CN112860929A

CN112860929A - Picture searching method and device, electronic equipment and storage medium

Info

Publication number: CN112860929A
Application number: CN202110178924.0A
Authority: CN
Inventors: 胥昕昂
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2021-02-09
Filing date: 2021-02-09
Publication date: 2021-05-28

Abstract

The application discloses a picture searching method, a device, electronic equipment and a storage medium, which relate to the technical field of data processing, and further relate to an intelligent searching technology and a natural language processing technology, wherein the picture searching method can be used for a full-text search engine and comprises the following steps: acquiring a picture search request; performing full text search according to the picture search request to obtain target picture engine storage data; and obtaining a picture search result according to the target picture engine storage data. According to the image searching method and device, the accuracy and the efficiency of image searching can be improved.

Description

Picture searching method and device, electronic equipment and storage medium

Technical Field

The application relates to the technical field of data processing, in particular to an intelligent search technology.

Background

The picture search is a service for providing related picture data on the internet or a picture storage database to a user through a search program. The purpose of picture search is to find out a specific picture required by a user, and belongs to an important branch of intelligent search technology.

Disclosure of Invention

The embodiment of the application provides a picture searching method and device, electronic equipment and a storage medium, so as to improve the accuracy and the efficiency of picture searching.

In a first aspect, an embodiment of the present application provides an image search method, which is applied to a full-text search engine, and includes:

acquiring a picture search request;

performing full text search according to the picture search request to obtain target picture engine storage data;

and obtaining a picture search result according to the target picture engine storage data.

In a second aspect, an embodiment of the present application provides an image search apparatus configured in a full-text search engine, including:

the image search request acquisition module is used for acquiring an image search request;

and the full text search module is used for performing full text search according to the picture search request to obtain target picture engine storage data.

In a third aspect, an embodiment of the present application provides an electronic device, including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the picture search method provided by the embodiment of the first aspect.

In a fourth aspect, the present application further provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the picture searching method provided in the first aspect.

In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the picture search method provided in the embodiment of the first aspect.

According to the image searching method and device, after the image searching request is obtained through the full-text searching engine, full-text searching is conducted according to the obtained image searching request to obtain the target image engine storage data, the image searching result is obtained according to the target image engine storage data obtained through searching, the problems that the existing image searching method is low in accuracy and efficiency and the like are solved, and therefore accuracy and efficiency of image searching are improved.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The drawings are included to provide a better understanding of the present solution and are not intended to limit the present application. Wherein:

fig. 1 is a flowchart of an image searching method provided in an embodiment of the present application;

fig. 2 is a schematic diagram of a terminal interface of a gallery according to an embodiment of the present application;

fig. 3 is a flowchart of an image searching method provided in an embodiment of the present application;

FIG. 4 is a schematic flow chart illustrating a process for obtaining data stored in a picture engine according to an embodiment of the present application;

fig. 5 is a flowchart of an image searching method provided in an embodiment of the present application;

fig. 6 is a schematic flowchart of an image searching method according to an embodiment of the present application;

fig. 7 is a flowchart of an image searching method provided in an embodiment of the present application;

fig. 8 is a structural diagram of an image search apparatus according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of an electronic device for implementing the picture search method according to the embodiment of the present application.

Detailed Description

The following description of the exemplary embodiments of the present application, taken in conjunction with the accompanying drawings, includes various details of the embodiments of the application for the understanding of the same, which are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In an example, fig. 1 is a flowchart of a picture search method provided in an embodiment of the present application, and the present embodiment may be applied to a case where picture search is implemented by using a full-text search technology, and the method may be performed by a picture search apparatus, which may be implemented by software and/or hardware, and may be generally integrated in an electronic device. The electronic device may be a server device capable of providing full text search engine functionality. Accordingly, as shown in fig. 1, the method comprises the following operations:

and S110, acquiring a picture searching request.

The picture search request may be a search request submitted by the gallery to a full-text search engine, and is used for requesting pictures required by the full-text search engine.

At present, a general gallery stores data information of pictures by using a relational database to provide a search function of picture data. However, when the amount of data to be stored in the gallery increases greatly, the relational database is prone to have problems of difficulty in capacity expansion, difficulty in data structure modification and the like, so that the accuracy and efficiency of picture search are reduced.

In the embodiment of the application, the gallery function can be realized by adopting technologies such as a picture file server, a full-text search engine and the like. The full-text search engine may support data storage of pictures, so as to implement a picture search operation in the gallery, and optionally, the full-text search engine may be, for example, Solr (enterprise-level search application server), an Elasticsearch (Lucene-based search server), and the like, and the engine type of the full-text search engine is not limited in the embodiment of the present application. The picture file server may store all pictures (also referred to as stored pictures) in the gallery for the searching user to search. The searching user can be a user with a picture searching requirement, and can search the required picture by using the gallery. It should be noted that all pictures stored in the picture file server have usage rights. That is, after the searching user searches for the corresponding picture by using the gallery, the searching user can have the right of using the searched picture so as to be allowed to use and spread in writing or in the network, thereby avoiding the corresponding legal problem.

Optionally, the gallery may directly send the picture search request submitted by the search user to the full-text search engine for picture search. Or, in order to improve the search efficiency of the full-text search engine, the gallery may also perform data processing, such as word segmentation processing, on the picture search request by a back-end service of the gallery after receiving the picture search request submitted by the search user, so as to obtain a new picture search request, and send the new picture search request to the full-text search engine for picture search.

Fig. 2 is a schematic view of a terminal interface of a gallery provided in an embodiment of the present application, and in a specific example, as shown in fig. 2, the gallery may be displayed to a search user in the form of a search interface at the terminal, and the search interface of the gallery includes functional modules such as "local picture", "gallery picture", "material library", and "web disk picture", it can be understood that the functional modules of the gallery may be expanded according to a business requirement, and the embodiment of the present application does not limit this. That is, the gallery can summarize and sort the pictures of different data sources, and can also open the picture search function for the user side, and the picture search requirement of the search user is supported by a large number of pictures stored by the picture file server.

Accordingly, the search user may enter a picture search request in the search interface, for example, the user may enter a picture search request for "animal images". Accordingly, the full-text search engine can directly obtain the picture search request. Or processing the picture search request of the animal image by the back-end service to obtain a new picture search request formed by the animal and sending the new picture search request to the full-text search engine.

And S120, carrying out full text search according to the picture search request to obtain target picture engine storage data.

The target picture engine storage data may be a search result obtained by the full-text search engine performing full-text search on all picture engine storage data according to the picture search request, and the search result is picture engine storage data in which the stored pictures matched with the search requirement of the search user are correspondingly stored in the full-text search engine. The picture engine storage data may be associated data that is stored in the full-text search engine for each stored picture, and the associated data may be related data describing associated information of the stored pictures. For example, the picture engine storage data may include, but is not limited to, a picture link or a picture identifier, a picture tag, a picture description, a picture shelf time, and the like, as long as the information related to the stored picture can be reflected, and the specific data content of the picture engine storage data is not limited in the embodiments of the present application.

A full-text search engine is a search engine tool for comprehensively retrieving the text contents of the pages of a website. Its advantages are automatic update of information, and full inquiry. In the embodiment of the application, the gallery stores the picture engine storage data of each storage picture in the picture file server by using the full-text search engine, so that the capacity expansion or scene modification of a large data volume can be better supported, and the processing speed of the search sequencing of the picture data is improved, thereby ensuring the accuracy and expandability of the picture data and further improving the accuracy and efficiency of picture search.

Correspondingly, after the full-text search engine acquires the picture search request, full-text search can be performed on all stored picture engine storage data according to the picture search request, so that target picture engine storage data matched with the picture search request can be obtained. Optionally, the target picture engine storage data may be one or more pieces of picture engine storage data, and the data amount of the target picture engine storage data is not limited in the embodiment of the present application.

And S130, obtaining a picture search result according to the target picture engine storage data.

The picture search result can be one or more stored pictures acquired from the picture file server according to the target picture engine storage data.

Correspondingly, after the target picture engine storage data is obtained, the full-text search engine can obtain the corresponding storage picture from the picture file server according to the target picture engine storage data obtained through searching and use the corresponding storage picture as a picture search result.

In an optional embodiment of the present application, the obtaining of the picture search result according to the target picture engine storage data may include: and downloading the corresponding target search picture from a picture file server according to each picture link in the target picture engine storage data.

The picture link may be a link for storing the picture in the picture file server. The target search picture may be a stored picture in which each picture link in the target picture engine stored data is correspondingly stored in the picture file server.

Optionally, if the target picture engine storage data includes a picture link of a stored picture in the picture file server, the full-text search engine may download corresponding target search pictures from the picture file server according to each picture link in the target picture engine storage data, and take each target search picture as a picture search result.

Optionally, the target picture engine storage data may further include a picture identifier of the picture stored in the picture file server. Correspondingly, the full-text search engine can retrieve corresponding target search pictures from the picture file server according to the picture identifications in the target picture engine storage data, and take the target search pictures as picture search results.

Correspondingly, after the full-text search engine obtains the picture search result matched with the picture search request, the obtained picture search result can be displayed to a search user, and the picture search process is completed.

According to the image searching method and device, after the full-text searching engine obtains the image searching request, full-text searching is conducted according to the obtained image searching request to obtain the target image engine storage data, the image searching result is obtained according to the target image engine storage data obtained through searching, the problems that an existing image searching method is low in accuracy and efficiency and the like are solved, and therefore accuracy and efficiency of image searching are improved.

In an example, fig. 3 is a flowchart of a picture search method provided in an embodiment of the present application, and the embodiment of the present application performs optimization and improvement on the basis of the technical solutions of the above embodiments, and provides a specific optional implementation manner for a full-text search engine to acquire picture engine storage data.

Correspondingly, the picture searching method shown in fig. 3 includes:

and S310, acquiring the picture data to be updated generated by the back-end service calculation.

And S320, storing the picture data to be updated as picture engine storage data.

The picture data to be updated is generated by the back-end service by using the picture updating data subjected to the validity check, and can be used for updating picture engine storage data in a full-text search engine. The picture update data may be data for updating pictures stored in a picture file server in the gallery and picture engine storage data of a full-text search engine. Optionally, the update operation may include, but is not limited to, adding a new picture or modifying a picture, and the like, and the embodiment of the present application does not limit the operation type of updating with the picture update data.

Fig. 4 is a schematic flowchart of a process for obtaining picture engine storage data according to an embodiment of the present disclosure, as shown in fig. 4, in the embodiment of the present disclosure, a backend service of a gallery may periodically access, through a data obtaining script (e.g., a timing script shown in fig. 4), a picture data obtaining interface provided by a third-party picture provider (e.g., provider 1, provider 2 shown in fig. 4) to periodically obtain picture update data. The data acquisition script is also a script for acquiring the image update data. The picture data acquisition interface may be an interface provided by a third party picture provider for providing pictures with access rights to the gallery. Optionally, the data acquisition period of the data acquisition script may be in units of days, which is not limited in this embodiment of the application.

Correspondingly, after the data acquisition script accesses the picture data acquisition interface to acquire the picture updating data, the validity of the picture updating data can be checked, so that the picture updating data is prevented from including unavailable updating data. Optionally, the content of the validity check may include, but is not limited to, whether the picture can be normally displayed, whether a label and description information of the picture are meaningful, whether the time for putting the picture on shelf is valid, and the like. If the picture updating data is determined to be subjected to validity check, the picture updating data is reasonable and normal data, and at the moment, the picture data to be updated can be obtained according to the picture updating data. If the picture updating data is determined not to be subjected to the validity check, the picture updating data is abnormal data, and at the moment, the abnormal picture updating data can be deleted.

After the picture update data is subjected to validity check, in order to avoid the problem of data accumulation caused by overlarge data volume or downstream service delay, the back-end service can store the normal picture update data into the message queue to be published through the message queue for asynchronous consumption. The backend service corresponding to the gallery can communicate with the message queue to update the data in the full text search engine and the picture file server.

In an optional embodiment of the present application, the picture update data may include, but is not limited to, original picture links, picture shelf times, original tag data, and original description data.

The original picture link may be a link for storing the original picture by a third-party picture provider, and the picture shelf time may be a time for publishing the original picture by the third-party picture provider. The original tag data may be tag data of the original picture provided by a third party picture provider, which may record features in the picture. The raw description data may be description data of the raw picture provided by a third party picture provider, the description data may be a sentence explaining the picture, and may generally contain the context of the picture, for example, the description data may be "girl training dog learning to hold things".

In an optional embodiment of the present application, the backend service is configured to: downloading a storage picture to be updated corresponding to the original picture according to the storage picture link; calculating the attractiveness of the picture and the pixel identification of the picture according to the stored picture to be updated; generating main label data and auxiliary label data according to the original label data; generating updated description data according to the original description data; and generating the picture data to be updated according to the picture link, the picture on-shelf time, the main label data, the auxiliary label data, the updating description data, the picture aesthetic measure and the picture pixel identification.

The stored picture link may be a link of a stored picture corresponding to the original picture stored in the picture file server. And the stored picture to be updated is also the picture of the original picture which is correspondingly stored in the picture file server. The aesthetic degree of the picture can be an aesthetic degree score of the stored picture to be updated, which is calculated by using a related aesthetic degree calculation method. The picture pixel identifier may be a pixel feature of the stored picture to be updated, such as a variance of an average value of pixel points in each row of the picture, and is used for calculating the picture similarity. The main tag data and the sub tag data may jointly constitute updated tag data as tag data stored by the full-text search engine. The update description data may be description data recalculated for a stored picture to be updated.

Correspondingly, when the back-end service finds that the image updating data in the message queue needs to be processed, the image updating data can be obtained from the message queue, so that the original image is downloaded from a third-party image provider to the local according to the original image link included in the image updating data, and then the locally stored original image is uploaded to an image file server of the intranet for storage. The image file server can preprocess the original image according to the image storage requirement of the image file server and then store the original image, generate a new image link, the width and height data of the image and the like, and obtain the stored image corresponding to the original image. The back-end service may calculate the attribute of the stored picture according to the stored picture generated in the picture file server corresponding to the original picture and the related data of the stored picture.

Optionally, a database may be used to replace the message queue to store the picture update data.

Specifically, the back-end service may download, from the picture file server, a stored picture corresponding to the original picture in the picture update data according to the stored picture link, as the stored picture to be updated. Further, the back-end service may calculate new main tag data, auxiliary tag data, update description data, and the like of the stored picture to be updated according to the calculation method of the image-text correlation. The main tag data may describe a main feature of the stored picture to be updated, such as a main feature of an animal, such as "dog", and the sub-tag data may describe an auxiliary feature of the stored picture to be updated, such as a background feature, such as "grass", "running", and "baseball". Meanwhile, the back-end service can also calculate new description data corresponding to the stored picture to be updated as the updated description data. In order to further ensure the accuracy of full-text search, the back-end service can also calculate the image attractiveness and the image pixel identification corresponding to the stored image to be updated. Correspondingly, the back-end service can generate the picture data to be updated according to the stored picture link, the picture shelf life, the main label data, the auxiliary label data, the update description data, the picture attractiveness and the picture pixel identification, and send the picture data to be updated to the full-text search engine. Accordingly, the full-text search engine may store the picture data to be updated, which is generated by the back-end service calculation, as picture engine storage data (such as the storage data shown in fig. 4). The image-text correlation calculation method and the image aesthetic degree and image pixel identification calculation method belong to image characteristic service contents. The application of the picture feature service improves AI (Artificial Intelligence) capability of the gallery.

Optionally, the stored picture link in the picture data to be updated may also be replaced by a picture identifier. For example, a back-end service or a picture file server generates a unique picture identifier for a stored picture, so that after a later full-text search engine searches, the stored picture is obtained from the picture file server according to the obtained picture identifier in the target picture engine storage data.

It should be noted that, in the prior art, the image data stored in the relational database is used to store the tag data of the image, and the tag data stored in the relational database is described by a sentence, which only supports the image search according to the completely or partially matched keywords, resulting in a lower accuracy of the image search.

And S330, acquiring a picture searching request.

And S340, carrying out full text search according to the picture search request to obtain target picture engine storage data.

And S350, obtaining a picture search result according to the target picture engine storage data, and displaying the picture search result to the search user.

By adopting the technical scheme, the picture data to be updated is generated according to the picture updating data through the back-end service of the picture library, so that the full-text search engine stores the picture data to be updated as the picture engine storage data, the accuracy of the picture engine storage data can be improved, and the accuracy of picture search is further improved.

In an example, fig. 5 is a flowchart of an image search method provided in an embodiment of the present application, and fig. 6 is a flowchart of an image search method provided in an embodiment of the present application.

In the embodiment of the present application, optionally, the full-text search engine may obtain and analyze picture search data submitted by a search user to obtain a search data analysis result, and use the search data analysis result as the picture search request. Correspondingly, the full-text search engine can perform full-text search according to the search data analysis result and the picture engine storage data for storing pictures.

The search data analysis result may be a result obtained by analyzing picture search data submitted by a search user. The image search data may be keyword text data or image data, and the like, and the data type of the image search data is not limited in the embodiment of the present application.

In the embodiment of the application, the full-text search engine can acquire and analyze the picture search data submitted by the search user to obtain the analysis result of the search data, and then the analysis result of the search data is used as the picture search request to perform full-text search on the picture engine storage data for storing pictures.

It should be noted that the operation of analyzing the picture search data submitted by the search user may be executed by a full-text search engine, or may also be executed by a backend service of a gallery, and the embodiment of the present application does not limit an execution main body of analyzing the picture search data submitted by the search user.

For example, the operation of parsing the picture search data submitted by the search user through the backend service is taken as an example. The back-end service can acquire picture search data submitted by a search user through a search interface. Correspondingly, the back-end service can analyze the picture search data according to the type of the picture search data, for example, natural language processing is performed on keyword text data, or the picture size of the picture data is analyzed, so as to obtain a search data analysis result corresponding to the picture search data, and the search data analysis result is sent to a full-text search engine as a picture search request. And after receiving a search data analysis result sent by the back-end service, the full-text search engine performs full-text search on the picture engine storage data of the stored picture by taking the search data analysis result as a picture search request.

The data of the pictures submitted by the search user are analyzed through the back-end service to obtain the picture search request, so that the process that the full-text search engine analyzes the picture search data can be avoided, and the full-text search efficiency is improved.

Correspondingly, if the full-text search engine is used for analyzing the picture search data submitted by the search user, the full-text search engine can directly obtain the picture search data submitted by the search user through the search interface, the subsequent analysis operation is carried out to obtain a search data analysis result corresponding to the picture search data, and the search data analysis result is used as a picture search request to carry out full-text search on the picture engine storage data for storing pictures.

In an optional embodiment of the present application, the picture search data may include a picture search keyword. Accordingly, an image searching method as shown in fig. 5 and 6 includes:

s510, obtaining a search keyword set obtained by natural language processing of the picture search keywords.

The search keyword set is generated by performing natural language processing such as word segmentation processing on the picture search keywords to obtain keyword word segmentation results. Optionally, the word segmentation processing may be performed on the picture search keyword by a backend service or a full-text search engine, which is not limited in the embodiment of the present application.

Wherein the picture search keyword may be keyword text data. The keyword word segmentation result can be one or more word segmentation results obtained by performing word segmentation processing on the picture search keyword.

In the embodiment of the application, optionally, the search user may submit the picture search keyword to the search interface as the picture search data. Illustratively, the picture search keyword may be, for example, "picture of dog" or "picture of training dog", or the like. Optionally, when the back-end service or the full-text search engine acquires the picture search keyword submitted by the search user, word segmentation processing may be performed on the picture search keyword to obtain a plurality of word segmentation results, that is, word segmentation results of the keyword. Correspondingly, when the word segmentation is performed through the back-end service, the back-end service can generate a search keyword set according to the keyword word segmentation result and send the search keyword set to the full-text search engine. When the word segmentation is performed through the full-text search engine, the full-text search engine can directly generate a search keyword set according to the keyword word segmentation result.

Specifically, after word segmentation processing is performed on the picture search keywords by a back-end service or a full-text search engine, word attributes of each participle can be calculated according to a part-of-speech tagging method, and meaningless pronouns, auxiliary words and the like are filtered, so that a group of search keywords is obtained to form a search keyword set.

In an optional embodiment of the present application, in the obtaining and analyzing the picture search data submitted by the search user to obtain a search data analysis result, the method may further include: acquiring a legality check word list through a back-end service or a full-text search engine; and matching and checking the image search keywords and the legality checking word list through a back-end service or a full-text search engine.

The legality check word list can be used for storing illegal words, such as sensitive words or forbidden words, and the specific word content stored in the legality check word list is not limited in the embodiment of the application.

In the embodiment of the application, after the picture search keyword is obtained, the backend service or the full-text search engine can obtain the legality check word list so as to perform matching check on the picture search keyword according to the legality check word list and judge whether the picture search keyword comprises illegal words.

Exemplarily, the validity check of the image search keyword by the backend service is described as follows: if the back-end service determines that the picture search keyword passes the matching check, which indicates that the picture search keyword does not include illegal words, the picture search data can be further analyzed to obtain a search data analysis result, so that a full-text search engine can quickly perform full-text search according to the search data analysis result, and the full-text search efficiency is improved. If the back-end service determines that the picture search keyword does not pass the matching check, which indicates that the picture search keyword comprises illegal words, the picture search data can be refused to be analyzed, and correspondingly, the full-text search engine cannot acquire the picture search request. The mode of carrying out validity check on the picture search keyword through the back-end service can reduce the data processing load of the full-text search engine, so that the full-text search engine can intensively process the search function, and the search efficiency and the search quality of the full-text search engine are further improved.

Correspondingly, if the picture search keyword is subjected to legality verification through the full-text search engine, the full-text search engine can directly execute subsequent search operation according to the legality verification result.

It should be noted that, in order to reduce the workload of the full-text search engine and improve the search efficiency and the search quality of the full-text search engine, the parsing of the picture search data submitted by the search user is preferably performed by the back-end service, so that the picture search request which can be directly processed without further parsing is generated by the back-end service with full-text search efficiency.

In an optional embodiment of the present application, the picture engine storage data comprises tag data and description data; the tag data includes primary tag data and secondary tag data; obtaining a search keyword set obtained by performing natural language processing on the picture search keyword may include: acquiring a label keyword set obtained by performing word segmentation processing on the label data according to the separators; and acquiring a description keyword set obtained by performing word segmentation processing on the description data according to a unit semantic word segmentation strategy.

The separator may be configured to separate each feature word in the tag data. The tag keyword set may be a keyword set obtained by dividing tag data by a delimiter. The unit semantic word segmentation strategy can adopt a minimum semantic word segmentation mode. The description keyword set may be a keyword set obtained by performing word segmentation processing on the description data in a minimum semantic word segmentation mode.

In the embodiment of the application, the search result can be obtained according to the search priority. Correspondingly, in order to support the search priority, the full-text search engine or the back-end service can perform word segmentation processing on the tag data and the description data in the picture engine storage data respectively, so that the full-text search engine can acquire different types of keyword sets according to the picture engine storage data. Specifically, the full-text search engine may obtain a set of tag keywords and a set of description keywords. The label keyword set is obtained by performing word segmentation processing on the main label data and the auxiliary label data according to the separators, and the description keyword set is obtained by performing word segmentation processing on the description data according to a unit semantic word segmentation strategy. For example, the set of label keywords may be "dog, grass, running, baseball" and the set of description keywords may be "girls training dogs to learn to hold things".

It should be noted that the above-mentioned operations of performing word segmentation on the tag data and the description data may be performed after the data stored in the picture engine is obtained, or may be performed in real time after the picture search request is obtained.

According to the technical scheme, the label data and the description data in the picture engine storage data are subjected to word segmentation processing and can be used as matching data of the search keyword set, so that the data matching efficiency of the full-text search engine is improved.

S520, full-text search is carried out according to the search keyword set and the picture engine storage data, and a full-text search result is obtained.

The full-text search result can be a result obtained by full-text search engine performing full-text search according to the search data analysis result and the picture engine storage data.

Optionally, the full-text search engine may perform full-text search according to the search keyword set and the picture engine storage data, to obtain a full-text search result in a form matched with the picture search keyword.

In an optional embodiment of the present application, the performing a full-text search according to the search keyword set and the picture engine storage data may include: matching the label keyword set according to the search keyword set to obtain first search picture engine storage data; matching the description keyword set according to the search keyword set to obtain second search picture engine storage data; and sequencing according to the first search picture engine storage data and the second search picture engine storage data according to search priority to obtain the full text search result.

The first search picture engine storage data may be picture engine storage data corresponding to target tag data obtained by matching the search keyword set with the tag keyword set. The second search picture engine storage data may be picture engine storage data corresponding to target description data obtained by matching the search keyword set with the description keyword set. The search priority may be a precedence priority determined based on the tag data and the description data.

Specifically, when the full-text search engine performs full-text search, the tag keyword set and the description keyword set corresponding to the picture engine storage data can be respectively matched according to the search keyword set, so that the first search picture engine storage data and the second search picture engine storage data obtained by matching are sorted according to the search priority, and a final full-text search result is obtained. Alternatively, the priority of the search result of the tag data may be set higher than the priority of the search result of the description data. That is, the priority of the first search picture engine to store data may be higher than the priority of the second search picture engine to store data.

In the related art, only the tag data of a picture can be searched by searching for the picture using a relational database. Because the label data of the pictures stored in the existing relational database is in a sentence form, the form of the attribute data of the pictures is simple, and the complete attributes of the pictures cannot be embodied, only picture searching is supported according to completely or partially matched keywords, and the accuracy rate of the picture searching is low. Further, the prior art approach of using a complete match of the relational database has a problem of obtaining a small number of pictures, because the picture tags are required to consistently contain the input keywords. The partial match further reduces the accuracy of the picture search because the search results can also be hit by tags that may not be context-specific.

In the full-text searching mode based on the tag data and the description data, the stored pictures are searched according to the priorities of different attributes of the pictures. Because the label data stores the characteristics of the picture and describes the data stores the context background of the picture, the stored picture is searched according to the priority of different attributes of the picture, the main characteristics of the stored picture can be covered to the context background, the search requirement of a search user can be more comprehensively understood, and the search accuracy is improved.

In an optional embodiment of the present application, the set of tag keywords comprises a set of primary tag keywords and a set of secondary tag keywords; the sorting according to the first search picture engine storage data and the second search picture engine storage data according to search priority may include: acquiring first main search picture engine storage data corresponding to the main label keyword set and first auxiliary search picture engine storage data corresponding to the auxiliary label keyword set; and sequencing according to the sequence of the first main search picture engine storage data, the first auxiliary search picture engine storage data and the second search picture engine storage data.

The main tag keyword set may be a keyword set obtained by performing word segmentation on the main tag data, and the auxiliary tag keyword set may be a keyword set obtained by performing word segmentation on the auxiliary tag data. The first main search picture engine storage data may be part of the first search picture engine storage data obtained by matching the main tag keyword set according to the search keyword set. The second main search picture engine storage data may be part of the first search picture engine storage data obtained by matching the secondary tag keyword set according to the search keyword set.

Because the picture engine storage data comprises main label data and auxiliary label data, when the label keyword set is determined according to the label data, the main label keyword set can be determined according to the main label data, and the auxiliary label keyword set can be determined according to the auxiliary label data. Correspondingly, the full-text search engine can match the main label keyword set according to the search keyword set to obtain first main search picture engine storage data, and match the auxiliary label keyword set according to the search keyword set to obtain first auxiliary search picture engine storage data. The search priority may specifically be first main search picture engine storage data-first sub search picture engine storage data-second search picture engine storage data.

That is, the full-text search engine may search the main tag data according to the search keyword set, and when the main tag data is completely matched with the search keyword set, obtain the first main search picture engine storage data corresponding to the main tag data. Furthermore, the full-text search engine can search the sub tag data according to the search keyword set, and when the sub tag data is completely matched with the search keyword set, the first sub search picture engine storage data corresponding to the sub tag data is obtained. Finally, the full-text search engine can search the description data according to the search keyword set, and when the description data is completely matched with the search keyword set, the first main search picture engine storage data corresponding to the description data is obtained. In the embodiment of the present application, the perfect match can be understood as: the tag data or description data can be matched with the set proportion of keywords in the search keyword set. The set proportion may be, for example, 90% or 95%, and the like, and may be specifically set according to actual requirements, and in the embodiment of the present application, a specific numerical value of the set proportion is not limited, but a value of the set proportion needs to be close to 100%, so as to avoid reducing the search accuracy.

It will be appreciated that when a certain type of picture engine storage data is empty, that type of picture engine storage data may be omitted. For example, if the first primary search picture engine storage data is empty, the full-text search results may include only the first secondary search picture engine storage data and the second search picture engine storage data.

According to the technical scheme, the search priority is determined by utilizing the main tag keyword set, the auxiliary tag keyword set and the description keyword set, the search sorting rule according to the main features, the auxiliary features and the context background of the picture is realized, the search requirements of a search user can be met, and the search accuracy is improved.

S530, determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy.

The full-text search ranking policy may be a policy for ranking full-text search results.

In the embodiment of the application, after the full-text search engine obtains the full-text search result by using the search keyword set, the picture engine storage data of the full-text search result can be reordered according to the full-text search ordering strategy, so that the final target picture engine storage data is determined.

In an optional embodiment of the present application, the picture search request may further include a picture search tag; the determining the target picture engine storage data according to the full-text search result and the full-text search sorting policy may include: determining a tag type of the picture search tag; the tag types comprise a high-quality picture tag, a latest picture tag and a low-frequency picture tag; calculating the sorting data of the data stored in each search picture engine in the full-text search result according to the tag type; and sequencing the stored data of each search picture engine according to the sequencing data to obtain the stored data of the target picture engine.

The picture search tag may be a search tag selected by a search user in a search interface. The premium picture tab may be used to provide a premium picture to a searching user. The latest picture tag may provide a picture that is most recent in time for the searching user. The low frequency picture tag may provide a search user with a picture that is used less frequently. In an embodiment of the application, the search picture engine storage data may be a first main search picture engine storage data, a first sub search picture engine storage data, and a second search picture engine storage data.

It will be appreciated that different search users have different search requirements. Different search users input the same picture search data and have different ordering requirements, for example, a search user who is worried about that an article is heavily judged wants to add pictures which are used less frequently, a current search user wants to use the latest news pictures, and a search user who seeks the quality of the article wants to obtain more beautiful and better pictures. Therefore, the gallery can also set picture search tags for selection by the searching user in consideration of the search requirements of different searching users.

Therefore, the search user can select the picture search tag in addition to submitting the picture search data through the search interface, that is, the picture search request can include both the picture search data and the picture search tag. The picture search data is used for obtaining searched pictures, and the picture search tags can be used for sequencing the searched pictures. In the embodiment of the present application, the tag types of the picture search tag may include, but are not limited to, a premium picture tag, a latest picture tag, a low-frequency picture tag, and the like. Illustratively, the premium picture tab may be, for example, the "premium recommendation" option in fig. 1, the latest picture tab may be, for example, the "latest info" option in fig. 1, and the low-frequency picture tab may be, for example, the "cold goodness" option in fig. 1.

Specifically, after the full-text search engine obtains the full-text search result by using the search keyword set, the ranking data of the stored data of each search picture engine in the full-text search result can be calculated according to the tag type of the picture search tag selected by the search user, so that all the stored data of the search picture engines can be reordered according to the ranking data. Optionally, the full-text search engine may sort the data stored in the search picture engines according to the order from high to low of the value of the aesthetic measure of the picture according to the high-quality picture tags, may also sort the data stored in the search picture engines according to the order from near to far of the picture listing time according to the latest picture tags, and may also sort the data stored in the search picture engines according to the order from low to high of the value of the picture use frequency according to the low-frequency picture tags. Alternatively, the picture use frequency may be determined according to the download frequency of the stored pictures.

It should be noted that, when all the search picture engine storage data are reordered according to the ordering data, the search picture engine storage data can be respectively ordered according to the search priorities in the full-text search results, that is, the search priorities of the full-text search results are not disturbed. For example, when the search user selects the low-frequency picture tag, the full-text search engine may sort the first main search picture engine storage data in the order of the picture use frequency from low to high, to obtain first sorted data. And sorting the stored data of the first auxiliary search picture engine according to the sequence of the picture use frequency from low to high to obtain second sorted data. And sorting the stored data of the second search picture engine according to the sequence of the picture use frequency from low to high to obtain third sorted data. And finally, respectively sequencing the search picture engine storage data according to the first sequencing data, the second sequencing data and the third sequencing data to obtain final target picture engine storage data.

According to the technical scheme, the full-text search results are sequenced by using the full-text search sequencing strategy to obtain the target picture engine storage data, so that a multi-dimensional sequencing rule is supported, and the user experience can be better improved.

And S540, obtaining a picture search result according to the target picture engine storage data.

And S550, displaying the picture searching result to a searching user.

In an optional embodiment of the present application, the presenting the picture search result to the search user may include: adding target watermarks to all target search pictures in the picture search results; and displaying each target search picture with the target watermark to the search user.

The target watermark may be a watermark type and content determined according to service requirements.

In the embodiment of the application, in order to realize copyright protection of the stored pictures, before providing each target search picture for a search user, a target watermark may be added to each target search picture, and each target search picture with the target watermark is displayed to the search user. It should be noted that the operation of adding the target watermark and the operation of displaying the target search picture may be performed by a full-text search engine or a backend service, which is not limited in this embodiment of the present application.

According to the technical scheme, the picture search mode is realized according to the keyword text data by utilizing the picture search keywords and the full-text search engine to perform full-text search, the full-text search results are sorted according to the search priority in the full-text search process, and the full-text search results are sorted again by utilizing the full-text search sorting strategy, so that a multi-dimensional sorting rule is realized, the search accuracy and the search efficiency of the picture search can be improved, and the user experience can be improved at the same time.

In an example, fig. 7 is a flowchart of an image search method provided in an embodiment of the present application, and the embodiment of the present application performs optimization and improvement on the basis of the technical solutions of the foregoing embodiments, and provides a specific optional implementation manner for performing image search according to a reference search image.

In an optional embodiment of the present application, the picture search data may include a reference search picture. Accordingly, an image searching method as shown in fig. 6 and 7 includes:

and S710, acquiring a reference search picture obtained by analyzing the reference search picture.

And the reference search picture is generated by analyzing the picture size data of the reference search picture. Optionally, the image size data of the reference search image may be analyzed by a back-end service or a full-text search engine, which is not limited in this application.

The reference search picture may be a picture provided by a search user through a search interface of a gallery, the picture may be a local picture or an online picture in internet resources, and as long as the search user has a picture usage right, the picture may be used as the reference search picture, and the embodiment of the application does not limit the manner of obtaining the reference search picture.

In this embodiment of the present application, optionally, the search user may submit the reference search picture to the search interface as the picture search data. For example, the search user may select one of the locally stored pictures as a reference search picture. Optionally, when the back-end service or the full-text search engine acquires the reference search picture submitted by the search user, the reference search picture may be analyzed according to the picture size data of the reference search picture to obtain the reference search picture. Correspondingly, when the reference search picture is analyzed through the back-end service, the back-end service can send the reference search picture obtained through analysis to the full-text search engine. When the reference search picture is analyzed by the full-text search engine, the full-text search engine can directly acquire the reference search picture.

It should be noted that, in order to reduce the workload of the full-text search engine and improve the search efficiency and the search quality of the full-text search engine, the above-mentioned parsing of the reference search picture submitted by the search user is preferably executed by the backend service, so that the picture search request which can be directly processed without further parsing by the full-text search efficiency is generated by the backend service.

In an optional embodiment of the present application, the obtaining of the reference search picture obtained by analyzing the reference search picture may include: under the condition that the picture size data is determined to be larger than or equal to a preset size threshold, acquiring a target picture downloaded from a picture file server according to a picture search link of the reference search picture; taking the target picture as the reference search picture; and taking the reference search picture as the reference search picture when the picture size data is determined to be smaller than the preset size threshold.

The preset size threshold may be a threshold determined according to the processing performance of the gallery on the picture, and the embodiment of the present application does not limit a specific size value of the preset size threshold. The target picture may be a picture in which the reference search picture is correspondingly stored in the picture file server.

In the embodiment of the present application, in order to avoid that an oversized picture affects the calculation performance or the search performance of the gallery, after the reference search picture is acquired, it may be first determined whether picture size data of the reference search picture is greater than or equal to a preset size threshold. If the picture size data is determined to be larger than or equal to the preset size threshold, which indicates that the size of the reference search picture is too large, the reference search picture can be transferred to a picture file server at the moment so as to reduce the picture size data of the reference search picture and obtain a corresponding target picture. Accordingly, the full-text search engine may acquire the target picture downloaded from the picture file server according to the picture search link of the reference search picture as the reference search picture. If the picture size data is determined to be smaller than the preset size threshold, the size of the reference search picture is not larger than the processing requirement of the gallery, and the reference search picture can be directly used as the reference search picture at the moment. Alternatively, the reference search picture may be acquired by a full-text search engine or a backend service. For example, a full-text search engine or a back-end service acquires a picture search link according to a reference search picture, and downloads a target picture from a picture file server according to the picture search link. If the target picture is downloaded through the back-end service, the back-end service needs to send the downloaded target picture to a full-text search engine. Or, the full text search engine or the backend service directly takes the reference search picture as the reference search picture. If the reference search picture is directly used as the reference search picture through the back-end service, the back-end service needs to send the reference search picture to the full-text search engine.

S720, full-text search is carried out according to the reference search picture and the picture engine storage data, and a full-text search result is obtained.

Optionally, the full-text search engine may perform full-text search according to the reference search picture and the picture engine storage data to obtain another form of full-text search result matched with the reference search picture.

In an optional embodiment of the present application, the picture engine storage data comprises picture pixel identifications; the performing full-text search according to the reference search picture and the picture engine storage data may include: acquiring a reference picture pixel identifier of the reference search picture; and matching the picture pixel identification of the data stored in each picture engine according to the reference picture pixel identification.

The reference picture pixel identifier may be a picture pixel identifier corresponding to a reference search picture, and the third search picture engine storage data may be picture engine storage data corresponding to a target picture pixel identifier obtained by matching picture pixel identifiers of the picture engine storage data according to the reference picture pixel identifier.

In the embodiment of the present application, the reference picture pixel identifier of the reference search picture may be calculated by a full-text search engine or a backend service. Correspondingly, if the reference picture pixel identifier of the reference search picture is calculated by the back-end service, the back-end service needs to send the reference picture pixel identifier to the full-text search engine. If the reference picture pixel identification of the reference search picture is calculated by the full-text search engine, the full-text search engine can directly acquire the reference picture pixel identification. Correspondingly, after the full-text search engine acquires the reference picture pixel identifier, the reference picture pixel identifier can be matched with the picture pixel identifiers included in the picture engine storage data by using an image similarity calculation method, so that third search picture engine storage data can be obtained. The process of calculating the picture pixel identifier by using the image similarity calculation method belongs to the process of picture characteristic service. The application of the picture feature service improves the AI capability of the gallery.

According to the technical scheme, the full-text search engine utilizes the image pixel identification to realize the image search method based on the image by utilizing the image similarity calculation method, so that the image search function is enriched, the search requirement of a search user is met, and the user experience is improved.

It should be noted that the above operations that can be performed by either the full-text search engine or the backend service can be preferably performed by the backend service to reduce the workload of the full-text search engine.

And S730, determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy.

In the embodiment of the application, after the full-text search engine obtains the full-text search result by using the reference search picture, the picture engine storage data of the full-text search result can be reordered according to the full-text search ordering strategy, so that the final target picture engine storage data is determined.

The search user can submit the picture search data through the search interface and select the picture search tag, that is, the picture search request can include the picture search data and the picture search tag at the same time. The picture search data is used for obtaining searched pictures, and the picture search tags can be used for sequencing the searched pictures. In the embodiment of the present application, the tag types of the picture search tag may include, but are not limited to, a premium picture tag, a latest picture tag, a low-frequency picture tag, and the like. Illustratively, the premium picture tab may be, for example, the "premium recommendation" option in fig. 1, the latest picture tab may be, for example, the "latest info" option in fig. 1, and the low-frequency picture tab may be, for example, the "cold goodness" option in fig. 1.

Specifically, after the full-text search engine obtains the full-text search result by using the reference search picture, the ranking data of the stored data of each third search picture engine in the full-text search result can be calculated according to the tag type of the picture search tag selected by the search user, so that all the stored data of the third search picture engines are reordered according to the ranking data. Optionally, the full-text search engine may sort the stored data of the third search picture engines according to the order from high to low of the value of the aesthetic measure of the picture according to the high-quality picture tags, may also sort the stored data of the third search picture engines according to the order from near to far of the picture shelf time according to the latest picture tags, and may also sort the stored data of the third search picture engines according to the order from low to high of the value of the picture use frequency according to the low-frequency picture tags. Alternatively, the picture use frequency may be determined according to the download frequency of the stored pictures.

And S740, obtaining picture search results according to the target picture engine storage data.

And S750, displaying the picture searching result to a searching user.

According to the technical scheme, full-text search is carried out by utilizing the reference search picture and the full-text search engine, a picture search mode according to pictures is realized, full-text search results are sorted by utilizing a full-text search sorting strategy in the full-text search process, picture search functions are enriched, the search accuracy and the search efficiency of picture search can be improved, and user experience can be improved simultaneously.

In an example, fig. 8 is a structural diagram of a picture search apparatus provided in an embodiment of the present application, where the embodiment of the present application is applicable to a case where picture search is implemented by using full-text search technology, and the apparatus is implemented by software and/or hardware and is specifically configured in a full-text search engine.

An image search apparatus 800 as shown in fig. 8 comprises: an image search request acquisition module 810, a full-text search module 820, and an image search result acquisition module 830. Wherein the content of the first and second substances,

an image search request obtaining module 810, configured to obtain an image search request;

a full text search module 820, configured to perform full text search according to the picture search request to obtain target picture engine storage data;

and the picture search result acquisition module 830 is configured to acquire a picture search result according to the target picture engine storage data.

Optionally, the image search request obtaining module 810 is specifically configured to: acquiring and analyzing picture search data submitted by a search user to obtain a search data analysis result; taking the analysis result of the search data as the picture search request; the full-text search module 820 is specifically configured to: and carrying out full-text search according to the search data analysis result and the picture engine storage data for storing pictures.

Optionally, the picture search data includes a picture search keyword; the image search request obtaining module 810 is specifically configured to: acquiring a search keyword set obtained by performing natural language processing on the picture search keywords; the search keyword set is generated by keyword word segmentation results obtained by performing word segmentation on the picture search keywords; the full-text search module 820 is specifically configured to: performing full-text search according to the search keyword set and the picture engine storage data to obtain a full-text search result; and determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy.

Optionally, the picture engine storage data includes tag data and description data; the tag data includes primary tag data and secondary tag data; the image search request obtaining module 810 is specifically configured to: acquiring a label keyword set obtained by performing word segmentation processing on the label data according to the separators; and acquiring a description keyword set obtained by performing word segmentation processing on the description data according to a unit semantic word segmentation strategy.

Optionally, the full-text search module 820 is specifically configured to: matching the label keyword set according to the search keyword set to obtain first search picture engine storage data; matching the description keyword set according to the search keyword set to obtain second search picture engine storage data; and sequencing according to the first search picture engine storage data and the second search picture engine storage data according to search priority to obtain the full text search result.

Optionally, the tag keyword set includes a main tag keyword set and an auxiliary tag keyword set; the full-text search module 820 is specifically configured to: acquiring first main search picture engine storage data corresponding to the main label keyword set and first auxiliary search picture engine storage data corresponding to the auxiliary label keyword set; and sequencing according to the sequence of the first main search picture engine storage data, the first auxiliary search picture engine storage data and the second search picture engine storage data.

Optionally, the picture search data includes a reference search picture; the image search request obtaining module 810 is specifically configured to: acquiring a reference search picture obtained by analyzing the reference search picture; the reference search picture is generated by analyzing picture size data of the reference search picture; the full-text search module 820 is specifically configured to: performing full-text search according to the reference search picture and the picture engine storage data to obtain a full-text search result; and determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy.

Optionally, the image search request obtaining module 810 is specifically configured to: under the condition that the picture size data is determined to be larger than or equal to a preset size threshold, acquiring a target picture downloaded from a picture file server according to a picture search link of the reference search picture; taking the target picture as the reference search picture; and taking the reference search picture as the reference search picture when the picture size data is determined to be smaller than the preset size threshold.

Optionally, the picture engine storage data includes picture pixel identifiers; the full-text search module 820 is specifically configured to: acquiring a reference picture pixel identifier of the reference search picture; and matching the picture pixel identification of the stored data of each picture engine according to the reference picture pixel identification to obtain stored data of a third search picture engine.

Optionally, the picture search request further includes a picture search tag; the full-text search module 820 is specifically configured to: determining a tag type of the picture search tag; the tag types comprise a high-quality picture tag, a latest picture tag and a low-frequency picture tag; calculating the sorting data of the data stored in each search picture engine in the full-text search result according to the tag type; and sequencing the stored data of each search picture engine according to the sequencing data to obtain the stored data of the target picture engine.

Optionally, the image search result obtaining module 810 is specifically configured to: downloading corresponding target search pictures from a picture file server according to each picture link in the target picture engine storage data; the device also comprises a target search picture display module, a target watermark adding module and a target watermark adding module, wherein the target search picture display module is used for adding a target watermark to each target search picture in the picture search result; and displaying each target search picture with the target watermark to the search user.

Optionally, the device further includes a to-be-updated picture data obtaining module, configured to obtain to-be-updated picture data generated by back-end service calculation; the picture engine storage data storage module is used for storing the picture data to be updated as picture engine storage data; and the picture data to be updated is generated by the back-end service by using the picture updating data subjected to the validity check.

Optionally, the picture update data includes an original picture link, a picture shelf time, original tag data, and original description data; the backend service is to: downloading a storage picture to be updated corresponding to the original picture according to the storage picture link; calculating the attractiveness of the picture and the pixel identification of the picture according to the stored picture to be updated; generating main label data and auxiliary label data according to the original label data; generating updated description data according to the original description data; and generating the picture data to be updated according to the stored picture link, the picture on-shelf time, the main label data, the auxiliary label data, the updating description data, the picture aesthetic measure and the picture pixel identification.

The image searching device can execute the image searching method provided by any embodiment of the application, and has the corresponding functional modules and beneficial effects of the executing method. For details of the technology that are not described in detail in this embodiment, reference may be made to the picture search method provided in any embodiment of the present application.

Since the above-described image search apparatus is an apparatus capable of executing the image search method in the embodiment of the present application, based on the image search method described in the embodiment of the present application, a person skilled in the art can understand a specific implementation manner of the image search apparatus in the embodiment of the present application and various variations thereof, and therefore, a detailed description of how the image search apparatus implements the image search method in the embodiment of the present application is omitted here. The device used by those skilled in the art to implement the picture searching method in the embodiments of the present application is within the scope of the present application.

In one example, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

FIG. 9 illustrates a schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 9, the apparatus 900 includes a computing unit 901, which can perform various appropriate actions and processes in accordance with a computer program stored in a Read Only Memory (ROM)902 or a computer program loaded from a storage unit 908 into a Random Access Memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The calculation unit 901, ROM 902, and RAM 903 are connected to each other via a bus 904. An input/output (I/O) interface 905 is also connected to bus 904.

A number of components in the device 900 are connected to the I/O interface 905, including: an input unit 906 such as a keyboard, a mouse, and the like; an output unit 907 such as various types of displays, speakers, and the like; a storage unit 908 such as a magnetic disk, optical disk, or the like; and a communication unit 909 such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 909 allows the device 900 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

The computing unit 901 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 901 performs the respective methods and processes described above, such as the picture search method. For example, in some embodiments, the picture search method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 900 via ROM 902 and/or communications unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the picture search method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the picture search method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), blockchain networks, and the internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A picture search method is applied to a full-text search engine and comprises the following steps:

acquiring a picture search request;

2. The method of claim 1, wherein the obtaining the picture search request comprises:

acquiring and analyzing picture search data submitted by a search user to obtain a search data analysis result;

taking the analysis result of the search data as the picture search request;

the full text search according to the picture search request comprises:

and carrying out full-text search according to the search data analysis result and the picture engine storage data for storing pictures.

3. The method of claim 2, wherein the picture search data comprises picture search keywords;

the acquiring and analyzing the picture search data submitted by the search user to obtain a search data analysis result includes:

acquiring a search keyword set obtained by performing natural language processing on the picture search keywords; the search keyword set is generated by keyword word segmentation results obtained by performing word segmentation on the picture search keywords;

the full-text search is carried out according to the search data analysis result and the picture engine storage data for storing pictures, and the full-text search method comprises the following steps:

performing full-text search according to the search keyword set and the picture engine storage data to obtain a full-text search result;

and determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy.

4. The method of claim 3, the picture engine storage data comprising tag data and description data; the tag data includes primary tag data and secondary tag data; the acquiring of the search keyword set obtained by natural language processing of the picture search keywords comprises:

acquiring a label keyword set obtained by performing word segmentation processing on the label data according to the separators;

and acquiring a description keyword set obtained by performing word segmentation processing on the description data according to a unit semantic word segmentation strategy.

5. The method of claim 4, wherein the conducting a full-text search according to the set of search keywords and the picture engine storage data comprises:

matching the label keyword set according to the search keyword set to obtain first search picture engine storage data;

matching the description keyword set according to the search keyword set to obtain second search picture engine storage data;

and sequencing according to the first search picture engine storage data and the second search picture engine storage data according to search priority to obtain the full text search result.

6. The method of claim 5, wherein the set of tag keywords comprises a set of primary tag keywords and a set of secondary tag keywords;

the sorting according to the first search picture engine storage data and the second search picture engine storage data according to search priority comprises:

acquiring first main search picture engine storage data corresponding to the main label keyword set and first auxiliary search picture engine storage data corresponding to the auxiliary label keyword set;

and sequencing according to the sequence of the first main search picture engine storage data, the first auxiliary search picture engine storage data and the second search picture engine storage data.

7. The method of claim 2, wherein the picture search data comprises a reference search picture;

acquiring a reference search picture obtained by analyzing the reference search picture; the reference search picture is generated by analyzing picture size data of the reference search picture;

performing full-text search according to the reference search picture and the picture engine storage data to obtain a full-text search result;

8. The method according to claim 7, wherein the obtaining of the reference search picture obtained by parsing the reference search picture comprises:

under the condition that the picture size data is determined to be larger than or equal to a preset size threshold, acquiring a target picture downloaded from a picture file server according to a picture search link of the reference search picture;

taking the target picture as the reference search picture;

and taking the reference search picture as the reference search picture when the picture size data is determined to be smaller than the preset size threshold.

9. The method of claim 7, wherein the picture engine storage data comprises picture pixel identification;

the full-text search according to the reference search picture and the picture engine storage data comprises:

acquiring a reference picture pixel identifier of the reference search picture;

and matching the picture pixel identification of the stored data of each picture engine according to the reference picture pixel identification to obtain stored data of a third search picture engine.

10. The method of claim 3 or 7, wherein the picture search request further comprises a picture search tag;

the determining the target picture engine storage data according to the full-text search result and the full-text search sorting strategy comprises:

determining a tag type of the picture search tag; the tag types comprise a high-quality picture tag, a latest picture tag and a low-frequency picture tag;

calculating the sorting data of the data stored in each search picture engine in the full-text search result according to the tag type;

and sequencing the stored data of each search picture engine according to the sequencing data to obtain the stored data of the target picture engine.

11. The method of claim 1, wherein said retrieving picture search results from said target picture engine storage data comprises:

downloading corresponding target search pictures from a picture file server according to each picture link in the target picture engine storage data;

the method further comprises the following steps:

adding a target watermark to each target search picture in the picture search result;

and displaying each target search picture with the target watermark to the search user.

12. The method of claim 1, further comprising:

acquiring picture data to be updated generated by back-end service calculation;

storing the picture data to be updated as picture engine storage data;

and the picture data to be updated is generated by the back-end service by using the picture updating data subjected to the validity check.

13. The method of claim 12, wherein the picture update data comprises an original picture link, a picture on-shelf time, original tag data, and original description data;

the backend service is to:

downloading a storage picture to be updated corresponding to the original picture according to the storage picture link;

calculating the attractiveness of the picture and the pixel identification of the picture according to the stored picture to be updated;

generating main label data and auxiliary label data according to the original label data;

generating updated description data according to the original description data;

and generating the picture data to be updated according to the stored picture link, the picture on-shelf time, the main label data, the auxiliary label data, the updating description data, the picture aesthetic measure and the picture pixel identification.

14. An image search device configured in a full text search engine, comprising:

the full text search module is used for performing full text search according to the picture search request to obtain target picture engine storage data;

and the picture search result acquisition module is used for acquiring a picture search result according to the target picture engine storage data.

15. The apparatus according to claim 14, wherein the picture search request acquisition module is specifically configured to:

taking the analysis result of the search data as the picture search request;

the full-text search module is specifically configured to:

16. The apparatus of claim 15, wherein the picture search data comprises picture search keywords;

the image search request acquisition module is specifically configured to: acquiring a search keyword set obtained by performing natural language processing on the picture search keywords; the search keyword set is generated by keyword word segmentation results obtained by performing word segmentation on the picture search keywords;

the full-text search module is specifically configured to: performing full-text search according to the search keyword set and the picture engine storage data to obtain a full-text search result;

17. The apparatus of claim 16, wherein the picture engine storage data comprises tag data and description data; the tag data includes primary tag data and secondary tag data; the image search request acquisition module is specifically configured to:

18. The apparatus of claim 17, wherein the full text search module is specifically configured to:

19. The apparatus of claim 18, wherein the set of tag keywords comprises a set of primary tag keywords and a set of secondary tag keywords; the full-text search module is specifically configured to:

20. The apparatus of claim 15, wherein the picture search data comprises a reference search picture;

the image search request acquisition module is specifically configured to: acquiring a reference search picture obtained by analyzing the reference search picture; the reference search picture is generated by analyzing picture size data of the reference search picture;

the full-text search module is specifically configured to: performing full-text search according to the reference search picture and the picture engine storage data to obtain a full-text search result;

21. The apparatus according to claim 20, wherein the picture search request obtaining module is specifically configured to:

taking the target picture as the reference search picture;

22. The apparatus of claim 20, wherein the picture engine storage data comprises picture pixel identification; the full-text search module is specifically configured to:

acquiring a reference picture pixel identifier of the reference search picture;

23. The apparatus of claim 16 or 20, wherein the picture search request further comprises a picture search tag; the full-text search module is specifically configured to:

24. The apparatus according to claim 14, wherein the picture search result acquisition module is specifically configured to:

the device also comprises a target search picture display module, a target watermark adding module and a target watermark adding module, wherein the target search picture display module is used for adding a target watermark to each target search picture in the picture search result;

25. The apparatus of claim 15, further comprising:

the image data to be updated acquiring module is used for acquiring image data to be updated generated by back-end service calculation;

the picture engine storage data storage module is used for storing the picture data to be updated as picture engine storage data;

26. The apparatus of claim 25, wherein the picture update data comprises an original picture link, a picture on shelf time, original tag data, and original description data;

the backend service is to:

generating updated description data according to the original description data;

27. An electronic device, comprising:

at least one processor; and

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the picture search method of any one of claims 1-13.

28. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the picture search method of any one of claims 1 to 13.

29. A computer program product comprising a computer program which, when executed by a processor, implements a picture search method according to any one of claims 1-13.