WO2018122959A1

WO2018122959A1 - Photographer identification system, photographer identification method, and program

Info

Publication number: WO2018122959A1
Application number: PCT/JP2016/088887
Authority: WO
Inventors: 俊二菅谷
Original assignee: 株式会社オプティム
Priority date: 2016-12-27
Filing date: 2016-12-27
Publication date: 2018-07-05

Abstract

[Problem] To identify a photographer by simultaneously using an inner camera and an outer camera of a smart device. [Solution] Provided is a photographer identification system which identifies a photographer by simultaneously using an inner camera and an outer camera of a smart device, and which is provided with: an inner image acquisition means for acquiring an inner image photographed by the inner camera; an outer image acquisition means for acquiring an outer image photographed by the outer camera; an image analysis means for performing image analysis on the acquired inner image; a photographer identification means for identifying the photographer on the basis of the result of the image analysis; and a storage means in which the acquired outer image is associated with the identified photographer, and stored.

Description

Photographer identification system, photographer identification method and program

The present invention relates to a photographer identification system, a photographer identification method, and a program for identifying a photographer of an image photographed by a smart device.

In recent years, smart devices having camera functions such as digital cameras, smartphones, tablets, and smart glasses have become widespread, and photography using these smart devices has been performed. For example, a system is provided that makes it possible to classify captured images as easily as possible when a user classifies captured images for each photographer with respect to captured images that cannot be specified by the photographer when captured by these smart devices. (Patent Document 1).

Japanese Patent Application No. 2012-198034

However, the system of Patent Document 1 is premised on classification based on operation input from a user, and there is a problem that a photographer cannot be identified without such user operation. Therefore, the inventor pays attention to the fact that recent smart devices are equipped with two cameras, an inner camera and an outer camera, and it is not possible to classify captured images using these cameras. I thought.

In view of the above problems, an object of the present invention is to provide a photographer identification system, a photographer identification method, and a program for identifying a photographer by simultaneously using an inner camera and an outer camera of a smart device.

The present invention provides the following solutions.

The invention according to the first aspect is a photographer identification system that identifies a photographer by simultaneously using an inner camera and an outer camera of a smart device, and is an inner image that acquires an inner image captured by the inner camera. An acquisition means; an outer image acquisition means for acquiring an outer image taken by the outer camera; an image analysis means for image analysis of the acquired inner image; and the photographer based on the result of the image analysis. There is provided a photographer identification system comprising: a photographer identification unit for identifying; a storage unit for storing the acquired outside image and the identified photographer in association with each other.

The invention according to the first aspect is a photographer identification method for identifying a photographer by simultaneously using an inner camera and an outer camera of a smart device, and an inner image for acquiring an inner image photographed by the inner camera An acquisition step; an outer image acquisition step of acquiring an outer image taken by the outer camera; an image analysis step of analyzing the acquired inner image; and the photographer based on the result of the image analysis. There is provided a photographer specifying method comprising a photographer specifying step for specifying, a storing step for storing the acquired outside image and the specified photographer in association with each other.

The invention according to the first feature is an inner image acquisition step of acquiring an inner image taken by the inner camera by simultaneously using the inner camera and the outer camera of the smart device in the computer, and taken by the outer camera. An outer image acquisition step for acquiring an outer image, an image analysis step for image analysis of the acquired inner image, a photographer specifying step for specifying the photographer based on the result of the image analysis, and the acquired outer image And a storage step of storing the specified photographer in association with each other.

According to the present invention, when classifying images taken by a camera, the photographer is identified by using two cameras without any input from the user, and the resulting image is automatically classified. It becomes possible to do.

Hereinafter, the best mode for carrying out the present invention will be described. This is merely an example, and the technical scope of the present invention is not limited to this.

The photographer identification system of the present invention is a system for identifying a photographer by using an inner camera and an outer camera of a smart device at the same time. The smart device may be a device having a camera function such as a digital camera, a smartphone, a tablet, and a smart glass or a wearable terminal. In general, a smart device is configured by a flat surface (six surfaces) and includes a display display unit that displays an image on the surface, and the back side of the surface is the back surface. Here, the camera installed on the front surface is the inner camera (the direction of the photographer), and the camera installed on the back surface is the outer camera (the direction in which the object to be photographed is photographed, opposite to the photographer). To do.

The photographer identification system includes an inner image acquisition unit, an outer image acquisition unit, an image analysis unit, a photographer identification unit, and a storage unit that are realized by a control unit reading a predetermined program. Similarly, an embedding unit, an information display unit, an authentication unit, a signature reception unit, a signature display unit, and a copyright management unit may be provided. These may be realized by an application installed on a smart device or a network partially realized by a computer via a network (may be a cloud type in which two or more means are realized by another computer). May be. Each means described above may be realized by a single computer or may be realized by two or more computers (for example, a server and a terminal).

The inner image acquisition means acquires the inner image taken by the inner camera facing inward (the direction of the photographer). The inner image is a still image or a moving image captured by the photographer. It is assumed that the inner image and the outer image are taken simultaneously. That is, it is desirable that both the inner and outer images are taken with one shooting shutter button. Here, the acquisition may be that the smart device itself acquires an image captured by the smart device camera, or that another computer acquires an image acquired by the smart device via a network. May be.

The outer image acquisition means acquires an outer image captured by an outer camera facing outward (opposite to the photographer). The outer image is a still image or a moving image taken by the photographer. Similarly, here, acquisition may mean that the smart device itself acquires an image taken by the smart device camera, or another computer acquires an image acquired by the smart device via a network. It may be.

The image analysis means performs image analysis by collating the acquired inner image with a predetermined image database (a database stored in the image processing apparatus or another computer may be referred to). Here, as an algorithm for image analysis, the image may be identified by extracting features that stand out from the image captured by the photographer. Alternatively, three-dimensional face recognition that attempts to achieve accuracy in consideration of invisible parts may also be used. In addition, skin texture analysis may be used in which skin details are applied to face recognition. In the case of a smart glass, iris (iris) recognition (one of biometric authentication and pattern recognition for an image of the iris resolution of an individual's eyes) may be used. For example, when recognizing a photographer, the image analysis means may identify the photographer from data obtained by learning a person name and an image as teacher data in advance. That is, the accuracy of image analysis may be increased by machine learning, deep learning, or the like using AI (artificial intelligence).

The photographer specifying means specifies the photographer based on the result of image analysis. The algorithm for identifying the photographer may identify the photographer by collating with the face image database in which the photographer's face image is registered, or collating with the iris image database in which the iris image is registered. The photographer may be specified. These databases may be databases stored inside the smart device, or may be external databases stored in a computer to which the smart device is connected via a network.

The storage means temporarily stores the acquired outside image in association with the specified photographer's name and user ID. For example, it is stored as a correspondence table.

The embedding unit embeds the stored photographer information in the information (attribute information such as properties) of the photographed and stored outside image. For example, photographer data such as the photographer's name and user ID is embedded in the outer image data. By embedding in this manner, the photographer data can be read by reading the outside image data, so that the one-to-one correspondence between the outside image and the photographer is possible. Here, the photographer data may include additional data such as a photographer's organization and family in addition to the photographer's name and user ID. These additional data are stored in correspondence with the name and user ID of the photographer in a database stored in the smart device or a computer connected via a network. The photographer data is read.

The information display means displays the stored photographer information (photographer data) on the stored outer image. For example, in addition to the name of the photographer, the affiliated organization as additional data is displayed on the outer image. Here, as an example, in order to clearly express who photographed in the image, information (signature information) that becomes a signature indicating the photographer of the image is displayed like a signature indicating the creator of the painting. Also good. In other words, by displaying the signature information as a signature, it is possible to clearly show the photographer to other people by showing who has taken the image as a signature.

The authentication unit authenticates the specified photographer based on the stored result. For example, as a result of authentication, the stored outer image is deleted when the authentication is not successful, and the stored outer image is maintained when the authentication is successful. As an authentication algorithm, authentication may be performed by collating a face authentication list, an iris authentication list, or the like. These lists may be internal lists or external lists.

The signature accepting unit accepts a signature input from an authenticated photographer.

The signature display means displays the received signature on the stored outside image. For example, the photographer's signature is displayed on the image, like the signature of the creator of the painting. That is, displaying the signature makes it possible to clarify who has taken the image.

The copyright management means manages the copyright of the outer image based on the stored photographer data and additional data. Since the outside image and the specified photographer are stored in association with each other, when a user other than the specified photographer processes and edits the outside image, or performs copying, this processing / The editing software may issue a warning by comparing the photographer and user information.
[Description of operation]

Next, the method for identifying the photographer will be described. Although the description is duplicated in each of the above-described means, only the portions that need to be explained as processes will be supplementarily explained. The photographer is specified by using the inner camera and the outer camera of the smart device at the same time. A smart device refers to a device having a camera function, such as a digital camera, a smartphone, a tablet, and a smart glass.

The photographer identification method includes an inner image acquisition step, an outer image acquisition step, an image analysis step, a photographer identification step, and a storage step. Further, it includes at least an embedding step, an information display step, an authentication step, a signature reception step, a signature display step, and a copyright management step.

In the inner image acquisition step, an inner image captured by the inner camera facing inward (the direction of the photographer) is acquired. The inner image is a still image or a moving image captured by the photographer.

In the outer image acquisition step, an outer image captured by an outer camera facing outward (in the direction opposite to the photographer) is acquired. The outer image is a still image or a moving image taken by the photographer.

The image analysis step analyzes the acquired inner image. Recognize the photographer through image analysis. The image analysis algorithm may be identified by extracting features that stand out from the image captured by the photographer. Three-dimensional face recognition that attempts to achieve accuracy in consideration of invisible parts may also be used. Skin texture analysis may be used that applies skin appearance details to face recognition. In the case of smart glasses, iris recognition may be used. Any algorithm that can be recognized by image analysis is possible.

The photographer specifying step specifies the photographer based on the result of image analysis. The specific algorithm may specify a photographer by comparing with a face image database in which face images are registered, or specify a photographer by comparing with an iris image database in which iris images are registered. Good. These databases may be internal databases or external databases.

In the storing step, the acquired outside image and the specified photographer are stored in association with each other. For example, it is stored as a correspondence table.

In the embedding step, the stored photographer information is embedded in the stored outside image information. For example, the photographer data is embedded in the outside image data. Since the photographer data can be read by reading the outside image data, the outside image and the photographer can be in one-to-one correspondence.

In the information display step, the photographer's information is combined with the stored (embedded) image in a visible state, and when the stored outer image is displayed, the photographer's information is also displayed. indicate. For example, when an image is displayed, the photographer's name and organization are also displayed on the outer image. Information such as a signature indicating the creator of the picture is displayed instead of the signature indicating the photographer of the image. That is, by displaying information that can be used instead of a signature, it is possible to clarify who has taken the image.

In the authentication step, the identified photographer is authenticated based on the stored result. For example, as a result of authentication, the stored outer image is deleted when the authentication is not successful, and the stored outer image is maintained when the authentication is successful. As an authentication algorithm, authentication may be performed by collating a face authentication list, an iris authentication list, or the like. These lists may be internal lists or external lists.

The signature acceptance step accepts an input of a signature from an authenticated photographer.

In the signature display step, the received signature is displayed on the stored outer image. For example, the photographer's signature is displayed on the image, like the signature of the creator of the painting. That is, displaying the signature makes it possible to clarify who has taken the image.

In the copyright management step, the copyright of the outer image is managed based on the stored result. Since the outside image and the specified photographer are stored in association with each other, the copyright can be clearly managed. If a copyright infringement is found, it may have a function of warning a person who has infringed.

The means and functions described above are realized by a computer (including a CPU, an information processing apparatus, and various terminals) reading and executing a predetermined program. The program may be, for example, an application installed on a computer, or may be in the form of SaaS (software as a service) provided from a computer via a network, for example, a flexible disk, a CD It may be provided in a form recorded on a computer-readable recording medium such as a CD-ROM (DVD-ROM, DVD-RAM, etc.). In this case, the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, stores it, and executes it. The program may be recorded in advance in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to a computer via a communication line.

As a specific algorithm for learning such as the image analysis described above, a nearest neighbor method, a naive Bayes method, a decision tree, a support vector machine, or the like may be used. Further, deep learning may be used in which a characteristic amount for learning is generated by using a neural network.

As mentioned above, although embodiment of this invention was described, this invention is not limited to these embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

Claims

A photographer identification system for identifying a photographer by using an inner camera and an outer camera of a smart device at the same time,
An inner image acquisition means for acquiring an inner image taken by the inner camera;
Outer image acquisition means for acquiring an outer image captured by the outer camera;
Image analysis means for image analysis of the acquired inner image;
Based on the result of the image analysis, a photographer identifying means for identifying the photographer;
A photographer specifying system comprising storage means for storing the acquired outside image and the specified photographer in association with each other.
The photographer specifying system according to claim 1, further comprising an embedding unit that embeds the stored photographer information in the stored outside image information.
The photographer specifying system according to claim 2, further comprising information display means for displaying information of the embedded photographer when displaying the stored outer image.
2. The photographer specifying system according to claim 1, further comprising authentication means for authenticating the specified photographer based on the stored result.
5. The photographer specifying system according to claim 4, wherein when the authentication is not successful, the stored outer image is deleted, and when the authentication is successful, the stored outer image is maintained.
Signature accepting means for accepting an input of a signature from the authenticated photographer;
The photographer specifying system according to claim 4, further comprising signature display means for displaying the received signature on the stored outside image.
The photographer specifying system according to claim 1, further comprising copyright management means for managing the copyright of the outer image based on the stored result.
The photographer identification system according to claim 7, wherein the copyright management means warns a person who has infringed when a copyright infringement is found.
The photographer specifying system according to claim 1, wherein the photographer specifying means specifies the photographer by collating with a face image database or an iris image database.
A photographer identification method for identifying a photographer by using an inner camera and an outer camera of a smart device at the same time,
An inner image acquisition step of acquiring an inner image taken by the inner camera;
An outer image acquisition step of acquiring an outer image taken by the outer camera;
An image analysis step for image analysis of the acquired inner image;
A photographer identification step for identifying the photographer based on the result of the image analysis;
A photographer specifying method comprising: a storing step of storing the acquired outside image and the specified photographer in association with each other.
On the computer,
Using the smart device's inner and outer cameras at the same time,
An inner image acquisition step of acquiring an inner image taken by the inner camera;
An outer image acquisition step of acquiring an outer image captured by the outer camera;
An image analysis step for image analysis of the acquired inner image;
A photographer identification step for identifying the photographer based on the result of the image analysis;
A program for executing the storage step of storing the acquired outside image and the specified photographer in association with each other.