CN105243060B - A kind of method and device of retrieving image - Google Patents
A kind of method and device of retrieving image Download PDFInfo
- Publication number
- CN105243060B CN105243060B CN201410239452.5A CN201410239452A CN105243060B CN 105243060 B CN105243060 B CN 105243060B CN 201410239452 A CN201410239452 A CN 201410239452A CN 105243060 B CN105243060 B CN 105243060B
- Authority
- CN
- China
- Prior art keywords
- picture
- pictures
- retrieved
- matching degree
- terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 230000001815 facial effect Effects 0.000 claims 2
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 14
- 239000013598 vector Substances 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present disclosure discloses a kind of method and devices of retrieving image, belong to field of computer technology.The described method includes: obtaining the characteristic information that terminal is sent, the characteristic information includes at least one of characteristics of image, face characteristic or label characteristics；Retrieval and the characteristic information picture that match degree is greater than the preset threshold in picture library, and the picture retrieved is determined as search result；The picture for meeting preset condition is being chosen as query expansion in cluster where the picture retrieved as a result, cluster where the picture retrieved is the corresponding cluster of type of the characteristic information；The search result and the query expansion result are returned into the terminal.The disclosure passes through using the picture retrieved as search result, and it chooses in the corresponding cluster of picture retrieved and meets other pictures of preset condition as query expansion result, can retrieve be more in line with user requirement as a result, improving the accuracy and efficiency of retrieval.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for retrieving pictures.
Background
With the development of the internet, the cloud space storage has been popularized in the life and work of users, and users can store various pictures in mobile phones and personal computers in the cloud space, retrieve and acquire their own pictures from the cloud space through the network at any time and any place, and are not limited by regions.
When a user uses a cloud space, the situation of searching for a picture inevitably exists, and the picture searching mode under a general condition is as follows: a user can search the picture with the keyword label by inputting the keyword, and the picture completely matched with the keyword is taken as a retrieval result; or the user selects the picture to be detected, searches whether the picture with the similarity larger than a certain threshold exists or not by using the face feature or the image feature, and takes the searched picture as a search result.
The inventors found that the related art has at least the following problems:
for the way of the keyword, only pictures that completely match the keyword can be retrieved; for the way of identifying the face features or the image features, only pictures with similarity greater than a certain threshold can be retrieved. The two search modes are limited too much, and the searched pictures can not completely show the search requirements of the user.
Disclosure of Invention
In order to solve the problems of the related art, the embodiments of the present disclosure provide a method and an apparatus for retrieving a picture. The technical scheme is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided a method for retrieving a picture, including:
acquiring feature information sent by a terminal, wherein the feature information comprises at least one of image features, face features or label features;
searching pictures with matching degree greater than a preset threshold value with the characteristic information in a picture library, and determining the searched pictures as a searching result;
selecting pictures meeting preset conditions from the clusters where the retrieved pictures are located as extended retrieval results, wherein the clusters where the retrieved pictures are located are clusters corresponding to the types of the characteristic information;
and returning the retrieval result and the extended retrieval result to the terminal.
According to a second aspect of the embodiments of the present disclosure, there is provided an apparatus for retrieving a picture, including:
the first acquisition module is used for acquiring feature information sent by a terminal, wherein the feature information comprises at least one of image features, human face features or label features;
the retrieval module is used for retrieving pictures with the matching degree with the characteristic information being greater than a preset threshold value in a picture library and determining the retrieved pictures as retrieval results;
the first selection module is used for selecting pictures meeting preset conditions from the clusters of the retrieved pictures as extended retrieval results, and the clusters of the retrieved pictures are clusters corresponding to the types of the characteristic information;
and the return module is used for returning the retrieval result and the extended retrieval result to the terminal.
The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects: by taking the retrieved picture as a retrieval result and selecting other pictures meeting the preset condition in the cluster corresponding to the retrieved picture as extended retrieval results, the result more meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a flow chart illustrating a method of retrieving pictures according to an exemplary embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating a method of retrieving pictures according to an exemplary embodiment of the present disclosure;
FIG. 3 is a flow chart illustrating a method of retrieving pictures according to an exemplary embodiment of the present disclosure;
FIG. 4 is a flow chart illustrating a method of retrieving pictures according to an exemplary embodiment of the present disclosure;
FIG. 5 is a schematic diagram illustrating an apparatus for retrieving pictures according to an exemplary embodiment of the present disclosure;
fig. 6 is a schematic structural diagram illustrating an apparatus for retrieving pictures according to an exemplary embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the present disclosure more apparent, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.
Example one
The embodiment of the present disclosure provides a method for retrieving pictures, referring to fig. 1, the method flow includes:
in step 101, acquiring feature information sent by a terminal, where the feature information includes at least one of an image feature, a face feature, or a tag feature, and the method of this embodiment is used in a server;
in step 102, retrieving a picture whose matching degree with the feature information is greater than a preset threshold value from a picture library, and determining the retrieved picture as a retrieval result, where the method of the embodiment is used in a server;
in step 103, the picture meeting the preset condition is selected from the clusters where the retrieved pictures are located as the extended retrieval result, and the cluster where the retrieved pictures are located is the cluster corresponding to the type of the feature information.
In step 104, the search result and the expanded search result are returned to the terminal, and the method of the embodiment is used in the server.
Before the process of retrieving pictures is started in step 101, the pictures in the same data source need to be clustered in advance, and the pictures belonging to the same category need to be clustered in one cluster. The clustering method in the embodiment of the present disclosure may include, but is not limited to, the following ways: clustering based on image features, clustering based on face features, and clustering based on labels.
The clustering based on the image features is characterized in that feature vectors of key points in pictures are calculated and compared, the pictures with matching degrees larger than a certain threshold value are determined as a cluster, and the matching degrees of the pictures are stored in the same cluster. In the embodiment of the disclosure, the key points of the picture are extracted by using a Scale-invariant feature transform (SIFT) algorithm, the algorithm is an algorithm for detecting local features, each key point in the picture is calculated by the SIFT algorithm to obtain a 128-dimensional SIFT feature vector, and finally the feature vectors of all the key points constitute the image features. And during final alignment, the Euclidean distance between the feature vectors of the key points is used as the similarity judgment measurement of the key points in the two images. Alternative metrics may also include cosine similarity metric, which is not limited herein.
The clustering based on the human face features is characterized in that the matching degree of all pictures is stored in the clustering determined to be the same human face by extracting the features of the human face in the pictures and regarding the pictures containing the feature information of the same human face as a cluster. And finally, the characteristic structure of the face forms the face characteristic of the picture.
Label-based clustering can be divided into two categories:
the first one is: and clustering according to the time information and/or the position information labels included in the pictures, and determining the pictures belonging to a certain time range and a certain position range as a cluster. The time range and the location range are not limited. In the first case, the time information and the position information are obtained by extracting exif information in the picture. Finally, the label characteristics of the picture are obtained through the time information and/or the position information.
The second method is as follows: and clustering according to the character information in the label characteristics marked by the user and included in the picture. In the second case, the text information may include, but is not limited to: the names of people, scenery, place names and the like are obtained by calibrating in the picture by the user. And finally, obtaining the character information of the picture through the label characteristics calibrated by the user.
Furthermore, similarity exists between the pictures in each cluster, wherein for the clusters of image characteristics and face characteristics, the similarity between the pictures can be carried out through the image characteristics or the face characteristics; for the clustering of the label features, the similarity between the pictures can be calculated through the time span in the time information and the distance in the position information in the label features.
In the embodiment of the present disclosure, the server has the above-mentioned multiple clusters obtained based on different clustering manners, so that the user can find out a more accurate retrieval result by sending different feature information.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
Example two
The embodiment of the present disclosure provides a method for retrieving pictures, referring to fig. 2, the method flow includes:
in step 201, feature information sent by a terminal is obtained, the feature information includes an image feature, and the method of the embodiment is used in a server.
In the embodiment of the disclosure, a user selects a picture to be retrieved from a terminal, and determines to retrieve a picture similar to the picture to be retrieved. At the moment, SIFT feature vectors of all key points in the picture are calculated in the terminal through an SIFT algorithm, the image features are obtained, and the image features are sent to the server.
Optionally, after the user selects the picture to be retrieved at the terminal, the user can also directly send the picture to be retrieved to the server, and after the server receives the picture to be retrieved, the server performs a process of acquiring the feature information of the picture to be retrieved. The server obtains feature information of the picture to be retrieved, wherein the feature information comprises image features and/or face features.
In step 202, according to the identifier of the terminal, the pictures stored in the other picture libraries by the terminal and the feature information corresponding to the pictures stored in the other picture libraries by the terminal are obtained.
The identifier of the terminal may be content such as account information, and sends a request for obtaining a picture to other picture libraries corresponding to the terminal through the identifier.
The other picture libraries can be associated in advance, and the addresses of the other picture libraries are stored in advance in the case; or the terminal may carry the addresses of other picture libraries in the sent feature information or the retrieval request of the picture to be retrieved.
In step 203, the pictures stored in the other picture libraries by the terminal are clustered and stored in the picture library of the terminal according to the feature information corresponding to the pictures stored in the other picture libraries by the terminal.
The obtaining of the feature information corresponding to the pictures stored in the other picture libraries may include: at least one of image features, face features, or label features. The calculation process that can be performed on the image features and the face features is as in the first embodiment, and is not described herein again. For the label feature, the feature can be obtained in the attribute of the picture, such as: time information, location information, text information, etc.
Further, the process of clustering the pictures according to the feature information is the same as that in the first embodiment, and is not described herein again.
In step 204, when the feature information is an image feature, calculating a matching degree between the image feature and an image feature corresponding to each picture in the picture library, selecting a picture with the matching degree greater than a preset threshold as a retrieved picture, and determining the retrieved picture as a retrieval result.
And calculating the matching degree between the pictures by a selected Euclidean distance mode or a cosine similarity measurement mode. The degree of matching may be a value between 0 and 100. The higher the numerical value is, the more similar the two pictures are, and a preset threshold value is set as a judgment standard of the similar pictures.
In step 205, a picture meeting the preset condition is selected from the clusters where the retrieved pictures are located as the extended retrieval result, and the cluster where the retrieved pictures are located is the cluster corresponding to the type of the feature information.
And further selecting other pictures in the cluster corresponding to the image characteristics to which the retrieved picture belongs as an extended retrieval result, wherein the selection mode is to determine the selected picture as the extended retrieval result according to the fact that the similarity between the retrieved picture and the other pictures in the cluster is greater than the preset similarity.
For example: the matching degree of the picture E to be retrieved selected by the user and the retrieved picture A is 90, and the picture E and the picture A belong to the same cluster comprising B, C and D. Wherein, the matching degrees of B, C and D with A are respectively 90, 80 and 70. Then the matching degrees between E and B, C, D are: 90- (100-90) 80, 90- (100-80) 70, 90- (100-70) 60, wherein the preset similarity is 70, and then the picture with the similarity larger than 70 to the retrieved picture is selected as the extended retrieval result.
And generating an arrangement sequence according to the matching degrees of other pictures in the expansion result, wherein the sequence is displayed in the terminal display from high to low.
Therefore, step 205 may further be:
and selecting the picture with the similarity higher than the preset similarity with the retrieved picture as an extended retrieval result in the cluster of the retrieved pictures.
In step 206, the search result and the expanded search result are returned to the terminal, and the method of the embodiment is used in the server.
If the searched pictures exceed the display range of the current screen in the terminal display process, displaying the parts which can be displayed in the terminal screen according to the arrangement sequence in the extended search result, and sequentially displaying other pictures which are not displayed according to the sliding action of the user.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
EXAMPLE III
The embodiment of the present disclosure provides a method for retrieving pictures, referring to fig. 3, the method flow includes:
in step 301, feature information sent by a terminal is received, where the feature information includes a face feature, and the method of this embodiment is used in a server.
In the embodiment of the disclosure, a user selects a face in a picture to be retrieved from a terminal, and determines to retrieve a picture similar to the face in the picture to be retrieved. At this time, the face features in the picture to be retrieved are calculated in the terminal through a face recognition algorithm, and the face features are sent to the server.
Optionally, after the user selects the picture to be retrieved at the terminal, the user can also directly send the picture to be retrieved to the server, and after the server receives the picture to be retrieved, the server performs a process of acquiring the feature information of the picture to be retrieved. The server obtains feature information of the picture to be retrieved, wherein the feature information comprises image features and/or face features.
In step 302, according to the identifier of the terminal, the pictures stored in the other picture libraries by the terminal and the feature information corresponding to the pictures stored in the other picture libraries by the terminal are obtained.
The identifier of the terminal may be content such as account information, and sends a request for obtaining a picture to other picture libraries corresponding to the terminal through the identifier.
The other picture libraries can be associated in advance, and the addresses of the other picture libraries are stored in advance in the case; or the terminal may carry the addresses of other picture libraries in the sent feature information or the retrieval request of the picture to be retrieved.
In step 303, according to the feature information corresponding to the pictures stored in the other picture libraries by the terminal, the pictures stored in the other picture libraries by the terminal are clustered and stored in the picture library of the terminal.
The obtaining of the feature information corresponding to the pictures stored in the other picture libraries may include: at least one of image features, face features, or label features. The calculation process that can be performed on the image features and the face features is as in the first embodiment, and is not described herein again. For the label feature, the feature can be obtained in the attribute of the picture, such as: time information, location information, text information, etc.
Further, the process of clustering the pictures according to the feature information is the same as that in the first embodiment, and is not described herein again.
In step 304, when the feature information is a face feature, calculating a matching degree between the face feature and the face feature of each picture in the picture library, selecting a picture with the matching degree greater than a preset threshold as a retrieved picture, and determining the retrieved picture as a retrieval result.
The calculation process of the matching degree of the face features may be as follows:
and calculating the matching degree between each feature point in the human face features and each feature point in each picture in the picture library.
In step 305, a picture meeting a preset condition is selected from the clusters where the retrieved pictures are located as an extended retrieval result, and the clusters where the retrieved pictures are located are clusters corresponding to the types of the feature information.
And further selecting other pictures in the cluster corresponding to the face features to which the retrieved picture belongs as an extended retrieval result, wherein the selection mode is to determine the selected picture as the extended retrieval result according to the fact that the similarity between the retrieved picture and the other pictures in the cluster is greater than the preset similarity.
Thus, step 305 may further be:
and selecting the picture with the similarity higher than the preset similarity with the retrieved picture as an extended retrieval result in the cluster of the retrieved pictures.
In step 306, the search result and the expanded search result are returned to the terminal, and the method of the embodiment is used in the server.
If the searched pictures exceed the display range of the current screen in the terminal display process, displaying the parts which can be displayed in the terminal screen according to the arrangement sequence in the extended search result, and sequentially displaying other pictures which are not displayed according to the sliding action of the user.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
Example four
The embodiment of the present disclosure provides a method for retrieving pictures, referring to fig. 4, the method includes:
in step 401, feature information sent by a terminal is received, where the feature information includes a tag feature, and the method of this embodiment is used in a server.
In the embodiment of the disclosure, a user inputs a tag to be detected in a terminal, and determines to retrieve pictures similar to the tag. At this point, the tag characteristics are sent to the server at the terminal.
Optionally, after the user selects the picture to be retrieved at the terminal, the user can also directly send the picture to be retrieved to the server, and after the server receives the picture to be retrieved, the server performs a process of acquiring the feature information of the picture to be retrieved. The server obtains feature information of the picture to be retrieved, wherein the feature information comprises image features and/or face features.
In step 402, according to the identifier of the terminal, the pictures stored in the other picture libraries by the terminal and the feature information corresponding to the pictures stored in the other picture libraries by the terminal are obtained.
The identifier of the terminal may be content such as account information, and sends a request for obtaining a picture to other picture libraries corresponding to the terminal through the identifier.
The other picture libraries can be associated in advance, and the addresses of the other picture libraries are stored in advance in the case; or the terminal may carry the addresses of other picture libraries in the sent feature information or the retrieval request of the picture to be retrieved.
In step 403, according to the feature information corresponding to the pictures stored in the other picture libraries by the terminal, the pictures stored in the other picture libraries by the terminal are clustered and stored in the picture library of the terminal.
The obtaining of the feature information corresponding to the pictures stored in the other picture libraries may include: at least one of image features, face features, or label features. The calculation process that can be performed on the image features and the face features is as in the first embodiment, and is not described herein again. For the label feature, the feature can be obtained in the attribute of the picture, such as: time information, location information, text information, etc.
Further, the process of clustering the pictures according to the feature information is the same as that in the first embodiment, and is not described herein again.
In step 404, when the feature information is a tag feature, calculating a matching degree between the tag feature and the tag feature corresponding to each picture in the picture library, selecting a picture with the matching degree greater than a preset threshold as a retrieved picture, and determining the retrieved picture as a retrieval result.
And selecting a picture with the tag characteristic completely consistent with the received tag characteristic from the picture library as a retrieval result when the picture with the matching degree larger than a preset threshold value is selected for the time information, the position information and the text information in the tag characteristic.
For example: the label is characterized by 'position information', such as longitude and latitude, searching for pictures belonging to the longitude and latitude in the position information, and determining the searched pictures as the searched pictures;
or the label is characterized by time information '2010' and position information 'Beijing', and a picture which contains the Beijing in the position information and is shot at 2010 is selected from the picture library and determined as the retrieved picture;
or, the label is characterized by the character information 'Beijing', after semantic recognition, the 'Beijing' is determined as the position information, the longitude and latitude of the position information are obtained, the pictures belonging to the longitude and latitude in the position information are searched, and the searched pictures are determined as the searched pictures;
or the label is characterized by the character information 'zhang san', the label containing the name of the person is searched, and the searched picture is determined as the searched picture.
Accordingly, for step 404, embodiments of the present disclosure may further include:
calculating the matching degree of the time information in the tag characteristics and the time information in the tag characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as a retrieved picture; or,
and calculating the matching degree of the position information in the label characteristics and the position information in the label characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
In step 405, a picture meeting a preset condition is selected from the clusters where the retrieved pictures are located as an extended retrieval result, and the clusters where the retrieved pictures are located are clusters corresponding to the types of the feature information.
And further selecting other pictures in the cluster corresponding to the label characteristics of the retrieved picture as an extended retrieval result, wherein the selection mode is to determine the selected picture as the extended retrieval result according to the fact that the similarity between the retrieved picture and the other pictures in the cluster is greater than the preset similarity.
Thus, step 405 may be further:
and selecting the picture with the similarity higher than the preset similarity with the retrieved picture as an extended retrieval result in the cluster of the retrieved pictures.
In step 406, the retrieved picture belongs to the cluster corresponding to the image feature or other pictures in the cluster corresponding to the face feature are selected as the extended retrieval result.
For example: the label is characterized by position information of the 'Imperial palace', and after all pictures belonging to the 'Imperial palace' are found in the pictures marked by the label, pictures which are not marked by the label and contain 'Imperial palace' scenery may exist. At this time, the picture belonging to the scene of the 'home palace' is obtained by searching the cluster corresponding to the image feature to which the searched picture belongs, so as to obtain a more accurate retrieval result.
For another example: the label is characterized by the character information of three pictures, and after all pictures belonging to three pictures are found in the pictures marked by the label, pictures which are not marked by the label and contain three pictures may exist. At this time, the picture belonging to the person of "zhang san" is obtained by searching the cluster corresponding to the face feature to which the found picture belongs, so as to obtain a more accurate retrieval result.
In step 407, the search result and the expanded search result are returned to the terminal, and the method of this embodiment is used in the server.
If the searched pictures exceed the display range of the current screen in the terminal display process, displaying the parts which can be displayed in the terminal screen according to the arrangement sequence in the extended search result, and sequentially displaying other pictures which are not displayed according to the sliding action of the user.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
EXAMPLE five
An embodiment of the present disclosure provides an apparatus for retrieving a picture, referring to fig. 5, the apparatus includes:
a first obtaining module 501, configured to obtain feature information sent by a terminal, where the feature information includes at least one of an image feature, a face feature, or a tag feature;
the retrieval module 502 is configured to retrieve a picture in the picture library, where the matching degree with the feature information is greater than a preset threshold, and determine the retrieved picture as a retrieval result;
the first selecting module 503 is configured to select, as an extended search result, a picture meeting a preset condition from a cluster in which the retrieved picture is located, where the cluster in which the retrieved picture is located is a cluster corresponding to a type of the feature information;
and a returning module 504, configured to return the search result and the extended search result to the terminal.
The first obtaining module 501 is configured to:
and receiving the picture to be detected sent by the terminal, and calculating to obtain the characteristic information in the picture to be detected.
Wherein, the device still includes:
a second obtaining module 505, configured to obtain, according to the identifier of the terminal, pictures stored in other picture libraries by the terminal and feature information corresponding to the pictures stored in the other picture libraries by the terminal;
and the clustering module 506 is configured to cluster the pictures stored in the other picture libraries by the terminal according to the feature information corresponding to the pictures stored in the other picture libraries by the terminal, and store the clustered pictures in the own picture library.
Wherein, the retrieving module 502 comprises:
and the first retrieval unit is used for calculating the matching degree between the image characteristics and the image characteristics corresponding to each picture in the picture library when the characteristic information is the image characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
Wherein, the retrieving module 502 comprises:
and the second retrieval unit is used for calculating the matching degree between the human face characteristics and the human face characteristics of each picture in the picture library when the characteristic information is the human face characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
Wherein the second retrieval unit is configured to:
and calculating the matching degree between each feature point in the human face features and each feature point in each picture in the picture library.
Wherein, the retrieving module 502 comprises:
and the third retrieval unit is used for calculating the matching degree between the tag characteristics and the tag characteristics corresponding to each picture in the picture library when the characteristic information is the tag characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
Wherein, the third search unit includes:
the first retrieval subunit is used for calculating the matching degree of the time information in the tag characteristics and the time information in the tag characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture;
and the second retrieval subunit is used for calculating the matching degree of the position information in the label characteristics and the position information in the label characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
Wherein, the device still includes:
and a second selecting module 507, configured to select other pictures in the cluster corresponding to the image feature or the cluster corresponding to the face feature as extended retrieval results.
The first selecting module 503 is configured to:
and selecting the picture with the similarity higher than the preset similarity with the retrieved picture as an extended retrieval result in the cluster of the retrieved pictures.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
EXAMPLE six
Fig. 6 is a block diagram illustrating an apparatus 1900 for retrieving pictures according to an example embodiment. For example, the apparatus 1900 may be provided as a server. Referring to fig. 6, the device 1900 includes a processing component 1922 further including one or more processors and memory resources, represented by memory 1932, for storing instructions, e.g., applications, executable by the processing component 1922. The application programs stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processing component 1922 is configured to execute instructions to perform the above-described method of retrieving pictures.
The device 1900 may also include a power component 1926 configured to perform power management of the device 1900, a wired or wireless network interface 1950 configured to connect the device 1900 to a network, and an input/output (I/O) interface 1958. The device 1900 may operate based on an operating system stored in memory 1932, such as Windows Server, MacOS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
According to the embodiment of the invention, the retrieved picture is used as the retrieval result, and other pictures meeting the preset condition in the cluster corresponding to the retrieved picture are selected as the extended retrieval result, so that the result meeting the requirements of the user can be retrieved, and the retrieval accuracy and efficiency are improved.
The above-mentioned serial numbers of the embodiments of the present disclosure are merely for description and do not represent the merits of the embodiments.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, so that any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (20)
1. A method for retrieving pictures is applied to a server, and the method comprises the following steps:
the method comprises the steps of obtaining a picture to be retrieved sent by a terminal, and obtaining feature information of the picture to be retrieved, wherein the feature information comprises multiple image features, human face features or label features;
searching pictures with matching degree greater than a preset threshold value with the characteristic information in a picture library, and determining the searched pictures as a searching result;
selecting pictures with the similarity between the searched pictures larger than the preset similarity as an extended search result in the cluster where the searched pictures are located, wherein the cluster where the searched pictures are located is a cluster corresponding to the type of the characteristic information, and the similarity between the pictures in each cluster is stored in the server;
and returning the retrieval result and the extended retrieval result to the terminal.
2. The method of claim 1, wherein the obtaining the feature information sent by the terminal comprises:
and receiving a picture to be detected sent by a terminal, and calculating in the picture to be detected to obtain characteristic information.
3. The method according to claim 1, wherein before retrieving the picture with the matching degree with the feature information greater than a preset threshold from the picture library, the method further comprises:
acquiring pictures stored in other picture libraries by the terminal and characteristic information corresponding to the pictures stored in the other picture libraries by the terminal according to the identification of the terminal;
and clustering the pictures stored in other picture libraries by the terminal according to the characteristic information corresponding to the pictures stored in other picture libraries by the terminal and storing the pictures in the picture library of the terminal.
4. The method according to claim 1, wherein the retrieving the picture with the matching degree with the feature information greater than a preset threshold from the picture library comprises:
and when the characteristic information is an image characteristic, calculating the matching degree between the image characteristic and the image characteristic corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
5. The method according to claim 1, wherein the retrieving the picture with the matching degree with the feature information greater than a preset threshold from the picture library comprises:
and when the characteristic information is the face characteristic, calculating the matching degree between the face characteristic and the face characteristic of each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
6. The method of claim 5, wherein the calculating the degree of matching between the facial features and the facial features of each picture in the picture library comprises:
and calculating the matching degree between each feature point in the face features and each feature point in each picture in the picture library.
7. The method according to claim 1, wherein the retrieving the picture with the matching degree with the feature information greater than a preset threshold from the picture library comprises:
and when the characteristic information is a label characteristic, calculating the matching degree between the label characteristic and the label characteristic corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
8. The method according to claim 7, wherein the calculating a matching degree between the tag feature and a tag feature corresponding to each picture in the picture library, and selecting a picture with a matching degree greater than a preset threshold as the retrieved picture comprises:
calculating the matching degree of the time information in the label characteristics and the time information in the label characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as a retrieved picture; or,
and calculating the matching degree of the position information in the label characteristics and the position information in the label characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
9. The method according to claim 7, wherein after selecting other pictures in the cluster where the retrieved picture is located as the extended retrieval result, the method further comprises:
and selecting other pictures in the cluster corresponding to the image characteristics or the cluster corresponding to the face characteristics as the extended retrieval result.
10. An apparatus for retrieving pictures, applied to a server, the apparatus comprising:
the first acquisition module is used for acquiring feature information sent by a terminal, wherein the feature information comprises at least one of image features, human face features or label features;
the retrieval module is used for retrieving pictures with the matching degree with the characteristic information being greater than a preset threshold value in a picture library and determining the retrieved pictures as retrieval results;
the first selection module is used for selecting pictures with the similarity higher than the preset similarity with the retrieved pictures in the clusters where the retrieved pictures are located as extended retrieval results, the clusters where the retrieved pictures are located are clusters corresponding to the types of the characteristic information, and the similarity between the pictures in each cluster is stored in the server;
and the return module is used for returning the retrieval result and the extended retrieval result to the terminal.
11. The apparatus of claim 10, wherein the first obtaining module is configured to:
and receiving a picture to be detected sent by a terminal, and calculating in the picture to be detected to obtain characteristic information.
12. The apparatus of claim 10, further comprising:
the second acquisition module is used for acquiring pictures stored in other picture libraries by the terminal and the characteristic information corresponding to the pictures stored in the other picture libraries by the terminal according to the identifier of the terminal;
and the clustering module is used for clustering the pictures stored in the other picture libraries by the terminal according to the characteristic information corresponding to the pictures stored in the other picture libraries by the terminal and storing the pictures in the picture library of the terminal.
13. The apparatus of claim 10, wherein the retrieving module comprises:
and the first retrieval unit is used for calculating the matching degree between the image characteristics and the image characteristics corresponding to each picture in the picture library when the characteristic information is the image characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
14. The apparatus of claim 10, wherein the retrieving module comprises:
and the second retrieval unit is used for calculating the matching degree between the human face characteristics and the human face characteristics of each picture in the picture library when the characteristic information is the human face characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
15. The apparatus of claim 14, wherein the second retrieving unit is configured to:
and calculating the matching degree between each feature point in the face features and each feature point in each picture in the picture library.
16. The apparatus of claim 14, wherein the retrieving module comprises:
and the third retrieval unit is used for calculating the matching degree between the tag characteristics and the tag characteristics corresponding to each picture in the picture library when the characteristic information is the tag characteristics, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
17. The apparatus of claim 16, wherein the third retrieving unit comprises:
the first retrieval subunit is used for calculating the matching degree of the time information in the tag characteristics and the time information in the tag characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture;
and the second retrieval subunit is used for calculating the matching degree of the position information in the label characteristics and the position information in the label characteristics corresponding to each picture in the picture library, and selecting the picture with the matching degree larger than a preset threshold value as the retrieved picture.
18. The apparatus of claim 16, further comprising:
and the second selection module is used for selecting other pictures in the cluster corresponding to the image characteristics or the cluster corresponding to the face characteristics as the extended retrieval result.
19. A server, characterized in that the server comprises a processor and a memory, in which a program code is stored, which program code is loaded and executed by the processor to implement the method of retrieving pictures as claimed in any one of claims 1-9.
20. A storage medium having stored thereon instructions for execution by a processor to perform a method of retrieving a picture according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410239452.5A CN105243060B (en) | 2014-05-30 | 2014-05-30 | A kind of method and device of retrieving image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410239452.5A CN105243060B (en) | 2014-05-30 | 2014-05-30 | A kind of method and device of retrieving image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105243060A CN105243060A (en) | 2016-01-13 |
| CN105243060B true CN105243060B (en) | 2019-11-08 |
Family
ID=55040712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410239452.5A Active CN105243060B (en) | 2014-05-30 | 2014-05-30 | A kind of method and device of retrieving image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105243060B (en) |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10891019B2 (en) * | 2016-02-29 | 2021-01-12 | Huawei Technologies Co., Ltd. | Dynamic thumbnail selection for search results |
| CN106021704A (en) * | 2016-05-17 | 2016-10-12 | 京东方科技集团股份有限公司 | Dressing case, control method of dressing case and makeup system |
| CN105956186A (en) * | 2016-06-03 | 2016-09-21 | 捷开通讯科技(上海)有限公司 | Picture search system and method |
| CN106446108B (en) * | 2016-09-14 | 2019-11-19 | 维沃移动通信有限公司 | A method for sharing pictures, mobile terminal and server |
| CN106874513A (en) * | 2017-03-07 | 2017-06-20 | 维沃移动通信有限公司 | A kind of photo lookup method and mobile terminal |
| CN108664526B (en) * | 2017-04-01 | 2022-04-29 | 华为技术有限公司 | Method and device for retrieval |
| CN108108711B (en) * | 2017-12-29 | 2019-12-17 | 深圳云天励飞技术有限公司 | Face control method, electronic device and storage medium |
| CN108921034A (en) | 2018-06-05 | 2018-11-30 | 北京市商汤科技开发有限公司 | Face matching process and device, storage medium |
| CN109241314A (en) * | 2018-08-27 | 2019-01-18 | 维沃移动通信有限公司 | Method and device for selecting similar images |
| CN109308299B (en) * | 2018-09-12 | 2020-01-14 | 北京字节跳动网络技术有限公司 | Method and apparatus for searching information |
| CN111078924B (en) * | 2018-10-18 | 2024-03-01 | 深圳云天励飞技术有限公司 | Image retrieval method, device, terminal and storage medium |
| CN109711298B (en) * | 2018-12-14 | 2021-02-12 | 南京甄视智能科技有限公司 | Method and system for efficient face characteristic value retrieval based on faiss |
| CN110083732B (en) * | 2019-03-12 | 2021-08-31 | 浙江大华技术股份有限公司 | Image retrieval method, device and computer storage medium |
| CN110134815A (en) * | 2019-04-15 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Image processing method, device, computer equipment and storage medium |
| CN110750661B (en) * | 2019-09-04 | 2022-09-16 | 成都华为技术有限公司 | Method, device, computer equipment and storage medium for searching image |
| CN110851634B (en) * | 2019-11-18 | 2022-07-19 | 广东三维家信息科技有限公司 | Picture retrieval method and device and electronic equipment |
| CN111881322B (en) * | 2020-09-28 | 2020-12-25 | 成都睿沿科技有限公司 | Target searching method and device, electronic equipment and storage medium |
| CN112232290B (en) * | 2020-11-06 | 2023-12-08 | 四川云从天府人工智能科技有限公司 | Data clustering method, server, system and computer readable storage medium |
| CN112182262B (en) * | 2020-11-30 | 2021-03-19 | 江西师范大学 | Image query method based on feature classification |
| CN115146098A (en) * | 2021-03-31 | 2022-10-04 | 北京达佳互联信息技术有限公司 | Method, device, electronic device and storage medium for arranging pictures for text information |
| CN113286040B (en) * | 2021-05-17 | 2024-01-23 | 广州三星通信技术研究有限公司 | Method for selecting image and apparatus therefor |
| CN113360698A (en) * | 2021-06-30 | 2021-09-07 | 北京海纳数聚科技有限公司 | Picture retrieval method based on image-text semantic transfer technology |
| CN115168637B (en) * | 2022-06-17 | 2023-03-17 | 杭州晶彩数字科技有限公司 | Method, system and storage medium for adding label to picture |
| CN114880514B (en) * | 2022-07-05 | 2022-11-01 | 人民中科(北京)智能技术有限公司 | Image retrieval method, image retrieval device and storage medium |
| CN115455061B (en) * | 2022-11-14 | 2023-04-18 | 广东电网有限责任公司佛山供电局 | User characteristic quick retrieval method based on heterogeneous computing |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859326A (en) * | 2010-06-09 | 2010-10-13 | 南京大学 | An Image Retrieval Method |
| CN102842042A (en) * | 2011-06-07 | 2012-12-26 | 埃森哲环球服务有限公司 | Biometric authentication technology |
| CN103312730A (en) * | 2012-03-07 | 2013-09-18 | 腾讯科技（深圳）有限公司 | Picture storage method, server, terminal and picture storage system |
| CN103678661A (en) * | 2013-12-24 | 2014-03-26 | 中国联合网络通信集团有限公司 | Image searching method and terminal |
-
2014
- 2014-05-30 CN CN201410239452.5A patent/CN105243060B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859326A (en) * | 2010-06-09 | 2010-10-13 | 南京大学 | An Image Retrieval Method |
| CN102842042A (en) * | 2011-06-07 | 2012-12-26 | 埃森哲环球服务有限公司 | Biometric authentication technology |
| CN103312730A (en) * | 2012-03-07 | 2013-09-18 | 腾讯科技（深圳）有限公司 | Picture storage method, server, terminal and picture storage system |
| CN103678661A (en) * | 2013-12-24 | 2014-03-26 | 中国联合网络通信集团有限公司 | Image searching method and terminal |
Non-Patent Citations (1)
| Title |
|---|
| 基于内容的图象检索系统;唐立军等;《计算机应用研究》;20011231(第7期);41—45 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105243060A (en) | 2016-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105243060B (en) | A kind of method and device of retrieving image | |
| US11120078B2 (en) | Method and device for video processing, electronic device, and storage medium | |
| US11657084B2 (en) | Correlating image annotations with foreground features | |
| US20190188222A1 (en) | Thumbnail-Based Image Sharing Method and Terminal | |
| JP6759844B2 (en) | Systems, methods, programs and equipment that associate images with facilities | |
| US9928397B2 (en) | Method for identifying a target object in a video file | |
| CN110348362B (en) | Label generation method, video processing method, device, electronic equipment and storage medium | |
| WO2020006835A1 (en) | Customer service method, apparatus, and device for engaging in multiple rounds of question and answer, and storage medium | |
| US20090083275A1 (en) | Method, Apparatus and Computer Program Product for Performing a Visual Search Using Grid-Based Feature Organization | |
| US20080270378A1 (en) | Method, Apparatus and Computer Program Product for Determining Relevance and/or Ambiguity in a Search System | |
| US20130148898A1 (en) | Clustering objects detected in video | |
| JP2020042834A (en) | Account information acquisition method, terminal, server, and system | |
| US9652534B1 (en) | Video-based search engine | |
| US20120155717A1 (en) | Image search including facial image | |
| CN111046235A (en) | Method, system, equipment and medium for searching acoustic image archive based on face recognition | |
| CN102968473A (en) | Information retrieval method and system based on face image | |
| JP2017182628A (en) | Augmented reality user interface application apparatus and control method | |
| CN118051653B (en) | Multi-mode data retrieval method, system and medium based on semantic association | |
| CN107679128B (en) | Information display method and device, electronic equipment and storage medium | |
| US9977793B2 (en) | Information processing system, information processing method, and information processing apparatus | |
| KR20210058157A (en) | Apparatus and method for identificating user based on face image | |
| CN113869063A (en) | Data recommendation method, device, electronic device and storage medium | |
| WO2021196551A1 (en) | Image retrieval method and apparatus, computer device, and storage medium | |
| CN104850600B (en) | A kind of method and apparatus for searching for the picture comprising face | |
| CN112069331A (en) | Data processing method, data retrieval method, data processing device, data retrieval device, data processing equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |