Information processing device, method, and medium
Provided is an information processing device including: a specifying unit that specifies a plurality of parts where a character string is described in image data; a layout information acquiring unit that acquires, for each of the plurality of specified parts, layout information including the position of the part in the image data and the size of described characters; and an estimating unit that calculates, for each of the plurality of parts, a likelihood as a character string having a predetermined attribute that can be included in the image data based on a positional relationship with other parts and on a size relationship with other parts, and estimates a part where a character string having the predetermined attribute is described based on the likelihood.
1. An information processing device comprising:
a specifying unit to specify a plurality of parts where a character string is described in image data;
a type determining unit to determine a type of the image data;
a layout information acquiring unit to acquire, for each of the plurality of parts specified by the specifying unit, layout information including a position of the part in the image data and a size of described characters; and
an estimating unit to calculate, for each of the plurality of parts, a likelihood as a character string having a predetermined attribute that can be included in the image data based on a positional relationship with other parts and on a size relationship with other parts, and estimating a part where a character string having the predetermined attribute is described based on the likelihood,
wherein the estimating unit estimates a part where a character string having the predetermined attribute is described, by using an equation for calculating the likelihood indicating a possibility of being a character string suitable for classifying image data of the type determined, the equation is chosen according to the type determined by the type determining unit.
2. The information processing device according to claim 1 , further comprising:
a character recognizing unit to perform character recognition on at least a part estimated by the estimating unit to be a part where a character string having the predetermined attribute is described in the image data; and
a name generating unit to generate, using the character string acquired as a result of the character recognition, a name of the image data or data generated based on the image data.
3. The information processing device according to claim 1 , wherein
the type that is determined by the type determining unit includes a document, and
when the type that has been determined by the type determining unit is a document, the estimating unit calculates, for each of the plurality of parts, the likelihood as a character string including a title or a character string including a date, and estimates a part where a character string including a title or a character string including a date is described based on the likelihood.
4. The information processing device according to claim 1 , wherein
the type that is determined by the type determining unit includes a name card, and
when the type that has been determined by the type determining unit is a name card, the estimating unit calculates, for each of the plurality of parts, the likelihood as a character string including a name or a character string including an organization name, and estimates a part where a character string including a name or a character string including an organization name is described based on the likelihood.
5. The information processing device according to claim 1 , wherein
the type that is determined by the type determining unit includes a receipt, and
when the type that has been determined by the type determining unit is a receipt, the estimating unit calculates, for each of the plurality of parts, the likelihood as a character string including a store name or a character string including a date, and estimates a part where a character string including a store name or a character string including a date is described based on the likelihood.
6. The information processing device according to claim 1 , wherein
the type that is determined by the type determining unit includes at least any of a document, a name card, a receipt, and a photograph/illustration.
7. The information processing device according to claim 2 , further comprising:
a displaying unit to display the plurality of parts specified by the specifying unit so that the part estimated by the estimating unit to be a part where a character string having the predetermined attribute is described is recognizable by a user;
a user input accepting unit to accept an input of a modification by the user having confirmed display contents by the displaying unit; and
a modifying unit to modify, based on the input of the modification, a character string used to generate a name by the name generating unit.
8. The information processing device according to claim 7 , wherein
the estimating unit calculates the likelihood by a process including an operation of adjusting a score related to the positional relationship and a score related to the size relationship using parameters, and
the information processing device further comprises a parameter updating unit to update the parameters so that, when the input of the modification by the user is accepted by the user input accepting unit, the likelihood that is calculated with respect to a part specified by the user to be a part where a character string having the predetermined attribute is described takes a higher value.
9. The information processing device according to claim 7 , further comprising:
a layout information accumulating unit to accumulate, when the input of the modification by the user is accepted by the user input accepting unit, a part specified by the user to be a part where a character string having the predetermined attribute is described and the layout information acquired by the layout information acquiring unit in association with each other; and
a correcting unit to correct, when layout information of the image data matches or approximates the layout information accumulated by the layout information accumulating unit, the part estimated by the estimating unit based on the part that is associated with the matching or approximating layout information and is specified by the user to be a part where a character string having the predetermined attribute is described.
10. The information processing device according to claim 7 , wherein
the estimating unit corrects the likelihood based on a result of a collation between a character string acquired by the character recognizing unit and format information registered in advance.
11. The information processing device according to claim 10 , wherein
the user input accepting unit further accepts a user input of a character string having the predetermined attribute, and
the information processing device further comprises a format information registering unit to register, when the user input of the character string having the predetermined attribute is accepted by the user input accepting unit, format information that is specified from the accepted character string.
12. The information processing device according to claim 1 , further comprising an image receiving unit to receive, via a network, image data which is transmitted from and acquired by an imaging device.
13. The information processing device according to claim 12 , further comprising
a user authenticating unit to authenticate a user of the imaging device, wherein
the image receiving unit receives, via the network, image data which is transmitted from and acquired by the user-authenticated imaging device.
14. The information processing device according to claim 13 , further comprising:
a permission acquiring unit to acquire, from a predetermined service that provides a service via a network, access permission to an account of the user in the predetermined service; and
a data transmitting unit to transmit, when the image data is determined to be image data of a predetermined type by the type determining unit, the image data or data generated based on the image data to the account of the user in the predetermined service, using the access permission acquired from the predetermined service with respect to a user who has been authenticated as the user of the imaging device that has transmitted the image data.
15. The information processing device according to claim 14 , wherein
the user authenticating unit further authenticates a user of a user terminal, and
the permission acquiring unit transmits an access permission request to the predetermined service and acquires the access permission in accordance with a request transmitted from the user-authenticated user terminal.
16. The information processing device according to claim 13 , wherein
the user authenticating unit authenticates a user of the imaging device in response to an authentication request which is transmitted by the imaging device and which includes authentication information, the notification of which is issued from a user terminal to the imaging device.
17. A method that causes a computer to execute:
specifying a plurality of parts where a character string is described in image data;
determining a type of the image data;
acquiring, for each of the plurality of parts specified in the specification of the plurality of parts, layout information including a position of the part in the image data and a size of described characters; and
calculating, for each of the plurality of parts, a likelihood as a character string having a predetermined attribute that can be included in the image data based on a positional relationship with other parts and on a size relationship with other parts, and estimating a part where a character string having the predetermined attribute is described based on the likelihood,
wherein the calculating involves estimating a part where a character string having the predetermined attribute is described, by using an equation for calculating the likelihood indicating a possibility of being a character string suitable for classifying image data of the type determined, the equation is chosen according to the type determined by the determining.
18. A computer-readable non-transitory medium on which is recorded a program that causes a computer to function as:
a specifying unit to specify a plurality of parts where a character string is described in image data;
a type determining unit to determine a type of the image data;
a layout information acquiring unit to acquire, for each of the plurality of parts specified by the specifying unit, layout information including the position of the part in the image data and the size of described characters; and
an estimating unit to calculate, for each of the plurality of parts, a likelihood as a character string having a predetermined attribute that can be included in the image data based on a positional relationship with other parts and on a size relationship with other parts, and estimating a part where a character string having the predetermined attribute is described based on the likelihood,
wherein the estimating unit estimates a part where a character string having the predetermined attribute is described, by using an equation for calculating the likelihood indicating a possibility of being a character string suitable for classifying image data of the type determined, the equation is chosen according to the type determined by the type determining unit.