WO2017067485A1

WO2017067485A1 - Picture management method and device, and terminal

Info

Publication number: WO2017067485A1
Application number: PCT/CN2016/102762
Authority: WO
Inventors: 贾霞
Original assignee: 中兴通讯股份有限公司
Priority date: 2015-10-22
Filing date: 2016-10-20
Publication date: 2017-04-27
Also published as: CN106610983A

Abstract

Provided are a picture management method and device, and a terminal. The picture management device (4) creates picture classification indexes, wherein the picture classification indexes include at least two of an attribute index, a content index and an event index; then the value of each picture classification index of each picture is obtained and stored; then the picture management device (4) receives a user query instruction, wherein the query instruction contains a classification query formula, and the classification query formula is composed of at least one target classification index; and then the picture management device (4) selects a qualified picture according to the classification query formula. The classification query formula at least comprises one target classification index; a picture needed by a user is more accurately located by adding a query condition; and the problems in the prior art that pictures only can be singly classified and queried, and classification and query are not fine and accurate enough are solved, and technical effects that the query or the classification on pictures is carried out according to requirements of the user so that the pictures can be more finely and accurately classified or queried are realized.

Description

Picture management method, device and terminal

Technical field

The present invention relates to the field of communications and information, and in particular, to a picture management method and apparatus, and a terminal.

Background technique

The development of digital multimedia and networks has enriched the entertainment experience in people's daily lives. People can take various pictures anytime and anywhere through digital cameras, mobile phones, tablets and other devices, and can easily obtain image resources generated by others through the network. After a period of time, each person's personal computer and handheld devices often accumulate a large number of pictures taken by themselves, or favorite pictures, ranging from a few hundred to more than a thousand. Therefore, in order to manage a large number of pictures, users have more and more demands on picture operations. For example, classifying all pictures can realize different classification according to different needs, thereby visually displaying pictures and facilitating management of many pictures; To more accurately retrieve images based on your needs, quickly locate the images you want to view.

In the prior art, there are various image management tools for handheld devices such as computers and mobile phones, but these image management tools have relatively simple functions for managing pictures. For example, only the classification by time is simple, and the retrieval method is simple, which also leads to The search is not accurate enough, and the target image cannot be located more accurately after inputting the keyword.

Summary of the invention

The main technical problem to be solved by the embodiments of the present invention is to provide a picture management method, a device, and a terminal, which can solve the problem that only a single classification and query can be performed on a picture in the prior art, resulting in a single classification, classification, and query. The results are not accurate enough, the user experience is not good, and the satisfaction is low.

To solve the above technical problem, an embodiment of the present invention provides a picture management method, including:

Setting a picture classification index, where the picture classification index includes at least two of an attribute index, a content index, and an event index;

Obtaining an index value of each picture classification index of each picture for storage;

Receiving a query instruction, the query instruction comprising a classification query formula consisting of at least one target image classification index;

Extracting a target index value of each target image classification index in the classification query formula, and selecting, from each of the pictures, all the target image classification indexes in the classification query formula and the index values and the target of the classification query formula An image with the same index value.

Optionally, when the picture classification index includes an attribute index, the attribute index includes at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index.

Optionally, when the picture classification index includes a content index, the content index includes at least one of a person index, a plant index, and an animal index.

Optionally, before acquiring the index value of the content index of the image, the method further includes: identifying the content in the image, and adding an index value of the corresponding content index to the image according to the recognition result.

Optionally, when the recognition result is a person, adding a corresponding person index value to the picture includes: identifying each face information on the picture, and comparing the face information with the existing face information. Determining whether the face information already exists and having a character index value, and if so, directly using the existing person index value as a person index value of the picture; if not, newly adding a person index corresponding to the face value.

Optionally, before the image classification index includes the event index, before acquiring the index value of the event index of each picture, the method further includes:

Defining at least one event index value;

Add the appropriate event index value for each image.

The embodiment of the invention further provides a picture management device, including:

a creating unit: configured to set a picture classification index, where the picture classification index includes at least two of an attribute index, a content index, and an event index;

An index value obtaining unit: configured to acquire an index value of each picture classification index of each picture;

An instruction receiving unit: configured to receive a query instruction, where the query instruction includes a classified query formula composed of at least one target image classification index;

a classification query unit: for extracting target index values of each target image classification index in the classification query formula, selecting, from the respective pictures, all target image classification indexes in the classification query formula, and index values and the A picture with the same target index value in the classification query.

Optionally, the index value obtaining unit includes an attribute index value obtaining sub-unit, configured to obtain an index value of each attribute index of each picture when the picture classification index includes an attribute index, where the attribute index includes an image format index, At least one of a picture size index, a picture acquisition time index, and a picture acquisition location index.

Optionally, the index value obtaining unit further includes a content index value obtaining subunit, configured to acquire an index value of each content index of each picture when the picture classification index includes a content index, where the content index includes a person index At least one of a plant index and an animal index.

Optionally, the content index value obtaining subunit includes: a content identification module and a content index value determining module, wherein the content identification module is configured to identify content in a picture; and the content index value determining module is configured to identify The result is an index value of the corresponding content index for the picture.

Optionally, the index value obtaining unit further includes an event index value obtaining subunit, configured to acquire an index value of an event index of each picture; and the event index value obtaining subunit includes an event index value determining module, configured to: Add the appropriate event index value for each image.

An embodiment of the present invention further provides a terminal, where the terminal includes the picture management apparatus.

According to still another embodiment of the present invention, a storage medium is also provided. The storage medium is arranged to store program code for performing the following steps:

Obtaining an index value of each picture classification index of each picture for storing; receiving a query instruction, where the query instruction includes a classification query formula consisting of at least one target picture classification index; extracting each target picture classification index in the classification query formula Target index value from the respective images A picture having all the target picture classification indexes in the classification query formula and having the same index value as each target index value of the classification query formula is selected.

Optionally, the storage medium is further arranged to store program code for performing the following steps:

Before obtaining the index value of the content index of the image, the method further includes: identifying the content in the image, and adding an index value of the corresponding content index to the image according to the recognition result.

When the picture classification index includes the event index, before acquiring the index value of the event index of each picture, at least one event index value is defined; and a corresponding event index value is added for each picture.

An embodiment of the present invention provides a picture management method and apparatus, and a terminal. The picture management apparatus sets a classification index, where the classification index includes at least two of an attribute index, a content index, and an event index. Then, the index values of the classification indexes of each picture are obtained and stored, and each picture has an index value corresponding to the classification index; after receiving the query instruction of the user, the query instruction includes a classification query formula, and the classification The query formula is composed of at least one target classification index, and the picture management device extracts a target index value of all target classification indexes in the classification query formula, and the target classification index is compared with the classification index of each picture, and the target index value and each picture are compared. The index values are compared, and a picture having all the target classification indexes in the classification query and having the same index value as the target index value is selected. It can be seen that, by using the solution of the present invention, the user can set the content of the classification index according to his own needs, and input the classification query formula according to the requirement, and the image management apparatus obtains the image that satisfies the query formula according to the issued query instruction, because the query formula includes at least one target classification. The index may be one or more of the queries or conditions, so that the present invention can more accurately locate the pictures required by the user, and can query or classify the pictures according to the user requirements, thereby classifying or querying the pictures more finely and accurately. Technical effects.

DRAWINGS

The drawings described herein are provided to provide a further understanding of the invention and constitute a In part, the illustrative embodiments of the invention and the description thereof are used to explain the invention and are not intended to limit the invention. In the drawing:

1 is a flowchart of a picture management method according to Embodiment 1 of the present invention;

2 is a flowchart of acquiring a character index value by a picture management method according to Embodiment 1 of the present invention;

3 is a flowchart of acquiring an event index value by a picture management method according to Embodiment 1 of the present invention;

4 is a schematic structural diagram of a picture management apparatus according to Embodiment 2 of the present invention;

5 is a schematic structural diagram of an index value acquiring unit in a picture management apparatus according to Embodiment 2 of the present invention;

FIG. 6 is a schematic structural diagram of a content index value obtaining subunit in a picture management apparatus according to Embodiment 2 of the present invention.

detailed description

The present invention will be further described in detail below with reference to the accompanying drawings.

Embodiment 1:

This embodiment provides a picture management method, which has the advantages of querying or classifying pictures according to user requirements, thereby classifying or querying pictures more finely and accurately, as shown in FIG. 1 , including:

S101: Create a picture classification index, where the created picture classification index includes at least two of an attribute index, a content index, and an event index.

S102: Acquire an index value of each picture classification index of each picture for storage.

S103: Receive a query instruction.

