CN111339335A - Image retrieval method, image retrieval device, storage medium and electronic equipment - Google Patents
Image retrieval method, image retrieval device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111339335A CN111339335A CN202010152844.3A CN202010152844A CN111339335A CN 111339335 A CN111339335 A CN 111339335A CN 202010152844 A CN202010152844 A CN 202010152844A CN 111339335 A CN111339335 A CN 111339335A
- Authority
- CN
- China
- Prior art keywords
- retrieval
- text
- image
- standby
- target
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 239000013598 vector Substances 0.000 claims description 76
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000003287 optical effect Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Library & Information Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses an image retrieval method, an image retrieval device, a storage medium and electronic equipment, wherein the image retrieval method comprises the following steps: receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request; acquiring a plurality of standby retrieval texts; determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text; and retrieving a first target image matched with the target retrieval text. The image retrieval scheme provided by the embodiment performs image retrieval based on the target retrieval text with the similarity to the candidate retrieval text meeting the preset condition, and can improve the success rate of image retrieval of the electronic device.
Description
Technical Field
The present application belongs to the field of image technologies, and in particular, to an image retrieval method, an image retrieval device, a storage medium, and an electronic device.
Background
Along with the continuous development of the intelligent terminal, the shooting function of the intelligent terminal is increasingly powerful, and a user can record the drips of daily life through the shooting function of the intelligent terminal. This is when convenience of customers shoots, leads to image pile up in a large number easily among the intelligent terminal, increases the degree of difficulty that the user looked for the image.
In the related art, a user can search for an image through an image retrieval function of an intelligent terminal. However, this image search method has a low search success rate.
Disclosure of Invention
The embodiment of the application provides an image retrieval method, an image retrieval device, a storage medium and electronic equipment, which can improve the success rate of image retrieval.
In a first aspect, an embodiment of the present application provides an image retrieval method, including:
receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request;
acquiring a plurality of standby retrieval texts;
determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and retrieving a first target image matched with the target retrieval text.
In a second aspect, an embodiment of the present application provides an image retrieval apparatus, including:
the first determining module is used for receiving an image retrieval request and determining candidate retrieval texts according to the image retrieval request;
the acquisition module is used for acquiring a plurality of standby retrieval texts;
the second determining module is used for determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and the first retrieval module is used for retrieving a first target image matched with the target retrieval text.
In a third aspect, a storage medium is provided in this application, and a computer program is stored thereon, and when the computer program runs on a computer, the computer is caused to execute an image retrieval method as provided in any embodiment of this application.
In a fourth aspect, an electronic device provided in an embodiment of the present application includes a processor and a memory, where the memory has a computer program, and the processor is configured to execute the image retrieval method provided in any embodiment of the present application by calling the computer program.
According to the image retrieval scheme provided by the embodiment of the application, when an image retrieval request is received, the electronic equipment determines candidate retrieval texts according to the image retrieval request, obtains a plurality of standby retrieval texts, determines the standby retrieval texts with the similarity meeting the preset condition with the candidate retrieval texts from the standby retrieval texts to serve as target retrieval texts, and finally retrieves a first target image matched with the target retrieval texts. According to the scheme, the image retrieval is carried out on the basis of the target retrieval text with the similarity meeting the preset condition with the candidate retrieval text, so that the image retrieval success rate of the electronic equipment can be improved.
Drawings
The technical solutions and advantages of the present application will become apparent from the following detailed description of specific embodiments of the present application when taken in conjunction with the accompanying drawings.
Fig. 1 is a scene schematic diagram of an image retrieval method according to an embodiment of the present application.
Fig. 2 is a schematic first flowchart of an image retrieval method according to an embodiment of the present application.
Fig. 3 is a second flowchart of the image retrieval method according to the embodiment of the present application.
Fig. 4 is a schematic structural diagram of an image retrieval device according to an embodiment of the present application.
Fig. 5 is a schematic view of a first structure of an electronic device according to an embodiment of the present application.
Fig. 6 is a second structural schematic diagram of an electronic device provided in the embodiment of the present application.
Detailed Description
The following description is based on illustrated embodiments of the application and should not be taken as limiting the application with respect to other embodiments that are not detailed herein. The term "module" as used herein may be considered a software object executing on the computing system. The various modules, engines, and services herein may be considered as objects of implementation on the computing system.
According to the image retrieval method provided by the embodiment of the application, the execution main body of the image retrieval method can be the image retrieval device provided by the embodiment of the application or the electronic equipment integrated with the image retrieval device. The electronic device may be a smart phone, a tablet computer, a Personal Digital Assistant (PDA), or the like.
Referring to fig. 1, fig. 1 is a scene schematic diagram of an image retrieval method according to an embodiment of the present application. For example, if the user inputs "single car" to perform image search, it is assumed that the electronic device fails to perform image search based on "single car" as a search criterion. According to the image retrieval method provided by the embodiment of the application, when the user inputs the candidate retrieval text 'bicycle' in the electronic equipment, the electronic equipment can determine the target retrieval text 'bicycle' with the similarity meeting the preset condition with the candidate retrieval text 'bicycle' from the plurality of standby retrieval texts. Then the electronic equipment searches images by taking the target search text 'bicycle' as an image search basis, searches images with the image characteristics of 'bicycle', and can improve the success rate of image search of the electronic equipment.
Referring to fig. 2, fig. 2 is a first flowchart of an image retrieval method according to an embodiment of the present disclosure, where the image retrieval method includes the following steps:
101. and receiving an image retrieval request, and determining candidate retrieval texts according to the image retrieval request.
In the embodiment of the application, when an image retrieval request of a user is received, the electronic device can analyze the image retrieval request to obtain a candidate retrieval text. It should be noted that the candidate search texts may be presented in the form of short sentences or words.
Wherein, the candidate search text may be an input text of the user in the image search bar. For example, the candidate search text is the input text of the user in the image search field: "Xiaowang Beijing". For example, the candidate search text is the input text of the user in the image search field: "Beijing tourist image of King", etc.
Alternatively, the candidate search text may be a text extracted from the input text of the image search field by the user. For example, the candidate search text is a text extracted from the input text "beijing tourism image of queen" of the user in the image search field: "Xiaowang Beijing Tourism" and so on.
In addition, the embodiment of the present application is not particularly limited as to the manner of extracting the candidate search text from the input text. For example, the electronic device performs a word segmentation operation on an input text to obtain a plurality of words, then combines the words to make a sentence, and takes a word corresponding to a sentence expressing the closest meaning to the input text as a candidate search text.
102. A plurality of alternate search texts are obtained.
In the embodiment of the application, when receiving an image retrieval request, the electronic device may obtain a plurality of pre-stored standby retrieval texts. Alternatively, the electronic device may obtain a plurality of alternative search texts at the same time.
For example, the electronic device may collect a plurality of input texts used by the user for image retrieval as a plurality of alternative retrieval texts at intervals of a preset duration. Then, the data stored in the preset storage area is deleted, and a plurality of standby retrieval texts are stored in the preset storage area. The preset storage area is used for storing a plurality of standby retrieval texts acquired each time.
The standby retrieval text is a text obtained by the electronic equipment in a specific range, and is mainly used as a retrieval basis of the first target image. For example, the alternate retrieval text is a preset tag text corresponding to a preset image stored in the electronic device. For example, the alternative search text is all input text that the user of the electronic device enters in the image search field. For example, the alternative search text is an input text that is frequently searched by the user among all input texts input in the image search field.
103. And determining the alternative retrieval texts with the similarity meeting the preset condition with the candidate retrieval texts from the plurality of alternative retrieval texts as target retrieval texts.
In the embodiment of the application, after determining the candidate retrieval texts and obtaining the plurality of standby retrieval texts, the electronic device may obtain the similarity between each standby retrieval text and the candidate retrieval text, and determine, from the plurality of standby retrieval texts, a target retrieval text whose similarity with the candidate retrieval text meets a preset condition.
The similarity satisfying the preset condition may refer to a similarity with a larger value among the plurality of similarities. The similarity satisfying the preset condition may refer to a similarity having a value greater than a preset threshold among the plurality of similarities, and the like.
104. And retrieving a first target image matched with the target retrieval text.
In the embodiment of the application, after the target retrieval text with the similarity meeting the preset condition with the candidate retrieval text is determined, the electronic device may retrieve a preset image matched with the target retrieval text from a plurality of preset images as the first target image.
For example, the electronic device retrieves a preset image, the image content of which matches the target retrieval text, as the first target image according to the image content of the preset image. For example, assuming that the target search text is a "small wind smile", the preset image whose image content matches the candidate search text refers to a preset image having a "small wind smile", or the like.
For example, the electronic device retrieves a preset image, as a second target image, in which the preset tag text matches the target retrieval text according to the preset tag text of the preset image. The electronic equipment stores a plurality of preset images in advance, and each preset image is correspondingly provided with at least one preset label text.
In addition, as for the source of the plurality of preset images, the embodiment of the present application is not particularly limited, for example, the electronic device uses an image obtained by the camera as the preset image. For example, the electronic device takes an image downloaded by the target server as a preset image.
The first target image matched with the target retrieval text in the plurality of preset images may be a preset image in which image contents in the plurality of preset images are consistent with the target retrieval text, or may be a preset image in which a preset label text in the plurality of preset images is consistent with the target retrieval text.
The scheme is used for searching the image according to the target search text with the similarity meeting the preset condition with the candidate search text, so that the image search failure caused by the fact that the text input by the user does not meet the specification can be effectively reduced. For example, the user's input text is: the relevant specification text for a bicycle, but an electronic device, is: the bicycle, if the electronic device performs image retrieval based on the input text "bicycle", the electronic device may not retrieve the image related to "bicycle". And if the electronic equipment determines the target retrieval text bicycle with the similarity meeting the preset condition with the candidate retrieval text bicycle according to the scheme of the application, image retrieval is carried out by taking the target retrieval text bicycle as an image retrieval basis, and the image of bicycle is retrieved.
As can be seen from the above, in the image retrieval method provided in the embodiment of the present application, when receiving an image retrieval request, the electronic device determines a candidate retrieval text according to the image retrieval request, obtains a plurality of standby retrieval texts, then determines, from the plurality of standby retrieval texts, a standby retrieval text whose similarity with the candidate retrieval text satisfies a preset condition, as a target retrieval text, and finally retrieves a first target image matched with the target retrieval text. According to the scheme, the image retrieval is carried out on the basis of the target retrieval text with the similarity meeting the preset condition with the candidate retrieval text, so that image retrieval failure caused by the fact that the text input by a user does not meet the specification can be effectively reduced, and the image retrieval success rate of the electronic equipment is improved.
In some embodiments, after determining, as the target search text, an alternative search text whose similarity to the candidate search text satisfies a preset condition from the plurality of alternative search texts, the electronic device may perform the following steps:
displaying the target retrieval text for a user to determine an actual retrieval text from the target retrieval text;
the first target image matched with the target retrieval text is retrieved, and the electronic equipment can execute the following steps:
and retrieving a third target image matched with the actual retrieval text.
In the scheme, after the electronic device determines the target retrieval text, the electronic device can display the target retrieval text to enable a user to determine the actual retrieval text of the image which the user actually wants to retrieve. When the actual retrieval text determined by the user is received, the electronic device may retrieve a preset image matching the actual retrieval text from the plurality of preset images as a third target image.
For example, the electronic device retrieves a preset image, the image content of which matches the actual retrieval text, as the third target image according to the image content of the preset image. For example, assuming that the actual search text is "clear sky", the preset image whose image content matches the actual search text is a preset image having an image characteristic of "blue sky", and the like.
For example, the electronic device retrieves a preset image, as a third target image, in which the preset tag text matches the actual retrieved text according to the preset tag text of the preset image. The electronic equipment stores a plurality of preset images in advance, and each preset image is correspondingly provided with at least one preset label text.
It should be noted that, after the target retrieval text is determined, the electronic device does not directly perform image retrieval based on the target retrieval text, but displays the target retrieval text to allow the user to select the actual retrieval text desired by the user, so that the accuracy of image retrieval can be improved.
In addition, in the scheme, even if the target retrieval text displayed by the electronic equipment does not have the actual retrieval text wanted by the user, the intelligent sense of the electronic equipment in the endeavor can be created for the user, and the use experience of the user is improved. For example, after the electronic device displays the target search text, if the actual search text determined by the user is not received within the set time length, the electronic device may perform image search and the like by using the target search text as a search basis.
Referring to fig. 3, fig. 3 is a second flow chart of the image retrieval method according to the embodiment of the present application, where the image retrieval method includes the following steps:
201. and receiving an image retrieval request, and determining candidate retrieval texts according to the image retrieval request.
In the embodiment of the application, when an image retrieval request of a user is received, the electronic device can analyze the image retrieval request to obtain a candidate retrieval text.
Wherein, the candidate search text may be an input text of the user in the image search bar. Alternatively, the candidate search text may be a text extracted from the input text of the image search field by the user.
202. And retrieving the second target image matched with the candidate retrieval text.
In the embodiment of the application, after determining the candidate retrieval text, the electronic device may use the candidate retrieval text as an image retrieval basis, and retrieve a preset image matched with the candidate retrieval text from a plurality of preset images as a second target image. The second target image is an image retrieval result based on the candidate retrieval text.
For example, the electronic device retrieves a preset image, the image content of which matches the candidate retrieval text, as the second target image according to the image content of the preset image. For example, assuming that the candidate search text is "sheetlet", the preset image whose image content matches the candidate search text is referred to as a preset image having "sheetlet", or the like.
For example, the electronic device retrieves a preset image, as the second target image, in which the preset tag text matches the candidate retrieval text, according to the preset tag text of the preset image. The electronic equipment stores a plurality of preset images in advance, and each preset image is correspondingly provided with at least one preset label text.
The electronic device searches the second target image matched with the candidate search text, and two results can be obtained: search success and search failure. The successful retrieval means that the electronic equipment retrieves the second target image matched with the candidate retrieval text. The retrieval failure means that the electronic device does not retrieve the second target image matched with the candidate retrieval text.
In order to improve the success rate of the image search of the electronic device, when the second target image search is successful, the image search of the electronic device is successful, and at this time, the electronic device may continue to execute 203 to expand the searched image, or may end the image search process. When the second target image search fails, it indicates that the image search of the electronic device has not been successful, and at this time, the electronic device needs to continue to execute 203 to improve the success rate of the image search.
In some embodiments, after retrieving the second target image that matches the candidate retrieved text, the electronic device may display the second target image.
203. And when the second target image matched with the candidate retrieval text is not retrieved, acquiring a plurality of standby retrieval texts.
In the embodiment of the application, when the second target image matched with the candidate retrieval text is not retrieved, the electronic device may acquire a plurality of pre-stored standby retrieval texts. Alternatively, the electronic device may obtain a plurality of alternative search texts at the same time.
The standby retrieval text is a text obtained by the electronic equipment in a specific range, and is mainly used as a retrieval basis of the first target image. For example, the alternate retrieval text is a preset tag text corresponding to a preset image stored in the electronic device. For example, the alternative search text is all input text that the user of the electronic device enters in the image search field. For example, the alternative search text is an input text that is frequently searched by the user among all input texts input in the image search field.
In some embodiments, when obtaining the plurality of alternate search texts, the electronic device may perform the following:
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images.
The historical viewing record refers to the viewing record of the preset image by the user before the current moment. The standby image may be a preset image viewed by all users or a preset image viewed by a part of users. It can be understood that the preset label text corresponding to the standby image in the scheme actually refers to the preset label text corresponding to the preset image.
For example, when the standby image is determined from the plurality of preset images according to the history viewing record, the electronic device may determine the history viewing image with the viewing time within a preset time period taking the current time as the end time as the standby image, where the history viewing record includes the viewing time and the history viewing image. The history viewing image refers to a preset image viewed by the user.
In this scheme, each time the user viewing an image is detected, the electronic device records the image viewed by the user and the viewing time in the history viewing record. It is understood that the image viewed by the user in the history viewing record is the history viewing image.
For example, the first target image matching the target search text is searched. After a first target image matched with the target retrieval text is retrieved, the electronic equipment displays the first target image for a user to view; and adding the first target image viewed by the user as a history viewing image into the history viewing record, and adding the viewing time corresponding to the first target image viewed by the user into the history viewing record.
It should be noted that, in image retrieval, a user often bases on a certain rule. For example, when the user wants to compose a photo of the king and the plum about Beijing, the user first searches for "photos of the king to travel to Beijing", then searches for "photos of the plum to travel to Beijing", and so on. Based on this, the historical viewing image of the viewing time within the preset time period taking the current time as the end time is determined as the standby image, that is, the preset image which is recently viewed by the user is taken as the standby image, so that the image retrieval time of the electronic equipment can be reduced, and the image retrieval efficiency of the electronic equipment is improved.
For another example, when the alternative image is determined from the plurality of preset images according to the historical view record, the electronic device may obtain the user representation, and determine the alternative image from the historical view image in the historical view record according to the user representation. Wherein the history viewing record comprises history viewing images. The history viewing image refers to a preset image viewed by the user.
The user portrait is an effective tool for outlining the user and connecting the user appeal and the design direction. The electronic device can collect data of a user, tag the data, and embody the user image through the tag, thereby obtaining the user portrait. A user may have one portrait of one type, or a plurality of portraits of different types, such as basic portraits including basic data of age, sex, occupation, academic calendar, behavior characteristics portraits including behavior data of application, trip, etc.
It will be appreciated that different types of data constituting the tags of the representation will result in different types of representations being obtained. Therefore, the electronic equipment can continuously update the user portrait by using different target data so as to improve the accuracy of the user portrait.
For example, the electronic device obtains a new user portrait according to the target data, using the data collected within a preset time as the target data, so as to replace the original user portrait, thereby realizing the updating of the user portrait. For example, in the electronic device of day 10/2, data a collected within a preset time (day 10/1 to day 9/2) is used as target data, and the user portrait a is obtained according to the target data, so that the electronic device takes the user portrait a as the portrait of the user of the electronic device in day 10/2 to day 9/3. And collecting data B as target data within a preset time (from 2 month 10 to 3 month 9) in 3 month 10, and obtaining a user portrait B according to the target data, wherein the user portrait B is taken as a portrait of a user of the electronic equipment in 2 month 10 to 3 month 9 by the electronic equipment, and the like.
It should be noted that the user image delineates the user from multiple aspects, and some preset images that the user is interested in can be selected from all the preset images that the user has viewed according to the user image, so that the image retrieval time of the electronic device can be reduced, and the image retrieval efficiency of the electronic device can be improved.
In addition, the embodiment of the present application is not particularly limited to the manner of acquiring the plurality of alternative search texts.
In some embodiments, when obtaining the plurality of alternate search texts, the electronic device may perform the following:
and acquiring a history retrieval record of successful retrieval, and determining a plurality of spare retrieval texts from the history retrieval texts of the history retrieval record.
The history search record is a search record of image search performed by the user before the current time. The history search text refers to an input text that the user has entered in the search field before the current time.
It should be noted that, in image retrieval, a user often bases on a certain rule. Based on the above, the history search text which is searched successfully once is used as the standby search text, so that the image search time of the electronic equipment can be reduced, and the image search efficiency of the electronic equipment can be improved.
In some embodiments, when obtaining the plurality of alternate search texts, the electronic device may perform the following:
acquiring a history retrieval record of successful retrieval;
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images and the history retrieval texts of the history retrieval records.
The historical viewing record refers to the viewing record of the preset image by the user before the current moment. The standby image may be a preset image viewed by all users or a preset image viewed by a part of users. It can be understood that the preset label text corresponding to the standby image in the scheme actually refers to the preset label text corresponding to the preset image.
The history search record is a search record of image search performed by the user before the current time. The history search text refers to an input text that the user has entered in the search field before the current time.
It should be noted that the alternative search text in this scheme has two sources: one is a preset label text corresponding to the standby image, and the other is a history retrieval text recorded in the history retrieval. The accuracy of image retrieval can be ensured to a certain extent.
204. The candidate search texts are converted into a first text vector using a pre-trained text model, and each of the alternative search texts is converted into a corresponding second text vector.
In this embodiment of the application, after obtaining a plurality of alternative search texts, the electronic device may convert the candidate search texts into a first text vector and convert each alternative search text into a corresponding second text vector using a pre-trained text model. The first text vector and the second text vector can be used as a calculation basis of the similarity.
Where a pre-trained text model may be used to convert text into vectors. It should be noted that the pre-trained text model may also be other models capable of converting text into vectors, and the embodiment of the present application is not limited in particular. For example, the pre-trained text model may be a model obtained by training a pre-constructed convolutional neural network model using sample text data. For example, the pre-trained text model may be a model obtained by training a pre-built word2vec model using sample text data, and the like.
The vector (Word Embedding) is a vector obtained by mapping a text to a real number, and the vector can be used for calculating the similarity between two texts.
205. And calculating the similarity of the candidate search texts and each standby search text according to the first text vector and the second text vector.
In this embodiment of the application, after obtaining the first text vector and the second text vector, the electronic device may calculate, by using the first text vector and the second text vector, a similarity between the candidate search text and each of the alternative search texts. It is understood that the number of alternative search texts is the same as the number of similarities.
Note that, the embodiment of the present application is not particularly limited as to the manner of calculating the similarity.
In some embodiments, when calculating the similarity between the candidate search text and each of the alternative search texts according to the first text vector and the second text vector, the electronic device may perform the following:
cosine calculation is carried out on each second text vector and each first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or performing Manhattan distance calculation on each second text vector and each first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or carrying out Euclidean distance calculation on each second text vector and each first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text.
Wherein, similarity1 represents the similarity between the candidate search text and each of the alternative search texts, P1 represents the cosine value between the candidate search text and each of the alternative search texts, AiRepresenting the ith first text vector corresponding to the candidate search text, BiAnd representing the ith second text vector corresponding to each standby retrieval text, wherein the cosine calculation formula is as follows:
wherein, similarity2 represents the similarity between the candidate search text and each of the alternative search texts, P2 represents the Manhattan distance value between the candidate search text and each of the alternative search texts, AiRepresenting the ith first text vector corresponding to the candidate search text, BiAnd expressing the ith second text vector corresponding to each standby retrieval text, wherein the Manhattan distance calculation formula is as follows:
wherein, similarity3 represents the similarity between the candidate search text and each of the alternative search texts, P3 represents the Euclidean distance value between the candidate search text and each of the alternative search texts, AiRepresenting the ith first text vector corresponding to the candidate search text, BiRepresenting the ith second text vector corresponding to each alternative search text,the euclidean distance calculation formula is:
in some embodiments, when calculating the similarity between the candidate search text and each of the alternative search texts according to the first text vector and the second text vector, the electronic device may further perform the following:
and for each second text vector, performing cosine calculation on the second text vector and the first text vector to obtain a cosine value, performing Manhattan distance calculation on the second text vector and the first text vector to obtain a Manhattan distance value, performing Euclidean distance calculation on the second text vector and the first text vector to obtain an Euclidean distance value, performing weighted summation on the cosine value, the Manhattan distance value and the Euclidean distance value, and taking the result of the weighted summation as the similarity between the candidate retrieval text and the standby retrieval text.
Wherein, similarity4 represents the similarity between each candidate search text and the alternative search text, p1Cosine values, p, representing each candidate search text and the alternative search text2Manhattan distance value, p, representing each candidate search text from the alternative search text3Representing the Euclidean distance value, k, of each candidate search text from the alternative search text1Represents a first weight value, k2Represents a second weight value, k3The third weight value is expressed, and the calculation formula of the similarity is as follows:
Similarity4＝k1p1+k2p2+k3p3
it should be noted that the cosine value p1Manhattan distance value p2Euclidean distance value p3The calculation formula is referred to the above formula, and is not described herein again. And the similarity between each candidate retrieval text and the standby retrieval text is calculated by integrating various indexes, so that the accuracy of the similarity can be improved, and the accuracy of image retrieval is further improved.
206. And sequencing according to the similarity of each standby retrieval text, and judging the standby retrieval text with the sequencing rank meeting the preset rank as the standby retrieval text with the similarity meeting the preset condition to obtain the target retrieval text.
In the embodiment of the application, after the similarity between the candidate retrieval text and each standby retrieval text is obtained, the electronic device can sequence the plurality of similarities, determine the standby retrieval text corresponding to the similarity with the ranking satisfying the preset ranking as the standby retrieval text with the similarity satisfying the preset condition, and take the standby retrieval text with the similarity satisfying the preset condition as the target retrieval text so as to perform image retrieval according to the target retrieval text.
For example, the electronic device sorts the plurality of similarities in order of the similarity from high to low, and determines that the similarity whose ranking satisfies a preset ranking (e.g., the top 5) satisfies a preset condition in the sorting. The preset name is preset in the electronic equipment.
For example, the electronic device sorts the plurality of similarities in order of similarity from low to high, and determines that the similarity whose ranking satisfies a preset ranking (e.g., the next 3) satisfies a preset condition in the sorting. The preset name is preset in the electronic equipment.
207. And retrieving a first target image matched with the target retrieval text.
In the embodiment of the application, after the target retrieval text is obtained, the electronic device may retrieve a preset image matched with the target retrieval text from a plurality of preset images as the first target image.
For example, the electronic device retrieves a preset image, the image content of which matches the target retrieval text, as the first target image according to the image content of the preset image. For example, assuming that the target search text is a "small wind smile", the preset image whose image content matches the candidate search text refers to a preset image having a "small wind smile", or the like.
For example, the electronic device retrieves a preset image, as a second target image, in which the preset tag text matches the target retrieval text according to the preset tag text of the preset image. The electronic equipment stores a plurality of preset images in advance, and each preset image is correspondingly provided with at least one preset label text.
In addition, as for the source of the plurality of preset images, the embodiment of the present application is not particularly limited, for example, the electronic device uses an image obtained by the camera as the preset image. For example, the electronic device takes an image downloaded by the target server as a preset image.
The first target image matched with the target retrieval text in the plurality of preset images may be a preset image in which image contents in the plurality of preset images are consistent with the target retrieval text, or may be a preset image in which a preset label text in the plurality of preset images is consistent with the target retrieval text.
In some embodiments, after retrieving the first target image that matches the target retrieval text, the electronic device may display the first target image.
Referring to fig. 4, fig. 4 is a schematic structural diagram of an image retrieval device according to an embodiment of the present disclosure. The image retrieval apparatus 300 includes: a first determining module 301, an obtaining module 302, a second determining module 303 and a first retrieving module 304.
A first determining module 301, configured to receive an image retrieval request, and determine a candidate retrieval text according to the image retrieval request;
an obtaining module 302, configured to obtain a plurality of standby retrieval texts;
a second determining module 303, configured to determine, from the multiple standby search texts, a standby search text whose similarity to the candidate search text meets a preset condition as a target search text;
and a first retrieving module 304, configured to retrieve a first target image matching the target retrieval text.
In some embodiments, when obtaining a plurality of alternative search texts, the obtaining module 302 may be further configured to:
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images.
In some embodiments, when determining the standby image from the plurality of preset images according to the history viewing record, the obtaining module 302 may further be configured to:
determining a historical viewing image with a viewing time within a preset time period taking the current time as an end time as a standby image, wherein the historical viewing record comprises the viewing time and the historical viewing image.
In some embodiments, after retrieving the first target image matching the target retrieval text, the image retrieval apparatus 300 further comprises:
the first display module is used for displaying the first target image for a user to view;
and the adding module is used for adding the first target image viewed by the user into the history viewing record as a history viewing image and adding the viewing time corresponding to the first target image viewed by the user into the history viewing record.
In some embodiments, when determining the standby image from the plurality of preset images according to the history viewing record, the obtaining module 302 may further be configured to:
obtaining a user portrait, and determining a standby image from a history viewing image in the history viewing record according to the user portrait, wherein the history viewing record comprises the history viewing image.
In some embodiments, when obtaining a plurality of alternative search texts, the obtaining module 302 may be further configured to:
and acquiring a history retrieval record of successful retrieval, and determining a plurality of standby retrieval texts from the history retrieval texts of the history retrieval record.
In some embodiments, when obtaining a plurality of alternative search texts, the obtaining module 302 may be further configured to:
acquiring a history retrieval record of successful retrieval;
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images and the history retrieval texts of the history retrieval records.
In some embodiments, when an alternative search text whose similarity to the candidate search text satisfies a preset condition is determined from the plurality of alternative search texts, as the target search text, the second determining module 303 may be further configured to:
converting the candidate search texts into first text vectors and converting each standby search text into corresponding second text vectors by using a pre-trained text model;
calculating the similarity between the candidate retrieval text and each standby retrieval text according to the first text vector and the second text vector;
and sequencing according to the similarity of each standby retrieval text, and judging the standby retrieval text with the sequencing rank meeting the preset rank as the standby retrieval text with the similarity meeting the preset condition so as to obtain the target retrieval text.
In some embodiments, when calculating the similarity between the candidate search text and each alternative search text according to the first text vector and the second text vector, the second determining module 303 may be further configured to:
performing cosine calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or performing Manhattan distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or carrying out Euclidean distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text.
In some embodiments, before acquiring the plurality of alternative search texts, the image search apparatus 300 further includes:
the second retrieval module is used for retrieving a second target image matched with the candidate retrieval text;
when a second target image matching the candidate search text is not retrieved, the obtaining module 302 performs the obtaining of the plurality of alternative search texts.
In some embodiments, after determining, as the target search text, an alternative search text whose similarity to the candidate search text satisfies a preset condition from the plurality of alternative search texts, the image search apparatus 300 further includes:
the second display module is used for displaying the target retrieval text so that a user can determine an actual retrieval text from the target retrieval text;
when the first target image matching with the target search text is retrieved, the first retrieval module 304 is further configured to:
the first retrieving module 304 is used for retrieving a third target image matching the actual retrieved text.
As can be seen from the above, in the image retrieval apparatus 300 provided in this embodiment of the application, the first determining module 301 receives an image retrieval request, determines candidate retrieval texts according to the image retrieval request, the obtaining module 302 obtains a plurality of standby retrieval texts, then the second determining module 303 determines, from the plurality of standby retrieval texts, a standby retrieval text whose similarity to the candidate retrieval text satisfies a preset condition as a target retrieval text, and finally the first retrieving module 304 retrieves a first target image matching the target retrieval text, which can improve the success rate of image retrieval of the electronic device.
An electronic device is further provided in an embodiment of the present application, please refer to fig. 5, and fig. 5 is a first structural schematic diagram of the electronic device provided in the embodiment of the present application. The electronic device 400 comprises a processor 401 and a memory 402. The processor 401 is electrically connected to the memory 402.
The processor 401 is a control center of the electronic device 400, connects various parts of the entire electronic device using various interfaces and lines, performs various functions of the electronic device 400 and processes data by running or loading a computer program stored in the memory 402 and calling data stored in the memory 402.
The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by operating the computer programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, a computer program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to use of the electronic device, and the like.
Further, the memory 402 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 access to the memory 402.
In this embodiment, the processor 401 in the electronic device 400 loads instructions corresponding to one or more processes of the computer program into the memory 402 according to the following steps, and the processor 401 runs the computer program stored in the memory 402, so as to implement various functions, as follows:
receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request;
acquiring a plurality of standby retrieval texts;
determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and retrieving a first target image matched with the target retrieval text.
Referring to fig. 6, fig. 6 is a second schematic structural diagram of an electronic device according to an embodiment of the present disclosure, which is different from the electronic device shown in fig. 5 in that the electronic device 400 further includes: a camera module 403, a display module 404, an input unit 405, and a power supply 406. The memory 402, the camera module 403, the rf circuit 404, the audio circuit 405, the input unit 405, and the power source 406 are electrically connected to the processor 401.
The camera assembly 403 may include Image Processing circuitry, which may be implemented using hardware and/or software components, and may include various Processing units that define an Image Signal Processing (Image Signal Processing) pipeline. The image processing circuit may include at least: a plurality of cameras, an Image Signal Processor (ISP Processor), control logic, an Image memory, and a display. Where each camera may include at least one or more lenses and an image sensor. The image sensor may include an array of color filters (e.g., Bayer filters). The image sensor may acquire light intensity and wavelength information captured with each imaging pixel of the image sensor and provide a set of raw image data that may be processed by an image signal processor.
The display component 404 may be used to display information entered by or provided to a user as well as various graphical user interfaces that may be composed of graphics, text, icons, video, and any combination thereof.
The input unit 405 may be used to receive input numbers, character information, or user characteristic information (e.g., fingerprint), and generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. Wherein, the input unit 405 may include a fingerprint recognition module.
The power supply 406 may be used to power various components of the electronic device 400. In some embodiments, power supply 406 may be logically coupled to processor 401 via a power management system, such that functions to manage charging, discharging, and power consumption management are performed via the power management system.
Although not shown, the electronic device may further include a radio frequency circuit, an audio circuit, and the like, which are not described in detail herein.
In this embodiment, the processor 401 in the electronic device 400 loads instructions corresponding to one or more processes of the computer program into the memory 402 according to the following steps, and the processor 401 runs the computer program stored in the memory 402, so as to implement various functions, as follows:
receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request;
acquiring a plurality of standby retrieval texts;
determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and retrieving a first target image matched with the target retrieval text.
In some embodiments, when obtaining a plurality of alternative search texts, processor 401 may perform:
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images.
In some embodiments, when determining the standby image from the plurality of preset images according to the history viewing record, the processor 401 may perform:
determining a historical viewing image with a viewing time within a preset time period taking the current time as an end time as a standby image, wherein the historical viewing record comprises the viewing time and the historical viewing image.
In some embodiments, after retrieving the first target image matching the target retrieval text, processor 401 may further perform:
displaying the first target image for a user to view;
and adding the first target image viewed by the user as a history viewing image into the history viewing record, and adding the viewing time corresponding to the first target image viewed by the user into the history viewing record.
In some embodiments, when determining the standby image from the plurality of preset images according to the history viewing record, the processor 401 may perform:
obtaining a user portrait, and determining a standby image from a history viewing image in the history viewing record according to the user portrait, wherein the history viewing record comprises the history viewing image.
In some embodiments, when obtaining a plurality of alternative search texts, processor 401 may perform:
and acquiring a history retrieval record of successful retrieval, and determining a plurality of standby retrieval texts from the history retrieval texts of the history retrieval record.
In some embodiments, when obtaining a plurality of alternative search texts, processor 401 may perform:
acquiring a history retrieval record of successful retrieval;
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images and the history retrieval texts of the history retrieval records.
In some embodiments, when an alternative search text whose similarity to the candidate search text satisfies a preset condition is determined from the plurality of alternative search texts as the target search text, the processor 401 may perform:
converting the candidate search texts into first text vectors and converting each standby search text into corresponding second text vectors by using a pre-trained text model;
calculating the similarity between the candidate retrieval text and each standby retrieval text according to the first text vector and the second text vector;
and sequencing according to the similarity of each standby retrieval text, and judging the standby retrieval text with the sequencing rank meeting the preset rank as the standby retrieval text with the similarity meeting the preset condition so as to obtain the target retrieval text.
In some embodiments, when calculating the similarity between the candidate search text and each alternative search text according to the first text vector and the second text vector, processor 401 may perform:
performing cosine calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or performing Manhattan distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or carrying out Euclidean distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text.
In some embodiments, before obtaining the plurality of alternative search texts, processor 401 may further perform:
retrieving a second target image matching the candidate retrieval text;
when a second target image matching the candidate search text is not retrieved, the processor 401 may execute the obtaining of the plurality of alternative search texts.
In some embodiments, after determining, as the target search text, an alternative search text whose similarity to the candidate search text satisfies a preset condition from the plurality of alternative search texts, processor 401 may further perform:
displaying the target retrieval text for a user to determine an actual retrieval text from the target retrieval text;
when a first target image matching the target search text is retrieved, the processor 401 may execute:
and retrieving a third target image matched with the actual retrieval text.
As can be seen from the above, when the electronic device receives an image retrieval request, the electronic device determines a candidate retrieval text according to the image retrieval request, obtains a plurality of standby retrieval texts, then determines a standby retrieval text having a similarity meeting a preset condition with the candidate retrieval text from the plurality of standby retrieval texts, uses the standby retrieval text as a target retrieval text, and finally retrieves a first target image matched with the target retrieval text. According to the scheme, the image retrieval is carried out on the basis of the target retrieval text with the similarity meeting the preset condition with the candidate retrieval text, so that the image retrieval success rate of the electronic equipment can be improved.
An embodiment of the present application further provides a storage medium, where the storage medium stores a computer program, and when the computer program runs on a computer, the computer is caused to execute the image retrieval method in any one of the above embodiments, such as: receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request; acquiring a plurality of standby retrieval texts; determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text; and retrieving a first target image matched with the target retrieval text.
In the embodiment of the present application, the storage medium may be a magnetic disk, an optical disk, a Read Only Memory (ROM), a Random Access Memory (RAM), or the like.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
It should be noted that, for the image retrieval method of the embodiment of the present application, it can be understood by a person skilled in the art that all or part of the process of implementing the image retrieval method of the embodiment of the present application can be completed by controlling the relevant hardware through a computer program, where the computer program can be stored in a computer readable storage medium, such as a memory of an electronic device, and executed by at least one processor in the electronic device, and during the execution process, the process of the embodiment of the image retrieval method can be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory, a random access memory, etc.
In the image retrieval device according to the embodiment of the present application, each functional module may be integrated into one processing chip, each module may exist alone physically, or two or more modules may be integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented as a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium such as a read-only memory, a magnetic or optical disk, or the like.
The image retrieval method, the image retrieval device, the storage medium and the electronic device provided by the embodiments of the present application are described in detail above, and a specific example is applied in the present application to explain the principle and the implementation of the present application, and the description of the above embodiments is only used to help understanding the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (14)
1. An image retrieval method, comprising:
receiving an image retrieval request, and determining a candidate retrieval text according to the image retrieval request;
acquiring a plurality of standby retrieval texts;
determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and retrieving a first target image matched with the target retrieval text.
2. The image retrieval method of claim 1, wherein the obtaining a plurality of alternative retrieval texts comprises:
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images.
3. The image retrieval method of claim 2, wherein the determining a backup image from a plurality of preset images according to the historical viewing record comprises:
determining a historical viewing image with a viewing time within a preset time period taking the current time as an end time as a standby image, wherein the historical viewing record comprises the viewing time and the historical viewing image.
4. The image retrieval method according to claim 3, wherein after retrieving the first target image matching the target retrieval text, further comprising:
displaying the first target image for a user to view;
and adding the first target image viewed by the user as a history viewing image into the history viewing record, and adding the viewing time corresponding to the first target image viewed by the user into the history viewing record.
5. The image retrieval method of claim 2, wherein the determining a backup image from a plurality of preset images according to the historical viewing record comprises:
obtaining a user portrait, and determining a standby image from a history viewing image in the history viewing record according to the user portrait, wherein the history viewing record comprises the history viewing image.
6. The image retrieval method of claim 1, wherein the obtaining a plurality of alternative retrieval texts comprises:
and acquiring a history retrieval record of successful retrieval, and determining a plurality of standby retrieval texts from the history retrieval texts of the history retrieval record.
7. The image retrieval method of claim 1, wherein the obtaining a plurality of alternative retrieval texts comprises:
acquiring a history retrieval record of successful retrieval;
acquiring a history viewing record of a preset image, and determining a standby image from a plurality of preset images according to the history viewing record;
and obtaining a plurality of standby retrieval texts based on the preset label texts corresponding to the standby images and the history retrieval texts of the history retrieval records.
8. The image retrieval method according to any one of claims 1 to 7, wherein the determining, as the target retrieval text, the alternative retrieval text whose similarity with the candidate retrieval text satisfies a preset condition from the plurality of alternative retrieval texts comprises:
converting the candidate search texts into first text vectors and converting each standby search text into corresponding second text vectors by using a pre-trained text model;
calculating the similarity between the candidate retrieval text and each standby retrieval text according to the first text vector and the second text vector;
and sequencing according to the similarity of each standby retrieval text, and judging the standby retrieval text with the sequencing rank meeting the preset rank as the standby retrieval text with the similarity meeting the preset condition so as to obtain the target retrieval text.
9. The image retrieval method of claim 8, wherein the calculating the similarity of the candidate retrieval text and each alternative retrieval text according to the first text vector and the second text vector comprises:
performing cosine calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or performing Manhattan distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text;
or carrying out Euclidean distance calculation on each second text vector and the first text vector to obtain the similarity between the candidate retrieval text and each standby retrieval text.
10. The image retrieval method according to any one of claims 1 to 7, wherein before the obtaining of the plurality of alternative retrieval texts, the method further comprises:
retrieving a second target image matching the candidate retrieval text;
when a second target image matched with the candidate search text is not searched out, the obtaining of the plurality of standby search texts is executed.
11. The image retrieval method according to claim 10, wherein after determining, as the target retrieval text, the alternative retrieval text whose similarity with the candidate retrieval text satisfies a preset condition from the plurality of alternative retrieval texts, the method further comprises:
displaying the target retrieval text for a user to determine an actual retrieval text from the target retrieval text;
the retrieving of the first target image matched with the target retrieval text comprises the following steps:
and retrieving a third target image matched with the actual retrieval text.
12. An image retrieval apparatus, comprising:
the first determining module is used for receiving an image retrieval request and determining candidate retrieval texts according to the image retrieval request;
the acquisition module is used for acquiring a plurality of standby retrieval texts;
the second determining module is used for determining a standby retrieval text with the similarity meeting a preset condition with the candidate retrieval text from the standby retrieval texts as a target retrieval text;
and the first retrieval module is used for retrieving a first target image matched with the target retrieval text.
13. A storage medium having stored thereon a computer program, characterized in that, when the computer program runs on a computer, it causes the computer to execute an image retrieval method according to any one of claims 1 to 11.
14. An electronic device comprising a processor, a memory, said memory having a computer program, wherein said processor is adapted to perform the image retrieval method of any of claims 1 to 11 by invoking said computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010152844.3A CN111339335A (en) | 2020-03-06 | 2020-03-06 | Image retrieval method, image retrieval device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010152844.3A CN111339335A (en) | 2020-03-06 | 2020-03-06 | Image retrieval method, image retrieval device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111339335A true CN111339335A (en) | 2020-06-26 |
Family
ID=71184091
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010152844.3A Pending CN111339335A (en) | 2020-03-06 | 2020-03-06 | Image retrieval method, image retrieval device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111339335A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766269A (en) * | 2021-03-04 | 2021-05-07 | 深圳康佳电子科技有限公司 | Picture text retrieval method, intelligent terminal and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462212A (en) * | 2014-11-04 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Information exhibiting method and device |
| CN104572889A (en) * | 2014-12-24 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method, device and system for recommending search terms |
| WO2018040503A1 (en) * | 2016-08-30 | 2018-03-08 | 北京百度网讯科技有限公司 | Method and system for obtaining search results |
| CN108804550A (en) * | 2018-05-21 | 2018-11-13 | 北京奇艺世纪科技有限公司 | A kind of query word expanding method, device and electronic equipment |
| CN109783727A (en) * | 2018-12-24 | 2019-05-21 | 东软集团股份有限公司 | Retrieve recommended method, device, computer readable storage medium and electronic equipment |
| CN110597987A (en) * | 2019-08-21 | 2019-12-20 | 微梦创科网络科技（中国）有限公司 | Search recommendation method and device |
| CN110765348A (en) * | 2019-09-17 | 2020-02-07 | 五八有限公司 | Hot word recommendation method and device, electronic equipment and storage medium |
-
2020
- 2020-03-06 CN CN202010152844.3A patent/CN111339335A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462212A (en) * | 2014-11-04 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Information exhibiting method and device |
| CN104572889A (en) * | 2014-12-24 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method, device and system for recommending search terms |
| WO2018040503A1 (en) * | 2016-08-30 | 2018-03-08 | 北京百度网讯科技有限公司 | Method and system for obtaining search results |
| CN108804550A (en) * | 2018-05-21 | 2018-11-13 | 北京奇艺世纪科技有限公司 | A kind of query word expanding method, device and electronic equipment |
| CN109783727A (en) * | 2018-12-24 | 2019-05-21 | 东软集团股份有限公司 | Retrieve recommended method, device, computer readable storage medium and electronic equipment |
| CN110597987A (en) * | 2019-08-21 | 2019-12-20 | 微梦创科网络科技（中国）有限公司 | Search recommendation method and device |
| CN110765348A (en) * | 2019-09-17 | 2020-02-07 | 五八有限公司 | Hot word recommendation method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 夏太寿等: "科技查新案例评析", 东南大学出版社, pages: 95 - 100 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766269A (en) * | 2021-03-04 | 2021-05-07 | 深圳康佳电子科技有限公司 | Picture text retrieval method, intelligent terminal and storage medium |
| CN112766269B (en) * | 2021-03-04 | 2024-03-12 | 深圳康佳电子科技有限公司 | Picture text retrieval method, intelligent terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163234B (en) | Model training method and device and storage medium | |
| CN109117777B (en) | Method and device for generating information | |
| CN110163115B (en) | Video processing method, device and computer readable storage medium | |
| CN111062871B (en) | Image processing method and device, computer equipment and readable storage medium | |
| CN111209970B (en) | Video classification method, device, storage medium and server | |
| EP2551792B1 (en) | System and method for computing the visual profile of a place | |
| CN111581433A (en) | Video processing method and device, electronic equipment and computer readable medium | |
| CN108319723A (en) | A kind of picture sharing method and device, terminal, storage medium | |
| CN110083729B (en) | Image searching method and system | |
| CN111209897B (en) | Video processing method, device and storage medium | |
| CN114332680A (en) | Image processing method, video searching method, image processing device, video searching device, computer equipment and storage medium | |
| CN113434716B (en) | Cross-modal information retrieval method and device | |
| CN114358203A (en) | Training method and device for image description sentence generation module and electronic equipment | |
| CN112765387A (en) | Image retrieval method, image retrieval device and electronic equipment | |
| CN112418327A (en) | Training method and device of image classification model, electronic equipment and storage medium | |
| CN112069338A (en) | Picture processing method and device, electronic equipment and storage medium | |
| CN111046203A (en) | Image retrieval method, image retrieval device, storage medium and electronic equipment | |
| CN114168768A (en) | Image retrieval method and related equipment | |
| CN111339335A (en) | Image retrieval method, image retrieval device, storage medium and electronic equipment | |
| CN112926300A (en) | Image searching method, image searching device and terminal equipment | |
| CN112328833A (en) | Label processing method and device and computer readable storage medium | |
| CN110489592B (en) | Video classification method, apparatus, computer device and storage medium | |
| CN113569094A (en) | Video recommendation method and device, electronic equipment and storage medium | |
| CN111177493A (en) | Data processing method, device, server and storage medium | |
| CN117938951B (en) | Information pushing method, device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |