person other than the person shown in the image used in the composite image The type of object.
For example, if the composite image evaluation item is the number of images used in the composite image and the number of images used in the composite image is 5 as a result of the image analysis of the composite image, 5 images are supported. A composite image analysis evaluation value is calculated.

  Subsequently, the composite image user evaluation value acquisition unit 32 evaluates the composite image for each of the one or more composite image evaluation items from the terminal devices 14 of a plurality of users via the network 16. Get each of the values.

For example, when the composite image evaluation item is the number of images used in the composite image, the user is “too little”, “less”, “just right”, “more”, or “ Evaluate whether it is too much. Note that the evaluation may be at least two stages.
When the composite image analysis evaluation value of the number of images used in the composite image is five and it is evaluated that the number of images is large, a composite image user evaluation value corresponding to the large number of images is acquired.

Subsequently, the composite image user evaluation value calculation unit 34, for each composite image evaluation item, based on the composite image user evaluation values from a plurality of users acquired by the composite image user evaluation value acquisition unit 32. Calculate the total value.
Moreover, the composite image user evaluation value calculation unit 34 calculates the standard deviation of the composite image user evaluation values from a plurality of users for each composite image evaluation item.

  The composite image user evaluation total value is a total value of composite image user evaluation values from a plurality of users, and is, for example, the sum, average, or product of composite image user evaluation values from a plurality of users. The composite image user evaluation total value may be the sum, average, or product of composite image user evaluation values from a plurality of users weighted according to the weight of each user calculated by the weight calculation unit 40 described later.

  Subsequently, the composition condition changing unit 36, for each composite image evaluation item, the composite image user evaluation total value calculated by the composite image user evaluation value calculation unit 34 and the composite image calculated by the composite image evaluation value calculation unit 30. When the difference from the image analysis evaluation value is larger than a threshold value (including the case where the threshold value is 0), the individual image evaluation condition, the image extraction condition, and the A change condition is created by changing at least one of the composite image creation conditions in accordance with the difference between the composite image user evaluation total value and the composite image analysis evaluation value.

  Also, the composition condition changing unit 36, for each composite image evaluation item, when the standard deviation calculated by the composite image user evaluation value calculation unit 34 is larger than the threshold, the individual image evaluation condition, the image extraction condition, and the composite image Two or more change conditions are created by changing at least one of the creation conditions according to the standard deviation.

  The change condition created by the composition condition changing unit 36 is used when the composite image creating unit 26 recreates the composite image.

  Subsequently, the personal information acquisition unit 38 acquires personal information of each user from the terminal devices 14 of the plurality of users via the network 16.

The personal information is, for example, personal information of each user including information such as the user's name, age, face image, hobby, etc. When the weight calculation unit 40 described later calculates the weight of each user. used.
For example, the personal information acquisition unit 38 may acquire registration information in an SNS (Social Networking Service) of the user of the terminal device 14 from which the image acquisition unit 18 has acquired each image.

  Subsequently, the weight calculation unit 40 calculates the weight of each of the plurality of users.

  For example, the weight calculation unit 40 can calculate the weight of each of a plurality of users based on the personal information of each user acquired by the personal information acquisition unit 38. For example, based on personal information, if user A's hobby is photography and user B's hobby is not photography, that is, if user A is considered to have more detailed knowledge about photography than user B, The weight of A is made heavier than that of user B.

Further, the weight calculation unit 40 can calculate the weight of each of the plurality of users so that the weight of the user whose more images are acquired by the image acquisition unit 18 becomes heavier.
Further, the weight calculation unit 40 calculates the weight of each of the plurality of users so that the weight of the user whose composite image user evaluation value is acquired at an earlier timing by the composite image user evaluation value acquisition unit 32 becomes heavier. be able to.

  As described above, the weight of each user is the sum, average or product of composite image evaluation values from a plurality of users weighted according to the weight of each user as the composite image user evaluation total value as described above. Used for.

Subsequently, the composite condition storage unit 42 changes at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition in accordance with the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value. A plurality of changed conditions are stored.
Further, the synthesis condition storage unit 42 stores a plurality of change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the standard deviation.

  The change condition stored in the composition condition storage unit 42 is used as a condition when the composite image creation unit 26 recreates the composite image.

Subsequently, the synthesis condition selection unit 44 selects a composite image evaluation in which the difference between the composite image user evaluation total value and the composite image analysis evaluation value is larger than the threshold value from among a plurality of change conditions stored in the synthesis condition storage unit 42. One change condition corresponding to the item is selected.
Further, the synthesis condition selection unit 44 selects two or more change conditions corresponding to a composite image evaluation item having a standard deviation larger than the threshold value from among a plurality of change conditions stored in the synthesis condition storage unit 42.

  The change condition selected by the composition condition selection unit 44 is used when the composite image creation unit 26 recreates the composite image.

  Subsequently, the composite image selection number acquisition unit 46 has information on the composite image selected by each user for the two or more composite images recreated by the composite image creation unit 26 based on the two or more change conditions. The number of each of the two or more composite images selected by the plurality of users is acquired based on the above.

  Subsequently, based on the number of each of two or more composite images selected by a plurality of users acquired by the composite image selection number acquisition unit 46, the composite image information acquisition unit 48 outputs 2 Information on a selected composite image that is a composite image having the largest number selected by a plurality of users from two or more composite images recreated based on the above change conditions is acquired.

  Subsequently, the composite image information storage unit 50 stores information on the selected composite image acquired by the composite image information acquisition unit 48.

  The information of the selected composite image stored in the composite image information storage unit 50 is used for creating the next composite image so that the composite condition changing unit 36 matches the tendency of selection of composite images by a plurality of users. This is used to change at least one of the image evaluation condition, the image extraction condition, and the composite image creation condition.

  Subsequently, based on the analysis result of each image, the object detection unit 52 detects an object other than a person shown in each image and creates a list of objects shown in a plurality of images.

  Subsequently, the object number acquisition unit 54 selects each object selected by a plurality of users based on the information on the object selected by each user from the list of objects created by the object detection unit 52. Get the number.

  The number of objects acquired by the object number acquisition unit 54 is determined when the composition condition changing unit 36 changes the weight of an image in which each object is captured when one or more images are extracted from a plurality of images. Used for.

  Subsequently, the composite image evaluation item storage unit 56 stores one or more composite image evaluation items.

  The one or more composite image evaluation items stored in the composite image evaluation item storage unit 56 are used as a composite image evaluation item when the composite image user evaluation value acquisition unit 32 acquires a composite image user evaluation value.

  Subsequently, the composite image evaluation item creation unit 58 selects a composite image evaluation item for an object whose appearance count in the composite image is larger than the threshold from one or more objects based on the individual image evaluation value of each image. create.

  The composite image evaluation item created by the composite image evaluation item creation unit 58 is used as a composite image evaluation item when the composite image user evaluation value acquisition unit 32 acquires a composite image user evaluation value.

  Finally, the composite image user evaluation value storage unit 59 calculates composite image user evaluations from a plurality of users calculated by the composite image user evaluation value calculation unit 34 as a composite image user evaluation total value for each composite image evaluation item. Remember the average of the values.

  The average of the composite image user evaluation values from a plurality of users stored in the composite image user evaluation value storage unit 59 is obtained when the composite image creation unit 26 creates the next composite image, the individual image evaluation condition, the image extraction condition And at least one value of the composite image creation conditions.

  Image acquisition unit 18, individual image analysis unit 20, individual image evaluation value calculation unit 22, image extraction unit 24, composite image creation unit 26, composite image analysis unit 28, composite image evaluation value calculation unit 30, composite image user evaluation value acquisition Unit 32, composite image user evaluation value calculation unit 34, composition condition change unit 36, personal information acquisition unit 38, weight calculation unit 40, composition condition storage unit 42, composition condition selection unit 44, composition image selection number acquisition unit 46, composition The image information acquisition unit 48, the composite image information storage unit 50, the object detection unit 52, the object number acquisition unit 54, the composite image evaluation item storage unit 56, the composite image evaluation item creation unit 58, and the composite image user evaluation value storage unit 59 For example, it is realized by a control device such as a CPU (Central Processing Unit) executing a program loaded in a memory. The data saved by each part is saved in storage devices such as HDD (Hard Disk Drive), SSD (Solid State Drive), SD (Secure Digital) cards, etc. Is done.

  Next, FIG. 3 is a block diagram of an embodiment showing the configuration of the user terminal device shown in FIG. The user terminal device 14 shown in the figure includes an image display unit 60 and an instruction input unit 62.

  In the terminal device 14, the image display unit 60 displays various images, a setting screen, a selection screen, a confirmation screen, an input screen, a creation screen, and the like, and is configured by a display device such as a liquid crystal display.

  Subsequently, the instruction input unit 62 acquires various setting instructions, selection instructions, confirmation instructions, input instructions, creation instructions, and the like input by the user, and includes, for example, an input device such as a keyboard and a mouse. The

  In the case of this embodiment, the image display unit 60 and the instruction input unit 62 are configured by a device in which a display device and an input device are integrated, such as a touch panel.

  Although not shown, the server 12 and the terminal device 14 include, for example, a transmission / reception unit that is a communication device that transmits and receives various data between the server 12 and the terminal device 14, a CPU that controls the operation of each part, and the like. The control part etc. which are are provided.

  Next, the operation of the image processing apparatus 10 when a composite image is recreated based on the difference between the composite image user evaluation total value and the composite image analysis evaluation value will be described with reference to the flowchart shown in FIG.

  First, the image acquisition unit 18 acquires a plurality of images from the terminal devices 14 of a plurality of users via the network 16 (step S1).

  Subsequently, each of the plurality of images acquired by the image acquisition unit 18 is analyzed by the individual image analysis unit 20 (step S2).

  Subsequently, the individual image evaluation value calculation unit 22 calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on the analysis result of each image (step S3).

  Subsequently, one or more images corresponding to one or more image extraction conditions are extracted from the plurality of images based on the individual image evaluation value of each image by the image extraction unit 24 (step S4).

  Subsequently, the composite image creation unit 26 creates a composite image obtained by combining one or more images extracted by the image extraction unit 24 based on one or more composite image creation conditions (step S5).

  Subsequently, the composite image created by the composite image creation unit 26 is sent to the terminal device 14 of each user via the network 16 and displayed on the image display unit 60.

  Each user refers to the composite image displayed on the image display unit 60 of the terminal device 14 of each user, and evaluates the composite image for each composite image evaluation item via the instruction input unit 62.

  On the other hand, in the server 12, the composite image created by the composite image creation unit 26 is analyzed by the composite image analysis unit 28 (step S6).

  Subsequently, each of the one or more composite image analysis evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items is calculated by the composite image evaluation value calculation unit 30 based on the analysis result of the composite image. (Step S7).

  When the evaluation of the composite image by the user is finished, the composite image user evaluation value acquisition unit 32 evaluated the composite image for each of one or more composite image evaluation items from the terminal devices 14 of a plurality of users via the network 16. Each of the one or more composite image user evaluation values is acquired (step S8).

  Note that if the composite image evaluation for the specific composite image evaluation item is not obtained from the user, for example, the evaluation of the composite image for the composite image evaluation item may be ignored, or “just right” May be regarded as an evaluation.

  Subsequently, the composite image user evaluation value calculation unit 34 calculates a composite image user evaluation total value for each composite image evaluation item based on the composite image user evaluation values from a plurality of users (step S9).

  Subsequently, the composition condition changing unit 36 checks whether or not the difference between the composite image user evaluation total value and the composite image analysis evaluation value is larger than the threshold value for each composite image evaluation item (step S10).

  As a result, when the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value is larger than the threshold (Yes in Step S10), the composite condition changing unit 36 causes the individual image evaluation condition, the image extraction condition, and the composite image to be A change condition is created by changing at least one of the creation conditions in accordance with the difference between the composite image user evaluation total value and the composite image analysis evaluation value (step S11).

  For example, when the combined image user evaluation total value is larger than the threshold with respect to the combined image analysis evaluation value, the combining condition changing unit 36 reduces the combined image analysis evaluation value according to the combined image user evaluation total value. Or, when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the individual image evaluation condition, so that the composite image analysis evaluation value increases according to the composite image user evaluation total value, A change condition in which at least one of the image extraction condition and the composite image creation condition is changed is created.

  On the other hand, when the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value is equal to or smaller than the threshold (No in step S10), the composite image user evaluation comprehensive value and the composite image analysis evaluation are performed for all composite image evaluation items. It is confirmed whether or not the confirmation of whether or not the difference from the value is larger than the threshold value has been completed (step S12).

  As a result, when the confirmation has not been completed for all the composite image evaluation items (No in step S11), it is confirmed whether or not the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value is larger than the threshold value. Will continue.

On the other hand, when the confirmation is completed for all the composite image evaluation items (Yes in step S11), the composite image is recreated by the composite image creation unit 26 based on the change condition created by the composite condition change unit 36 ( Step S13).
Note that when the change condition is not created by the composition condition changing unit 36, the composite image is not recreated.

  For example, when the composite image evaluation item is the number of images used in the composite image, each user uses the slider bar 64 as shown in FIG. 5 to determine the number of images used in the composite image. The composite image can be evaluated.

  In the case of the example in FIG. 5, the evaluation is performed in five stages (evaluation value) of “too little”, “less”, “just right”, “more” and “too much” from the left end to the right end of the slider bar 64. ) Is arranged. Each user evaluates the number of images used in the composite image by moving the slider 66 left and right along the slider bar 64 and selecting one of the five-level evaluations.

  In this case, when the evaluation of the composite image by the user is completed, the composite image user evaluation value acquisition unit 32 determines the composite image for the number of images used in the composite image from the terminal devices 14 of the plurality of users via the network 16. A composite image user evaluation value obtained by evaluating is obtained.

  Subsequently, the composite image user evaluation value calculation unit 34 calculates the composite image user evaluation total value for the number of images used in the composite image based on the composite image user evaluation values from a plurality of users.

  Subsequently, the composition condition changing unit 36 confirms whether or not the difference between the composite image user evaluation total value and the composite image analysis evaluation value is greater than a threshold for the number of images used in the composite image.

  As a result, when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value is decreased according to the composite image user evaluation total value by the composite condition changing unit 36. Alternatively, when the composite image user evaluation total value is smaller than the threshold value with respect to the composite image analysis evaluation value, the composite image is created so that the composite image analysis evaluation value increases according to the composite image user evaluation total value. The number of arranged images in the template used for the change is changed.

  Then, the composite image creating unit 26 recreates the composite image using the template whose number of arranged images has been changed by the composite condition changing unit 36.

  Even when the composite image evaluation item is the number of person images used in the composite image, each user uses the slider bar to display the person used in the composite image, as in the example of FIG. A composite image can be evaluated for the number of images.

  In this case, when the evaluation of the composite image by the user ends, the composite image user evaluation value acquisition unit 32 combines the number of person images used in the composite image from the terminal devices 14 of a plurality of users via the network 16. A composite image user evaluation value obtained by evaluating the image is acquired.

  Subsequently, a composite image user evaluation value calculation unit 34 calculates a composite image user evaluation total value based on composite image user evaluation values from a plurality of users for the number of person images used in the composite image.

  Subsequently, the composition condition changing unit 36 confirms whether or not the difference between the composite image user evaluation total value and the composite image analysis evaluation value is greater than a threshold for the number of person images used in the composite image.

  As a result, when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value is decreased according to the composite image user evaluation total value by the composite condition changing unit 36. Alternatively, when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the plurality of images are set such that the composite image analysis evaluation value increases according to the composite image user evaluation total value. The weight of the person image when extracting one or more images from the inside is changed.

  Then, the composite image creation unit 26 re-creates a composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the composition condition change unit 36.

  Similarly, when the composite image evaluation item is the atmosphere of the composite image, each user can evaluate the composite image with respect to the atmosphere of the composite image using a slider bar as shown in FIG.

  In the case of the example of FIG. 6, five levels of evaluation are arranged from the left end of the slider bar 64 toward the right end, namely “too dark”, “dark”, “just right”, “bright”, and “too bright”. The Similarly, each user evaluates the atmosphere of the composite image by moving the slider 66 to the left and right along the slider bar 64 and selecting one of the five levels of evaluation.

  In this case, when the evaluation of the composite image by the user ends, the composite image obtained by evaluating the composite image with respect to the atmosphere of the composite image from the plurality of user terminal devices 14 via the network 16 by the composite image user evaluation value acquisition unit 32. A user evaluation value is acquired.

  Subsequently, the composite image user evaluation value calculation unit 34 calculates a composite image user evaluation total value for the atmosphere of the composite image based on composite image user evaluation values from a plurality of users.

  Subsequently, the composition condition changing unit 36 confirms whether or not the difference between the composite image user evaluation comprehensive value and the composite image analysis evaluation value is greater than the threshold value for the atmosphere of the composite image.

  As a result, when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value is decreased according to the composite image user evaluation total value by the composite condition changing unit 36. Alternatively, when the composite image user evaluation total value is smaller than the threshold value with respect to the composite image analysis evaluation value, the composite image is created so that the composite image analysis evaluation value increases according to the composite image user evaluation total value. The background image of the template used for is changed.

  Then, the composite image creation unit 26 recreates the composite image using the template whose background image has been changed by the synthesis condition change unit 36.

  When the composite image evaluation item is the type of an object other than the person shown in the image used in the composite image, the object detection unit 52 uses the object annotation to make the analysis based on the analysis result of each image. Then, objects other than a person appearing in each image are detected, a list of objects appearing in a plurality of images is created, and displayed on the image display unit 60 of each user terminal device 14.

  Each user refers to the list displayed on the image display unit 60 of the terminal device 14 of each user, and selects an object that appears in the image that the user wants to actively use in the composite image from the list of objects. .

  When an object is selected by each user, the object number acquisition unit 54 acquires the number of each object selected by a plurality of users from the list of objects.

  Subsequently, according to the number of objects selected by the plurality of users from the list of objects by the synthesis condition changing unit 36, for example, the more objects selected by the plurality of users, the more The weight of the image in which each object is captured when one or more images are extracted from the plurality of images is changed so that the weight of the image in which the object is captured becomes heavy.

  Then, the composite image creation unit 26 re-creates a composite image using one or more images extracted from the plurality of images according to the object weight changed by the synthesis condition change unit 36.

  Moreover, you may acquire the evaluation of several users regarding the appearance frequency about a specific object. For example, for a photo album created at a club that eats and walks, the object is specified as “cooking”, and the evaluation item is a category of “cooking images” from “too few” to “too many”. Get a rating of. In this case, if the average value is biased toward “too many”, the composite image that the majority of multiple users are satisfied by recreating the composite image by changing the conditions so as to reduce the image of the dish Images can be created.

  In the present embodiment, a plurality of users do not directly edit a layout of a composite image created using a plurality of images acquired from a plurality of users, but calculate a composite image analysis evaluation value, and A composite image user evaluation value is acquired from the user, a composite image user evaluation total value is calculated, and a composite image is recreated based on the difference between the composite image user evaluation total value and the composite image analysis evaluation value.

  Thus, since a plurality of users do not directly edit the layout of the composite image, even if the number of users increases, the composite image can be recreated, that is, the layout of the composite image can be edited. In addition, since the image processing apparatus 10 instead of the user recreates the composite image, it is possible to create a composite image that can be generally understood by the majority of users without imposing a heavy burden on the manager user.

Note that instead of the composition condition changing unit 36 creating the change condition, the composition condition selecting unit 44 selects the composite image user evaluation total value and the composite image analysis from among the plurality of change conditions stored in the composition condition storage unit 42. One change condition corresponding to the composite image evaluation item whose difference from the evaluation value is larger than the threshold value may be selected.
In this case, the composite image creation unit 26 recreates the composite image based on the change condition selected by the composite condition selection unit 44.

Alternatively, each of a plurality of users may select one change condition from among a plurality of change conditions stored in the synthesis condition storage unit 42.
In this case, the composite image creation unit 26 re-creates a composite image based on one change condition selected by each of a plurality of users from the plurality of change conditions stored in the composite condition storage unit 42. .

Further, the synthesis condition changing unit 36 may create not only one change condition but also two or more change conditions.
In this case, the composite image creation unit 26 recreates two or more composite images based on the two or more change conditions created by the composition condition change unit 36.
Subsequently, the composite image selection number acquisition unit 46 acquires the number of each of the two or more composite images selected by the plurality of users for the two or more composite images recreated by the composite image creation unit 26. The
Then, the composite image information acquisition unit 48 acquires information of the selected composite image having the largest number selected by a plurality of users from two or more composite images acquired by the composite image selection number acquisition unit 46. .

  Next, the operation of the image processing apparatus 10 when a composite image is recreated based on the standard deviation will be described with reference to the flowchart shown in FIG.

  The operations in steps S21 to S25 and step S26 are the same as the operations in steps S1 to S5 and step S8 shown in the flowchart of FIG.

  Subsequently, the standard deviation of the composite image user evaluation values from a plurality of users is calculated by the composite image user evaluation value calculation unit 34 for each composite image evaluation item (step S27).

  Subsequently, the composition condition changing unit 36 confirms whether or not the standard deviation calculated by the composite image user evaluation value calculation unit 34 is larger than the threshold value for each composite image evaluation item (step S28).

  As a result, when the standard deviation is larger than the threshold (Yes in step S28), the synthesis condition changing unit 36 sets at least one of the individual image evaluation condition, the image extraction condition, and the synthesized image creation condition as the standard deviation. Two or more change conditions changed accordingly are created (step S29).

  For example, when the distribution of composite image user evaluation values from a plurality of users is largely divided into two groups, the composite condition changing unit 36 has a composite image analysis evaluation value for each of these two groups. The composite image user evaluation total value of, for example, at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is set to the standard deviation so as to approach the average of the composite image user evaluation values from a plurality of users. Two change conditions changed accordingly are created.

  On the other hand, when the standard deviation is equal to or smaller than the threshold value (No in step S28), it is confirmed whether or not the confirmation of whether the standard deviation is larger than the threshold value is completed for all the composite image evaluation items (step S28). S30).

  As a result, when the confirmation has not been completed for all the composite image evaluation items (No in step S30), it is continuously confirmed whether or not the standard deviation is larger than the threshold value.

On the other hand, when the confirmation is completed for all the composite image evaluation items (Yes in step S30), two or more composite images are generated by the composite image creation unit 26 based on the two or more change conditions created by the composite condition change unit 36. Is recreated (step S31).
Note that when the change condition is not created by the composition condition changing unit 36, the composite image is not recreated.

  Subsequently, two or more composite images recreated by the composite image creation unit 26 are sent to each user terminal device 14 via the network 16 and displayed on the image display unit 60.

  Each user refers to two or more composite images displayed on the image display unit 60 of the terminal device 14 of each user, and one composite image is selected from the two or more composite images via the instruction input unit 62. Select.

  Subsequently, the composite image selection number acquisition unit 46 acquires the number of each of the two or more composite images selected by the plurality of users for the two or more composite images (step S32).

  Subsequently, the composite image information acquisition unit 48 acquires information of a selected composite image that is the composite image having the largest number selected by a plurality of users from two or more composite images (step S33).

  For example, when the composite image evaluation item is the number of images used in the composite image, each user uses the slider bar to select the image used in the composite image as in the example of FIG. The composite image can be evaluated for numbers.

  In this case, when the evaluation of the composite image by the user is completed, the composite image user evaluation value acquisition unit 32 determines the composite image for the number of images used in the composite image from the terminal devices 14 of the plurality of users via the network 16. A composite image user evaluation value obtained by evaluating is obtained.

  Subsequently, the standard deviation of the composite image user evaluation values from a plurality of users is calculated by the composite image user evaluation value calculation unit 34 for the number of images used in the composite image.

  Subsequently, the composition condition changing unit 36 confirms whether or not the standard deviation of the number of images used in the composite image is larger than a threshold value.

  As a result, when the standard deviation is larger than the threshold, the composition condition changing unit 36 changes the number of arranged images in the template used for creating the composite image according to the standard deviation.

  Then, the composite image creating unit 26 recreates the composite image using the template whose number of arranged images has been changed by the composite condition changing unit 36.

  Even when the composite image evaluation item is the number of person images used in the composite image, each user is using the slider bar in the composite image as in the example shown in FIG. The composite image can be evaluated with respect to the number of person images.

  In this case, when the evaluation of the composite image by the user ends, the composite image user evaluation value acquisition unit 32 combines the number of person images used in the composite image from the terminal devices 14 of a plurality of users via the network 16. A composite image user evaluation value obtained by evaluating the image is acquired.

  Subsequently, the standard deviation of the composite image user evaluation values from a plurality of users is calculated by the composite image user evaluation value calculation unit 34 with respect to the number of person images used in the composite image.

  Subsequently, the composition condition changing unit 36 confirms whether or not the standard deviation of the number of person images used in the composite image is larger than a threshold value.

  As a result, when the standard deviation is larger than the threshold, the composition condition changing unit 36 changes the weight of the person image when extracting one or more images from the plurality of images.

  Then, the composite image creation unit 26 re-creates a composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the composition condition change unit 36.

  Even when the composite image evaluation item is the atmosphere of the composite image, each user can evaluate the composite image with respect to the atmosphere of the composite image using the slider bar, as in the example shown in FIG. .

  In this case, when the evaluation of the composite image by the user ends, the composite image obtained by evaluating the composite image with respect to the atmosphere of the composite image from the plurality of user terminal devices 14 via the network 16 by the composite image user evaluation value acquisition unit 32. A user evaluation value is acquired.

  Subsequently, the standard deviation of the composite image user evaluation values from a plurality of users is calculated by the composite image user evaluation value calculation unit 34 for the atmosphere of the composite image.

  Subsequently, the composition condition changing unit 36 confirms whether the standard deviation of the atmosphere of the composite image is larger than the threshold value.

  As a result, when the standard deviation is larger than the threshold value, the synthesis condition changing unit 36 changes the background image of the template used for creating the synthesized image according to the standard deviation.

  Then, the composite image creation unit 26 recreates the composite image using the template whose background image has been changed by the synthesis condition change unit 36.

  In this embodiment, a composite image user evaluation value is acquired from each user, a standard deviation of composite image user evaluation values from a plurality of users is calculated, and two or more composite images are recreated based on the standard deviation. . Then, information on a selected composite image that is a composite image having the largest number selected by a plurality of users from two or more composite images is acquired, and a selected composite image is adopted from the two or more composite images.

Thus, by recreating the composite image based on the standard deviation, even if the distribution of the composite image user evaluation values from a plurality of users varies, the majority of users can be convinced on average. A composite image can be created.
For example, if the average of the evaluations from a plurality of users is “too bright” for the atmosphere of the composite image, the value of the condition for determining the brightness of the composite image can be changed to “ Of course, I don't know if it can be "good" brightness. For example, even when the brightness of the composite image is lowered, there is a possibility that the brightness is lowered too much or the brightness is lowered too little.
Therefore, by re-creating two or more composite images in which the value of the condition for determining the brightness of the composite image is changed and determining a composite image to be adopted by majority vote by a plurality of users, the majority of the plurality of users Can reveal the desired brightness, and can also clarify the value of the condition that determines the brightness of the composite image, which can make the composite image “just right” for the majority of users. it can.

Note that instead of the composition condition changing unit 36 creating the change condition, the composition condition selecting unit 44 selects a composite image whose standard deviation is larger than the threshold from the plurality of change conditions stored in the composition condition storage unit 42. Two or more change conditions corresponding to the item may be selected.
In this case, the composite image creation unit 26 recreates the composite image based on the change condition selected by the composite condition selection unit 44.

Alternatively, each of a plurality of users may select two or more change conditions from among a plurality of change conditions stored in the synthesis condition storage unit 42.
In this case, the composite image creation unit 26 recreates the composite image based on two or more change conditions selected by each of the plurality of users from the plurality of change conditions stored in the composite condition storage unit 42. It is.

Note that the composition condition changing unit 36 may create a change condition based on both the difference between the composite image user evaluation total value and the composite image analysis evaluation value and the standard deviation.
For example, when evaluating the atmosphere of a composite image, even if the difference between the composite image user evaluation total value and the composite image analysis evaluation value is equal to or less than a threshold value, the evaluation from a plurality of users is “too dark” and “too bright. "May be many. In this case, since the composite image is not re-created, the majority of users cannot be satisfied if the re-creation of the composite image is determined based on the difference between the composite image user evaluation total value and the composite image analysis evaluation value. There is a fear.
On the other hand, when the average of the composite image user evaluation values from a plurality of users is calculated as the composite image user evaluation total value, for example, three corresponding to the average, the average + standard deviation, and the average-standard deviation By re-creating a composite image and determining a composite image to be adopted by majority vote by a plurality of users, it is possible to create a composite image that incorporates the opinions of a plurality of users to some extent.

  In addition, each user may evaluate a composite image recreated by the composite image creation unit 26 for each composite image evaluation item, and recreate a composite image. That is, the composite image may be recreated twice or more.

One or more composite image evaluation items may be specified (input) by the manager.
In this case, the composite image user evaluation value acquisition unit 32 acquires a composite image user evaluation value obtained by each user evaluating the composite image for each of the one or more composite image evaluation items specified by the secretary user.

One or more composite image evaluation items may be stored in the composite image evaluation item storage unit 56.
In this case, the composite image user evaluation value acquisition unit 32 acquires a composite image user evaluation value obtained by each user evaluating the composite image for each of the one or more composite image evaluation items stored in the composite image evaluation item storage unit 56. Is done.

Each of the plurality of users may select a composite image evaluation item to be evaluated from one or more composite image evaluation items stored in the composite image evaluation item storage unit 56.
In this case, each of the composite image evaluation items selected by each of a plurality of users from one or more composite image evaluation items stored in the composite image evaluation item storage unit 56 by the composite image user evaluation value acquisition unit 32. A composite image user evaluation value obtained by evaluating the composite image by the user is acquired.
For example, when the secretary user selects a composite image evaluation item to be evaluated, the secretary image user evaluation value acquisition unit 32 causes the secretary to select from among one or more composite image evaluation items stored in the composite image evaluation item storage unit 56. For the composite image evaluation item selected by the user, a composite image user evaluation value obtained by each user evaluating the composite image is acquired.
Further, when the composite image analysis evaluation value exceeds the threshold value, the composite image evaluation value calculation unit 30 recommends to the manager user to evaluate the composite image for the composite image evaluation item exceeding the threshold value. Good. The secretary user is not limited to one person, and may be at least one user among a plurality of users.

Also, the composite image evaluation item creation unit 58 automatically selects a composite image evaluation item for an object whose appearance count in the composite image is larger than the threshold from one or more objects based on the individual image evaluation value of each image. You may create it.
In this case, the composite image user evaluation value acquisition unit 32 acquires a composite image user evaluation value obtained by each user evaluating the composite image for the composite image evaluation item created by the composite image evaluation item creation unit 58.

  Further, the information of the selected composite image is stored in the composite image information storage unit 50, and a plurality of users are based on the information of the selected composite image stored in the composite image information storage unit 50 by the composition condition changing unit 36. At least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition used for creating the next composite image may be changed so as to match the selection tendency of the composite image.

  In addition, for each composite image evaluation item, an average of composite image user evaluation values from a plurality of users is stored in the composite image user evaluation value storage unit, and the composite image creation unit 26 determines each composite image evaluation item. The individual image evaluation condition, the image extraction condition, and the composite image creation condition so that the composite image analysis evaluation value approaches the average of the composite image user evaluation values from a plurality of users stored in the composite image user evaluation value storage unit. The next composite image may be created under the condition that at least one value of is determined.

  If the average of the composite image created under a certain condition and the composite image user evaluation value from a plurality of users who evaluated this composite image for each composite image evaluation item is determined, how much the condition value is determined, It is possible to learn whether a composite image that can satisfy the average satisfaction of a plurality of users can be created. Therefore, by storing the average of the composite image user evaluation values from a plurality of users in the composite image user evaluation value storage unit, the next composite image is more accurately the one closer to the average of the evaluations of the plurality of users. And it can be created quickly.

  Note that, when a composite image is re-created, for each composite image evaluation item, a condition that the difference between the composite image user evaluation total value and the composite image analysis evaluation value is equal to or less than a threshold, that is, evaluation from a plurality of users Value of the condition that the average of “just right” is stored in the composite image user evaluation value storage unit, the composite image user evaluation total value and the composite image analysis evaluation value stored in the composite image user evaluation value storage unit, The next synthesized image may be created under the condition that the difference between the two is equal to or less than the threshold.

  The image processing apparatus 10 may create a composite image such as a photo album (photo book) on a paper medium by printing and binding, and is not limited to this, and includes an image including a still image and a moving image on a template. And a composite image such as an electronic album to be viewed on the display may be created. Furthermore, a composite image such as a slide show that sequentially displays various contents including a still image and a moving image acquired from the user may be created without using a template.

  It is not essential that the image processing apparatus 10 includes the server 12 and the terminal devices 14 of a plurality of users. For example, an apparatus including all the components of the server 12 and the terminal device 14 can be configured. In this case, exchange between the network 16 and the server 12 and the terminal device 14 of the user is not necessary. For example, direct exchange from the user to the image processing apparatus 10 or from the image processing apparatus 10 to the user. It can be carried out.

  Image acquisition unit 18, individual image analysis unit 20, individual image evaluation value calculation unit 22, image extraction unit 24, composite image creation unit 26, composite image analysis unit 28, composite image evaluation value calculation unit 30, composite image user evaluation value acquisition Unit 32, composite image user evaluation value calculation unit 34, composition condition change unit 36, personal information acquisition unit 38, weight calculation unit 40, composition condition storage unit 42, composition condition selection unit 44, composition image selection number acquisition unit 46, composition Image information acquisition unit 48, composite image information storage unit 50, object detection unit 52, object number acquisition unit 54, composite image evaluation item storage unit 56, composite image evaluation item creation unit 58, and composite image user evaluation value storage unit 59 It is not essential that the server 12 is provided, and at least one of them may be provided in the user terminal device 14.

  In the apparatus of the present invention, each component included in the apparatus may be configured by dedicated hardware, or each component may be configured by a programmed processor and a computer including a processor. The method of the present invention can be implemented, for example, by a program for causing a computer to execute each step. It is also possible to provide a computer-readable recording medium in which this program is recorded.

  Although the present invention has been described in detail above, the present invention is not limited to the above-described embodiment, and it is needless to say that various improvements and modifications may be made without departing from the gist of the present invention.

10 image processing device 12 server 14 terminal device (client)
DESCRIPTION OF SYMBOLS 16 Network 18 Image acquisition part 20 Individual image analysis part 22 Individual image evaluation value calculation part 24 Image extraction part 26 Composite image creation part 28 Composite image analysis part 30 Composite image evaluation value calculation part 32 Composite image user evaluation value acquisition part 34 Composite image User evaluation value calculation unit 36 Composition condition change unit 38 Personal information acquisition unit 40 Weight calculation unit 42 Composition condition storage unit 44 Composition condition selection unit 46 Composition image selection number acquisition unit 48 Composition image information acquisition unit 50 Composition image information storage unit 52 Object Detection unit 54 Object number acquisition unit 56 Composite image evaluation item storage unit 58 Composite image evaluation item creation unit 59 Composite image user evaluation value storage unit 60 Image display unit 62 Instruction input unit 64 Slider bar 66 Slider

Claims (31)

An image acquisition unit for acquiring a plurality of images from a plurality of users;
An individual image analysis unit for analyzing each of the plurality of images;
An individual image evaluation value calculation unit that calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on the analysis result of each image;
An image extraction unit that extracts one or more images corresponding to one or more image extraction conditions from the plurality of images based on the individual image evaluation values of the respective images;
A composite image creating unit that creates a composite image by combining the one or more images based on one or more composite image creation conditions;
A composite image analysis unit for analyzing the composite image;
A composite image evaluation value calculation unit that calculates one or more composite image analysis evaluation values obtained by evaluating the composite image for each of one or more composite image evaluation items based on the analysis result of the composite image;
A composite image user evaluation value acquisition unit that acquires each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items from the plurality of users;
For each of the composite image evaluation items, a composite image user evaluation value calculation unit that calculates a composite image user evaluation total value based on composite image user evaluation values from the plurality of users;
For each of the composite image evaluation items, when the difference between the composite image user evaluation total value and the composite image analysis evaluation value is larger than a threshold, the composite image analysis evaluation value approaches the composite image user evaluation total value. And a synthesis condition change unit that creates a change condition in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference,
The composite image creation unit recreates the composite image based on the change condition created by the composition condition change unit.   The synthesis condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value. Or when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value increases according to the composite image user evaluation total value. The image processing apparatus according to claim 1, wherein a change condition is created by changing at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of images used in the composite image,
The synthesis condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value. Or when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value increases according to the composite image user evaluation total value. Change the number of arranged images in the template used to create the composite image,
The image processing apparatus according to claim 1, wherein the composite image creation unit re-creates the composite image using a template in which the number of arranged images has been changed by the synthesis condition change unit. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image with respect to the number of person images used in the composite image,
The synthesis condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value. Or when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value increases according to the composite image user evaluation total value. As described above, the weight of the person image when extracting the one or more images from the plurality of images is changed,
The composite image creation unit re-creates the composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the synthesis condition change unit. Item 4. The image processing device according to any one of Items 1 to 3. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the atmosphere of the composite image,
The synthesis condition changing unit may reduce the composite image analysis evaluation value according to the composite image user evaluation total value when the composite image user evaluation total value is larger than the threshold with respect to the composite image analysis evaluation value. Or when the composite image user evaluation total value is smaller than the threshold with respect to the composite image analysis evaluation value, the composite image analysis evaluation value increases according to the composite image user evaluation total value. Change the background image of the template used to create the composite image,
5. The image processing apparatus according to claim 1, wherein the composite image creation unit re-creates the composite image using a template in which the background image has been changed by the synthesis condition change unit.   The composite image user evaluation value calculation unit calculates a sum, an average, or a product of the composite image user evaluation values from the plurality of users as the composite image user evaluation total value. An image processing apparatus according to 1. A weight calculation unit for calculating the weight of each of the plurality of users;
The composite image user evaluation value calculation unit calculates a sum, an average, or a product of the composite image user evaluation values from the plurality of users weighted according to the weight of each user as the composite image user evaluation total value The image processing apparatus according to claim 1, wherein: A personal information acquisition unit for acquiring the personal information of each user;
The image processing apparatus according to claim 7, wherein the weight calculation unit calculates the weight of each of the plurality of users based on the personal information of each of the users.   The image processing device according to claim 7, wherein the weight calculation unit calculates the weight of each of the plurality of users such that the weight of the user whose more images are acquired by the image acquisition unit becomes heavier. .   The weight calculation unit calculates the weight of each of the plurality of users so that the weight of the user whose composite image user evaluation value is acquired at an earlier timing by the composite image user evaluation value acquisition unit becomes heavier. The image processing apparatus according to claim 7. A synthesis condition storage unit that stores a plurality of change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference;
A synthesis condition selection unit that selects one change condition corresponding to a composite image evaluation item in which the difference is larger than the threshold value from among a plurality of change conditions stored in the synthesis condition storage unit;
The image processing apparatus according to claim 1, wherein the composite image creation unit re-creates the composite image based on a change condition selected by the synthesis condition selection unit. A synthesis condition storage unit that stores a plurality of change conditions obtained by changing at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition according to the difference;
The composite image creation unit re-creates the composite image based on one change condition selected by each of the plurality of users from a plurality of change conditions stored in the composition condition storage unit. Item 11. The image processing apparatus according to any one of Items 1 to 10. The composition condition changing unit creates two or more change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference,
The composite image creation unit re-creates two or more composite images based on the two or more change conditions created by the composition condition change unit,
A composite image selection number acquisition unit that acquires the number of each of the two or more composite images selected by the plurality of users with respect to the two or more composite images;
13. The composite image information acquisition unit according to claim 1, further comprising: a composite image information acquisition unit configured to acquire information on a selected composite image having the largest number selected by the plurality of users from the two or more composite images. Image processing device. The composite image user evaluation value calculation unit further calculates a standard deviation of the composite image user evaluation values from the plurality of users,
The synthesis condition changing unit creates two or more change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the synthesized image creation condition is changed according to the difference and the standard deviation,
The composite image creation unit re-creates two or more composite images based on the two or more change conditions created by the composition condition change unit,
A composite image selection number acquisition unit that acquires the number of each of the two or more composite images selected by the plurality of users with respect to the two or more composite images;
14. A composite image information acquisition unit that acquires information of a selected composite image having the largest number selected by the plurality of users from the two or more composite images. 14. Image processing apparatus. An image acquisition unit for acquiring a plurality of images from a plurality of users;
An individual image analysis unit for analyzing each of the plurality of images;
An individual image evaluation value calculation unit that calculates an individual image evaluation value obtained by evaluating each image with respect to one or more individual image evaluation conditions based on the analysis result of each image;
An image extraction unit that extracts one or more images corresponding to one or more image extraction conditions from the plurality of images based on the individual image evaluation values of the respective images;
A composite image creating unit that creates a composite image by combining the one or more images based on one or more composite image creation conditions;
A composite image user evaluation value acquisition unit that acquires each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items from the plurality of users;
For each composite image evaluation item, a composite image user evaluation value calculation unit that calculates a standard deviation of composite image user evaluation values from the plurality of users,
For each of the composite image evaluation items, when the standard deviation is larger than a threshold, at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the standard deviation A composition condition changing unit for creating two or more changed conditions,
Compositing that acquires the number of each of the two or more composite images selected by the plurality of users for the two or more composite images recreated based on the two or more change conditions by the composite image creation unit An image selection number acquisition unit;
An image processing apparatus comprising: a composite image information acquisition unit that acquires information of a selected composite image having the largest number selected by the plurality of users from the two or more composite images. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the number of images used in the composite image,
When the standard deviation is larger than the threshold, the synthesis condition changing unit changes the number of arranged images in the template used for creating the synthesized image according to the standard deviation,
The image processing apparatus according to claim 15, wherein the composite image creation unit re-creates the composite image using a template in which the number of arranged images has been changed by the synthesis condition change unit. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image with respect to the number of person images used in the composite image,
The composition condition changing unit, when the standard deviation is larger than the threshold, changes the weight of the person image when extracting the one or more images from the plurality of images according to the standard deviation,
The composite image creation unit re-creates the composite image using one or more images extracted from the plurality of images according to the weight of the person image changed by the synthesis condition change unit. Item 15. The image processing apparatus according to Item 15 or 16. The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the atmosphere of the composite image,
When the standard deviation is larger than the threshold, the synthesis condition changing unit changes a background image of a template used for creating the synthesized image according to the standard deviation,
The image processing apparatus according to any one of claims 15 to 17, wherein the composite image creation unit re-creates the composite image using a template in which the background image has been changed by the synthesis condition change unit. A synthesis condition storage unit that stores a plurality of change conditions in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the standard deviation;
A synthesis condition selection unit that selects two or more change conditions corresponding to a composite image evaluation item in which the standard deviation is larger than the threshold value from among a plurality of change conditions stored in the synthesis condition storage unit;
The image processing apparatus according to any one of claims 15 to 18, wherein the composite image creation unit re-creates the composite image based on a change condition selected by the synthesis condition selection unit. A synthesis condition storage unit that stores a plurality of change conditions obtained by changing at least one of the individual image evaluation condition, the image extraction condition, and the synthesized image creation condition according to the standard deviation;
The composite image creation unit re-creates the composite image based on two or more change conditions selected by each of the plurality of users from a plurality of change conditions stored in the composition condition storage unit. Item 19. The image processing device according to any one of Items 15 to 18. An object detection unit that detects an object other than a person shown in each of the images based on an analysis result of each of the images, and creates a list of objects shown in the plurality of images;
An object number acquisition unit for acquiring the number of each object selected by the plurality of users from the list of objects;
The synthesis condition changing unit further extracts the one or more images from the plurality of images according to the number of each object selected by the plurality of users from the list of objects. Change the weight of the image in which each object is shown,
The composite image creation unit recreates the composite image using one or more images extracted from the plurality of images according to the weight of the object changed by the synthesis condition change unit. 21. The image processing apparatus according to any one of 1 to 20. A composite image evaluation item storage unit for storing the one or more composite image evaluation items;
The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for each of the one or more composite image evaluation items stored in the composite image evaluation item storage unit. 22. The image processing device according to any one of items 21 to 21. A composite image evaluation item storage unit for storing the one or more composite image evaluation items;
The composite image user evaluation value acquisition unit includes the composite image for a composite image evaluation item selected by each of the plurality of users from the one or more composite image evaluation items stored in the composite image evaluation item storage unit. The image processing apparatus according to any one of claims 1 to 21, which acquires a composite image user evaluation value obtained by evaluating   The composite image evaluation value calculation unit evaluates the composite image for a composite image evaluation item that exceeds the threshold when the composite image analysis evaluation value exceeds the threshold. The image processing apparatus according to claim 1, wherein the image processing apparatus recommends to at least one user. A composite image evaluation item creating unit that creates the composite image evaluation item for an object whose appearance count in the composite image is greater than a threshold value from among one or more objects based on the individual image evaluation value of each image; Prepared,
The composite image user evaluation value acquisition unit acquires a composite image user evaluation value obtained by evaluating the composite image for the composite image evaluation item created by the composite image evaluation item creation unit. The image processing apparatus according to item. A composite image information storage unit for storing information of the selected composite image;
The synthesis condition changing unit creates a next synthesized image so as to match the selection tendency of the synthesized image by the plurality of users based on information of the selected synthesized image stored in the synthesized image information storage unit. The image processing apparatus according to claim 13, wherein at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition used in the process is changed. A composite image user evaluation value storage unit that stores an average of composite image user evaluation values from the plurality of users for each of the composite image evaluation items,
The composite image creation unit approaches the average of the composite image user evaluation values from the plurality of users stored in the composite image user evaluation value storage unit for each composite image evaluation item. The next composite image is created under a condition in which at least one value of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is determined as described above. An image processing apparatus according to 1. An image acquisition unit acquiring a plurality of images from a plurality of users;
An individual image analysis unit analyzing each of the plurality of images;
An individual image evaluation value calculating unit calculating an individual image evaluation value obtained by evaluating each of the images with respect to one or more individual image evaluation conditions based on the analysis result of each of the images;
An image extraction unit extracting one or more images corresponding to one or more image extraction conditions from the plurality of images based on the individual image evaluation values of the respective images;
A step of generating a composite image obtained by combining the one or more images based on one or more composite image generation conditions;
A step of a composite image analysis unit analyzing the composite image;
A composite image evaluation value calculation unit calculating each of one or more composite image analysis evaluation values obtained by evaluating the composite image for each of one or more composite image evaluation items based on an analysis result of the composite image;
A composite image user evaluation value acquisition unit acquiring, from the plurality of users, each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items;
A step of calculating a composite image user evaluation total value based on a composite image user evaluation value from the plurality of users for each of the composite image evaluation items;
When the difference between the composite image user evaluation total value and the composite image analysis evaluation value is greater than a threshold value for each of the composite image evaluation items, the composite condition changing unit determines that the composite image analysis evaluation value is the composite image Creating a change condition in which at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition is changed according to the difference so as to approach the user evaluation total value;
And a step of re-creating the composite image based on the change condition created by the step of creating the change condition. An image acquisition unit acquiring a plurality of images from a plurality of users;
An individual image analysis unit analyzing each of the plurality of images;
An individual image evaluation value calculating unit calculating an individual image evaluation value obtained by evaluating each of the images with respect to one or more individual image evaluation conditions based on the analysis result of each of the images;
An image extraction unit extracting one or more images corresponding to one or more image extraction conditions from the plurality of images based on the individual image evaluation values of the respective images;
A step of generating a composite image obtained by combining the one or more images based on one or more composite image generation conditions;
A composite image user evaluation value acquisition unit acquiring, from the plurality of users, each of one or more composite image user evaluation values obtained by evaluating the composite image for each of the one or more composite image evaluation items;
A step of calculating a standard deviation of the composite image user evaluation values from the plurality of users for each of the composite image evaluation items;
For each of the composite image evaluation items, when the standard deviation is greater than a threshold, the composite condition changing unit sets at least one of the individual image evaluation condition, the image extraction condition, and the composite image creation condition to the Creating two or more change conditions changed according to the standard deviation;
The two or more composite images selected by the plurality of users with respect to the two or more composite images recreated based on the two or more change conditions by the step of creating the composite image Obtaining the number of each of the composite images;
A composite image information acquisition unit including a step of acquiring information of a selected composite image having the largest number selected by the plurality of users from the two or more composite images.   30. A program for causing a computer to execute each step of the image processing method according to claim 28 or 29.   A computer-readable recording medium on which a program for causing a computer to execute each step of the image processing method according to claim 28 or 29 is recorded.