S104: Select a picture that satisfies the condition according to the query instruction.

The image classification index created in S101 in this embodiment may be flexibly created according to user requirements or a specific application scenario, and may include any two of an attribute index, a content index, and an event index in order to improve the flexibility and accuracy of the classification query. Yes, the theoretically included image classification index is more comprehensive, so the preferred image classification index contains attribute index, content index, and things. Index.

The attribute index of the picture in this embodiment is used to represent some attribute information of the picture, including at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index. The index value of the picture format index is used to represent the specific format of the picture, for example, other picture formats such as JEPG format, GIF format, and PNG format. The index value of the image size index is used to indicate the size of the image file, that is, the size of the storage space. The index value of the picture acquisition time index is used to indicate the time when the picture is acquired. For example, when the picture is taken by the camera on the terminal, the picture acquisition time of the picture is the time of capturing the picture, and when the picture is downloaded from the network, The picture acquisition time of the picture is the time when the picture was downloaded. The index value of the picture acquisition location index is used to indicate the location where the picture is created. For example, if a picture is taken in Shenzhen, the index of the location index of the picture is Shenzhen.

The content index is used to represent a person or thing that the image content is objectively presented, including at least one of a character index, a plant index, and an animal index. The index value for the character index is used to represent the person included in the picture content. When the content index includes the person index, one picture may correspond to one or more person index values. The index value of the plant index is used to represent the plants contained in the image content. There are many ways to define the index value. For example, different markers may be used to indicate the presence or absence of a plant. For example, 01 indicates that there is a plant, and 00 indicates There are no plants; you can also define the specific types in the picture. The animal index is used to represent the animals contained in the image content, similar to the way the plant index is defined. Different markers can also be used to indicate whether the image exists in the animal, or the specific animal in the image can be defined.

The index value of the event index is used to describe the meaning expressed by the content of the picture, or the event recorded by the picture, or the occasion of obtaining the picture. For example, if a picture records a scene on a wedding, the index value of the event index can be defined as a wedding. It is worth noting that a picture can correspond to an event index value or multiple event index values.

For the step S102 of the embodiment, the index values of the image classification indexes of the respective pictures are obtained, and the index values of the respective classification indexes for acquiring the respective pictures are specifically described in the following.

After creating the image classification index, obtain the image classification index value of each image. When the image classification index includes an attribute index, the attribute index includes at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index, and each attribute index value of each picture needs to be obtained, for the picture format and The size is the basic information of the picture, and the format of the picture is directly extracted as the picture format index value, and the value of the size of the picture is extracted as the picture size index value. Obtain an event index value for the picture of the picture. When the picture is a local picture, the terminal obtains the creation time of the picture file by reading the creation time attribute of the picture file, and obtains the event index value as a picture. When the picture is a network download, Then, the terminal reads the time when the picture is downloaded as the picture acquisition event index value of the picture. For the specific way of obtaining the position index value of the picture, many terminal devices record the latitude and longitude coordinate information when taking the picture. The program obtains the latitude and longitude coordinates by reading the EXIF (Exchangeable Image File) attribute information of the picture, and the terminal obtains the latitude and longitude coordinates according to the latitude and longitude coordinates. The specific location information of the picture is captured, and the location information is used as the picture of the picture to obtain the location index value; for the picture that cannot obtain the location information, the picture acquisition location index value is defined as null.

When the picture classification index includes a content index, the content index includes at least one of a person index, a plant index, and an animal index. To get the character index value process, see Figure 2:

Step S201: Acquire a picture, and the obtained picture that can be taken in real time by the user or downloaded in real time may also exist before.

Step S202: The terminal detects all face information on the picture by using a face detection algorithm.

Step S203: Comparing the obtained face information with the existing face information, determining whether the face information already exists and having a corresponding person index value, and if so, executing S204, if not, executing S205;

Step S204: directly reading the character index value corresponding to the existing face information as the person index value of the picture;

Step S205: A new person index value is added.

In step S203, if the comparison result is non-existent, or although the face information exists in the comparison result but there is no person index value corresponding to the face information, perform S304, newly add and For the character index value corresponding to the face, it should be understood that the newly added character index value corresponding to the face may be a predefined person index value added, or the user may manually edit and add, for the predefined The character index value can be customized by the device or pre-defined by the user. Of course, the user can also choose not to add it and add it manually afterwards. The form of the added person index value may be the name of the person corresponding to the face, or any other form convenient for the user to memorize and operate. For the plant index value and the animal index value, it may be a pre-customized plant index value and an animal index value, or may be newly added by the user, when obtaining a picture, selecting a predefined plant index value and an animal index value, or according to the user The naming convention is to add a new plant index value and an animal index value to the image. The pre-defined plant index value and animal index value may be factory-customized by the device, or may be predefined by the user. For example, when the device is shipped from the factory, the plant and plant-free plant index values may be predefined. A picture detects that there is a plant on the picture, and the plant index value of the picture can be directly defined as having a plant, and the user can also modify the index value of the plant, which is similar for the animal index. The user can choose to define the plant index value and animal index value of the image when the image is acquired. It can also be added afterwards. The form of the plant index value and the animal index value can be added in any form according to the user's habit.

For the process of getting the index value of the event index, see Figure 3:

S301: Acquire a picture, obtain a picture that can be taken in real time by the user, or download it in real time, or it may exist before.

S302: Select an index value of the event index for the picture.

In step S302 of this embodiment, the index value of the selected event index may be a predefined event index value or a newly created event index value, and the predefined event index value may be an event index defined at the time of leaving the device at the factory. The value can also be predefined by the user. The specific form of the event index value may be any form that the user is accustomed to, such as numbers, symbols, Chinese characters, English, and the like.

After obtaining each picture classification index and index value of each picture, each picture, the classification index, and the index value of each classification index and the correspondence relationship between each picture and each index value are stored.

The query instruction received in S103 in this embodiment includes classification by at least one target picture A classification query formula composed of indexes; the classification query formula includes an image classification index that each user wants to query, for example, the user wants to search or classify the image according to the size of the image and the characters on the image, and the query instruction issued by the user The classification query formula included in the attribute index is the image size index in the attribute index and the person index in the content index. The classification query formula may include a target image classification index or multiple, and of course, the number of target image classification indexes included The more detailed the multi-category, the higher the accuracy of the search.

After obtaining the classification query formula in the query instruction, step S104 is performed: selecting a picture that satisfies the condition according to the query instruction. Extracting a target index value of each target image classification index in the classification query formula, performing an index value of the target image classification index and the target image classification index of the query formula, and an image classification index of each image and an index value of the image classification index. For example, if the picture classification index of the picture includes all the target classification indexes, and the index value of each picture classification index of the picture matches the index value of the target picture classification index, the picture is a picture that satisfies the query condition.

The entire flow of the above picture management method will be described below with reference to specific examples:

The created image classification index includes an attribute index, a content index, and an event index. The attribute index includes a time index and a position index; the content index includes a person index, and the index values of the respective indexes are as shown in Table 1 below:

Table 1

Obtaining each picture in the above table and each picture classification index is an index value and each picture The index value of the slice classification index will be stored, and when the user's query instruction is received, the classification query formula in the user query instruction is extracted. For example, the target image classification index in the classification query formula input by the user includes the attribute index and the content. The time index and the person index in the index, the index value of the time index is 2015, and the index value of the person index is Zhang San; the terminal selects all photos with the time index value of 2015 and the character index value of Zhang San, according to the above table. Comparing, it can be seen that picture 1 and picture 4 satisfy the query condition, and the terminal feeds back the query result to the user.

Embodiment 2:

The present embodiment provides a picture management apparatus. Referring to FIG. 4, the picture management apparatus 4 includes a creation unit 41, an index value acquisition unit 42, an instruction reception unit 43, and a classification inquiry unit 44. The creating unit 41 is configured to set a picture classification index; the index value obtaining unit 42 is configured to obtain an index value of each picture classification index of each picture; the instruction receiving unit 43 is configured to receive a query instruction; and the classification query unit 44 is configured to select according to the query instruction. A picture that satisfies the condition.

The creating unit 41 is configured to set a picture classification index, where the picture classification index includes at least two of an attribute index, a content index, and an event index. The index value of the attribute index is used to represent some attribute information of the picture, including at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index. The index value of the picture format index is used to represent the format of the picture. The index value of the picture size index is used to indicate the size of the picture file. The index value of the picture acquisition time index is used to indicate the time for obtaining the picture, and the index value of the picture acquisition position index is used. Indicates where the image was created. The index value of the content index is used to represent a person or thing whose image content is objectively presented, and the content index includes at least one of a person index, a plant index, and an animal index. The index value of the character index is used to represent the character contained in the image content. When the content index includes the character index, one picture may correspond to one or more character index values. The index value of the plant index is used to represent the plants contained in the image content. There are various ways to define the index value. For example, different markers may be used to indicate the presence or absence of a plant, and specific types in the image may be defined. . The index value of the animal index is used to represent the animal contained in the image content, similar to the way the plant index is defined. Different markers can also be used to indicate whether the image exists in the animal, or the specific animal in the image can be defined. The index value of the event index is used to describe the meaning expressed by the content of the picture, or the event recorded by the picture, or the occasion of obtaining the picture.

The instruction receiving unit 43 is configured to receive a query instruction, where the query instruction includes a classification query formula composed of at least one target picture classification index. The classification query includes an image classification index that each user wants to query. For example, if the user wants to search or classify the image according to the size of the image and the characters on the image, the classification query included in the query instruction issued by the user is The image size index in the attribute index and the person index in the content index. The classification query can include a target image classification index or multiple. Of course, the more the target image classification index is included, the finer the classification will be. The accuracy of the search is also higher.

The classification query unit 44 is configured to: after obtaining the classification query formula in the query instruction, extract a target index value of each target image classification index in the classification query formula, and select an index of the target image classification index of the query formula and the target image classification index. The value and the picture classification index of each picture are compared with the index value of the picture classification index, if the picture classification index of the picture includes all target classification indexes, and the index value of each picture classification index of the picture and the target picture classification index If the index value matches, the picture is a picture that satisfies the query condition.

Referring to FIG. 5, an index value obtaining unit 5 is configured to obtain an index value of the various types of picture classification indexes of each picture, including an attribute index value obtaining subunit 51, a content index value obtaining subunit 52, and an event index value obtaining sub Unit 53. The attribute index value obtaining sub-unit 51 is configured to obtain an index value of each attribute index of each picture when the picture classification index includes an attribute index, and the content index value obtaining sub-unit 52 is configured to acquire each picture when the picture classification index includes a content index. The index value of each content index is used; the event index value obtaining sub-unit 53 is configured to acquire an index value of an event index of each picture.

The attribute index includes at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index. For the basic information of the picture format and the size of the picture, the attribute index value obtaining subunit 51 directly extracts the format of the picture as The image format index value extracts the size value of the image as the image size index value. The specific obtaining manner of the image obtaining time index value of the picture is: when the picture is a local shooting picture, the attribute index value obtaining sub-unit 51 obtains the creation time of the picture file by reading the creation time attribute of the picture file, and uses it as a picture. Obtaining a time index value. When the picture is a network download, the attribute index value obtaining sub-unit 51 reads an event of downloading the picture as a picture acquisition event index value of the picture. For image acquisition The position index value, when many terminal devices take a picture, the latitude and longitude coordinate information is recorded, and the attribute index value acquisition sub-unit 51 obtains the latitude and longitude coordinates by reading the EXIF (Exchangeable Image File) attribute information of the picture, and the terminal obtains the photograph according to the latitude and longitude coordinates. The specific location information of the picture is obtained by using the location information as a picture of the picture; and the picture acquisition location index value is defined as null for the picture in which the location information cannot be obtained.

When a new picture or a new picture is added, the event index value acquisition sub-unit 53 completes the correspondence between the picture and the event index value by selecting a predefined event index value or a new time index value, and the predefined event index value may be When the device is shipped from the factory, the event index value defined by the sub-unit 53 is obtained by the event index value, or may be defined by the user through the event index value obtaining sub-unit 53 in advance. The specific form of the event index value may be any form that the user is accustomed to, such as numbers, symbols, Chinese characters, English, and the like. The user can choose to add the event index value when the image is obtained, or choose not to add when acquiring, and then add it later.

The content index includes at least one of a person index, a plant index, and an animal index. Referring to FIG. 6, the content index value acquisition sub-unit 6 includes a content recognition module 61 and a content index value determination module 62. The content identification module 61 is configured to identify the content in the picture, and the recognition result is a person, an animal or a plant. After the content identification module 61 recognizes the content of the picture, the content index value determining module 62 determines the corresponding index value according to the recognition result of the content. .

When the content identification module 61 recognizes that the content of the picture is a person, the content index value determining module 62 automatically detects all the face information on the picture by using the face detection algorithm, and the detected face information and the existing person. Whether the face information is compared before viewing, if the comparison result is existed, and then whether the existing face information has a corresponding human index value, and if so, directly reading the character index value as the image a person index value; if the comparison result is non-existent, or although the face information exists in the comparison result but there is no person index value corresponding to the face information, the person index value corresponding to the face is newly added, It should be understood that the newly added character index value corresponding to the face may be a pre-defined person index value added, or may be manually edited and added by the user. For a predefined character index value, the device may be a factory-time content index. The value determining module 62 has been defined, or may be pre-defined by the user in advance through the content index value determining module 62, of course, You can also choose not to add, and then afterwards add manully. The form of the added person index value may be the name of the person corresponding to the face, or any other form convenient for the user to memorize and operate. When the content identification module 61 recognizes that the content of the picture is a plant or an animal, the plant index value and the animal index value may be predefined by the content index value determining module 62, or may be newly added by the user. Selecting a predefined plant index value and an animal index value or setting the user in advance so that the content index value determining module 62 content index value determining module 62 automatically selects a predefined plant index value and an animal index value, or according to the user's naming convention The image adds a new plant index value and an animal index value. The pre-defined plant index value and the animal index value may be defined in the content index value determining module 62 when the device is shipped from the factory, or may be defined by the user in advance by the content index value determining module 62, for example, when the device is shipped from the factory. Two plant index values can be defined in advance and no plant. When a picture is obtained to detect that there is a plant on the picture, the plant index value of the picture can be directly defined as having a plant, and the user can also modify the index value of the plant. For animal indexing similar. The user can choose to define the plant index value and animal index value of the image when the image is acquired. It can also be added afterwards. The form of the plant index value and the animal index value can be added in any form according to the user's habit.

Optionally, the picture management apparatus further includes a network unit, configured to upload the local picture, each picture classification index, the index value of each picture classification index, and other picture related information to the network server.

Optionally, the device further includes a storage unit, configured to store the image and the image classification index corresponding to the image, and the index value of each image classification index, and store the storage unit on the device. The apparatus further includes an obtaining unit configured to acquire a picture and each picture classification index corresponding to the picture from the other storage device, and an index value of each picture classification index.

Embodiment 3:

The present embodiment provides a terminal, including the picture management device in this embodiment, for example, a computer, a server, a mobile electronic device, and the like, which are common to the picture management device in the second embodiment.

Specifically, the following will be described with reference to specific examples, for example, one includes the second embodiment. a mobile phone of the picture management device, the mobile phone includes a creation unit, an index value acquisition unit, an instruction receiving unit, and a classification query unit, the creation unit sets an image classification index, and the index value acquisition unit acquires an index value of each picture classification index of each picture; The receiving unit is configured to receive the query instruction, and the classification query unit selects the image that satisfies the condition according to the query instruction, and further includes the storage unit storing the related image information such as the image, the image classification index, and the index value. In order to reduce the workload of the mobile phone, only the creation unit, the instruction receiving unit, the classification query unit and the network unit may be set on the mobile phone, and the index value acquisition unit and the instruction receiving unit are set on the server, and the creation unit sets the picture classification. Index, when the instruction receiving unit of the mobile phone receives the classification query command issued by the user, the network unit wants the server to send the request, and the server receives the request to call up the stored related information for classification and query processing, saves the relevant information of the processing result, and the result Return to the phone for display to the user.

It can be seen from the above embodiments that the implementation of the present invention has at least the following beneficial effects:

The present invention provides a picture management method and apparatus, and a terminal, which first obtains a picture classification index of a picture, including at least two of an attribute index, a content index, and an event index, before acquiring and querying the picture, and acquiring each The value of the corresponding picture classification index of the picture is stored; after receiving the query command of the user, the picture matching the query condition is selected. It is worth noting that the query command input by the user includes a classification search formula and the classification search formula is composed of one or more target image classification indexes, that is, the user can input values of multiple target classification indexes according to needs. The invention obtains each category search value of each picture before classifying and retrieving the picture, so that the picture is searched by each category search value, and the picture required by the user can be more accurately located, and only a single picture can be solved in the prior art. Classification and query, classification and query are not detailed, accurate questions, to achieve the technical effect of more detailed and accurate classification or query of pictures according to user needs to query or classify pictures.

The above is a further detailed description of the present invention in connection with the specific embodiments, and the specific embodiments of the present invention are not limited to the description. It will be apparent to those skilled in the art that the present invention may be made without departing from the spirit and scope of the invention.

Industrial applicability

In the picture management process of the embodiment of the present invention, the classification index is set by the picture management apparatus, and the classification index includes at least two of an attribute index, a content index, and an event index; and then an index value of each category index of each picture is obtained and stored, and implemented. Each picture has an index value corresponding to the classification index; after receiving the query instruction of the user, the query instruction includes a classification query formula, the classification query formula is composed of at least one target classification index, and the picture management device extracts the classification The target index value of all target classification indexes in the query, the target classification index is compared with the classification index of each picture, and the target index value is compared with the index value of each picture, and the selection has the query query type All images with the same target index and the same index value as the target index value. It can be seen that, by using the solution of the present invention, the user can set the content of the classification index according to his own needs, and input the classification query formula according to the requirement, and the image management apparatus obtains the image that satisfies the query formula according to the issued query instruction, because the query formula includes at least one target classification. The index may be one or more of the queries or conditions, so that the present invention can more accurately locate the pictures required by the user, and can query or classify the pictures according to the user requirements, thereby classifying or querying the pictures more finely and accurately. Technical effects.

Claims

A picture management method, including:

Creating a picture classification index, where the picture classification index includes at least two of an attribute index, a content index, and an event index;

Obtaining an index value of each picture classification index of each picture for storage;

Receiving a query instruction, the query instruction comprising a classification query formula consisting of at least one target image classification index;

Extracting a target index value of each target image classification index in the classification query formula, and selecting, from each of the pictures, all the target image classification indexes in the classification query formula and the index values and the target of the classification query formula The image whose index value matches.
The picture management method according to claim 1, wherein when the picture classification index includes an attribute index, the attribute index includes at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition position index. .
The picture management method according to claim 1, wherein when the picture classification index includes a content index, the content index includes at least one of a person index, a plant index, and an animal index.
The picture management method according to claim 3, wherein before the index value of the content index of the picture is acquired, the method further comprises: identifying the content in the picture, and adding an index value of the corresponding content index to the picture according to the recognition result.
The picture management method according to claim 4, wherein when the recognition result is a person, adding a corresponding person index value to the picture comprises: identifying each face information on the picture, and using the face information and The existing face information is compared to determine whether the face information already exists and has a person index value, and if so, the existing person index value is directly used as the person index value of the picture; if not, the new addition and The character index value corresponding to the face.
The picture management method according to any one of claims 1 to 5, wherein, before the image classification index includes the event index, before acquiring the index value of the event index of each picture, the method further includes:

Defining at least one event index value;

Add the appropriate event index value for each image.
A picture management device comprising:

a creating unit: configured to set a picture classification index, where the picture classification index includes at least two of an attribute index, a content index, and an event index;

An index value obtaining unit: configured to acquire an index value of each picture classification index of each picture;

An instruction receiving unit: configured to receive a query instruction, where the query instruction includes a classified query formula composed of at least one target image classification index;

a classification query unit: for extracting target index values of each target image classification index in the classification query formula, selecting, from the respective pictures, all target image classification indexes in the classification query formula, and index values and the A picture with the same target index value in the classification query.
The picture management apparatus according to claim 7, wherein the index value acquisition unit includes an attribute index value acquisition subunit, and configured to acquire an index value of each attribute index of each picture when the picture classification index includes an attribute index, The attribute index includes at least one of a picture format index, a picture size index, a picture acquisition time index, and a picture acquisition location index.
The picture management apparatus according to claim 7, wherein the index value obtaining unit further comprises a content index value obtaining subunit, configured to acquire an index of each content index of each picture when the picture classification index includes a content index And a value index, the content index including at least one of a person index, a plant index, and an animal index.
The picture management device according to claim 9, wherein the content index value acquisition subunit comprises: a content recognition module and a content index value determination module, the content recognition module is configured to identify content in the picture; The content index value determining module is configured to add an index value of the corresponding content index to the picture according to the recognition result.
The picture management apparatus according to any one of claims 7 to 10, wherein the index value acquisition unit further includes an event index value acquisition subunit, configured to acquire an index value of an event index of each picture; the event index value. The obtaining subunit includes an event index value determining module for adding a corresponding event index value for each picture.
A terminal comprising the picture management device according to any one of claims 7-11.