CN113761256A

CN113761256A - Method, device, server and storage medium for determining article information

Info

Publication number: CN113761256A
Application number: CN202010769207.0A
Authority: CN
Inventors: 何云飞; 沙金; 白石
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Priority date: 2020-08-03
Filing date: 2020-08-03
Publication date: 2021-12-07

Abstract

The invention discloses a method, a device, a server and a storage medium for determining article information, wherein the method comprises the following steps: acquiring a target image corresponding to a target article, and extracting target text content and target image characteristics in the target image; determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image, and determining second item information of the target item based on the target similar image; and determining target item information of the target item based on the first item information and the second item information. According to the technical scheme of the embodiment of the invention, the text content of the target object and the processing result of the target image characteristic are comprehensively considered, so that the target object information corresponding to the target object can be rapidly, accurately and conveniently determined.

Description

Method, device, server and storage medium for determining article information

Technical Field

The embodiment of the invention relates to the technical field of computers, in particular to a method, a device, a server and a storage medium for determining article information.

Background

At present, when article information (such as article type) of a certain article needs to be determined, a target image of the article is obtained by means of shooting the article and the like, then an article image which is the same as or similar to the target image is determined by a picture search technology, and then the article information of the article is determined based on the same or similar article image.

In the process of implementing the invention, the inventor finds that the following problems exist in the prior art:

most of image searching technologies are searching for images similar to a target image, so that certain differences exist between the item information determined based on the similar images and the item information corresponding to the target image, that is, the image searching precision is low, and the accuracy of the determined item information is low.

Further, the image search technology also depends on the richness of the images in the gallery, and if the number or content of the images in the gallery is small, the target image cannot be matched, so that the article information corresponding to the target image cannot be determined.

Disclosure of Invention

The invention provides a method, a device, a server and a storage medium for determining article information, which are used for realizing the technical effects of optimizing the method for determining the article information and further efficiently and accurately determining the article information.

In a first aspect, an embodiment of the present invention provides a method for determining item information, where the method includes:

acquiring a target image corresponding to a target article, and extracting target text content and target image characteristics in the target image;

determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image, and determining second item information of the target item based on the target similar image;

and determining target item information of the target item based on the first item information and the second item information.

In a second aspect, an embodiment of the present invention further provides an apparatus for determining item information, where the apparatus includes:

the extraction module is used for acquiring a target image corresponding to a target article and extracting target text content and target image characteristics in the target image;

the information processing module is used for determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image, and determining second item information of the target item based on the target similar image;

and the article information determining module is used for determining the target article information of the target article based on the first article information and the second article information.

In a third aspect, an embodiment of the present invention further provides a server, where the server includes:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement a method of determining item information as in any one of the embodiments of the invention.

In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to perform the method for determining item information according to any one of the embodiments of the present invention.

According to the technical scheme of the embodiment of the invention, the text processing and the image characteristic processing are carried out on the target image corresponding to the target object, so that the first object information corresponding to the text information and the second object information corresponding to the image characteristic can be respectively determined, the target object information of the target object is determined by comprehensively considering the first object information and the second object information, and the technical effects of accuracy, convenience and high efficiency of determining the target object information are improved.

Drawings

In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, a brief description is given below of the drawings used in describing the embodiments. It should be clear that the described figures are only views of some of the embodiments of the invention to be described, not all, and that for a person skilled in the art, other figures can be derived from these figures without inventive effort.

Fig. 1 is a schematic flow chart of a method for determining item information according to an embodiment of the present invention;

fig. 2 is another schematic flow chart of a method for determining item information according to a second embodiment of the present invention;

fig. 3 is another schematic diagram of an item information determining method according to a third embodiment of the present invention;

fig. 4 is a schematic diagram of a framework for determining object information according to a third embodiment of the present invention;

fig. 5 is a schematic structural diagram of an apparatus for determining information of an article according to a fourth embodiment of the present invention;

fig. 6 is a schematic structural diagram of a server according to a fifth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a flowchart of a method for determining item information according to an embodiment of the present invention, where the embodiment is applicable to a case where the item information of a target item is determined based on an image of the target item, and the method may be performed by an apparatus for determining item information, where the apparatus may be implemented in a form of software and/or hardware, and optionally may be implemented based on a server.

As shown in fig. 1, the method of this embodiment includes:

and S110, acquiring a target image corresponding to the target object, and extracting target text content and target image characteristics in the target image.

If the item information of a certain item needs to be determined, the item can be used as a target item. Accordingly, the target image of the target object may be obtained by means of photographing or the like. The target image not only comprises a specific image of the target object, but also comprises text description information corresponding to the target object. And taking the text description information on the target image as the target text content.

In this embodiment, extracting target text content and target image features in a target image includes: acquiring target text content in a target image based on a graph-text recognition technology; and performing feature extraction on the target image based on a picture feature extraction model obtained by pre-training to obtain at least one target image feature of the target image.

The image-text recognition technology is mainly used for extracting texts in target images. The image feature extraction model is obtained by pre-training and is used for carrying out feature extraction on the image so as to obtain the target image feature of the target image. In this embodiment, the advantage of obtaining the characters and the image features in the target image is that the item information of the target item is comprehensively determined based on the text content and the image features, so that the technical effect of determining the accuracy of the target item information can be improved.

S120, determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image according to the target image characteristics, and determining second item information of the target item based on the target similar image.

The target text content includes description information of the target item, for example, a model and a specification of the target item. Correspondingly, the first item information may be an item name, a product specification, a product model, and a usage scenario of the target item, and optionally, the usage scenario may be medical or non-medical, and the like. At least one image including features of the target image may be obtained from a gallery. The target similar image is an image having the highest similarity with the target image. The photo library includes at least one image and associated information corresponding to the image, where the associated information may be item information corresponding to the target similar image, and the item information obtained at this time is used as second item information.

In this embodiment, determining the first item information of the target item based on the target text content includes: determining a target key word included in the target text content according to a predetermined key word library; determining associated content associated with the target key vocabulary from the target text content; determining first item information of the target item according to the associated content; the first item information includes at least one of a specification, a model, an applicable scene, and an item name of the target item.

The keyword library is predetermined and comprises at least one keyword. For example, the keywords of the keyword library include information such as product model, specification, and number. And taking preset keywords included in the target content as target key words. Correspondingly, the content corresponding to the target keyword in the target text content is used as the associated content. If the target keyword is the specification, the associated content is the content corresponding to the specification; if the target keyword is of an applicable type, the associated content correspondingly comprises medical or non-medical. The first item information may be associated content corresponding to a preset keyword, such as information of an item name, an item specification, an item type, and the like.

Specifically, the user may preset and determine a keyword corresponding to each article, and store the keyword in a keyword library established in advance. And acquiring character information in the target image based on the image-text recognition technology, and taking the character information as target text content. According to the key words in the pre-established key word library, the target key words included in the target text content are determined, the associated content corresponding to the target key words is determined from the target text content, and the determined associated content can be used as the first item information of the target item. By adopting the method to determine the first article information of the target article, the determined article information content can be more complete.

In this embodiment, determining a target similar image corresponding to the target image according to the target image feature, and determining second item information of the target item based on the target similar image includes: determining at least one to-be-processed similar image including target image characteristics from a pre-established image library; the image library comprises images and image characteristics matched with the images; determining a similarity value between at least one to-be-processed similar image and a target image; determining a target similar image based on the similarity value, and determining second item information of the target item based on the target similar image.

Wherein, at least one similar image to be processed comprises one, two or three images. The image library comprises a plurality of images, each image has image characteristics corresponding to the image, and the article type corresponding to the image. That is, the image library includes images and image features corresponding to the images. The similarity value may be a similarity value of the target image corresponding to each image in the image library. The second item information may be an item type of the target item.

Specifically, the pre-established image library includes a plurality of images, image characteristics corresponding to each image, and article types. Therefore, after determining the target image characteristics corresponding to the target image, at least one similar image to be processed including the target image characteristics can be determined from the pre-established image library. Calculating a similarity value between the similar image to be processed and the target similar image aiming at each similar image to be processed; according to the similarity value between each similar image to be processed and the target image, the target similar image to be processed, namely the target similar image, can be determined, and then second article information of the target article is determined based on the associated content corresponding to the target similar image.

In this embodiment, determining the target similar image based on the similarity value and determining the second item information of the target item based on the target similar image includes: acquiring a corresponding to-be-processed similar image when the similarity value is higher than a preset similarity threshold value, and taking the corresponding to-be-processed similar image as a target similar image; and determining second item information of the target item based on the item type corresponding to the target similar image.

Illustratively, the preset similarity threshold is preset, and optionally, the preset similarity threshold is ninety-five percent. And taking the corresponding image to be processed with the similarity value higher than ninety-five percent as the target similar image. According to the article type corresponding to the target similar image, the second article information of the target article can be determined, and convenience in determining the target article information is improved.

S130, determining target item information of the target item based on the first item information and the second item information.

Specifically, the target article information of the target article can be determined by comprehensively considering the first article information and the second article information, so that the accuracy of determining the target article information is improved.

It should be noted that, in the embodiment, not only the article information of the common article can be identified, but also the article information of the medical material can be identified, so that the technical effect of the accuracy of material classification is improved.

In order to improve user experience, after target text information corresponding to a target text is determined, the target text information can be converted into a target language type so as to meet user requirements of different users. Optionally, the target item information is translated into a target language type, and the target item information of the target language type is sent to the client to be displayed on a display interface of the client.

Specifically, the user may set a type of a target language, for example, japanese, on the client in advance, and may translate the determined target item information into japanese. And displaying the target article information corresponding to the target language type on a display interface of the client so as to achieve the technical effect of improving the user experience.

Example two

Fig. 2 is another schematic flow chart of a method for determining item information according to a second embodiment of the present invention. When the second item information corresponding to the target item is not determined based on the similarity value, the target image may be processed based on an item classification model obtained through pre-training, so as to obtain the second item information corresponding to the target image. The technical terms that are the same as or similar to those of the above embodiments are not repeated herein.

As shown in fig. 2, the method includes:

s210, acquiring a target image corresponding to the target object, and extracting target text content and target image characteristics in the target image.

S220, determining first article information of the target article based on the target text content, and if the target similar image corresponding to the target image is not determined, processing the target image based on an article classification model obtained through pre-training to obtain second article information corresponding to the target image.

It should be noted that, if the similarity values of the image in the image library and the target image are both smaller than the preset similarity threshold, it is indicated that the target similar picture corresponding to the target image is not determined from the picture library, that is, the second item information of the target item cannot be determined according to the target similar picture, and at this time, the second item information of the target item may be determined based on an item classification model obtained through pre-training.

The object classification model is obtained by training in advance, and is used for processing a target image of a target object to determine second object information of the target object, for example, the second object information may be an object type.

Specifically, after the second item information corresponding to the target item is not determined based on the picture search technology, the target image may be input into an item classification model trained in advance, and the second item information of the target item may be determined based on an output result of the item classification model, so that the target item information of the target item is determined based on the second item information.

In this embodiment, the target image is input into an article classification model obtained by pre-training, and the article type of the target article and the confidence level of the article type can be output; and determining second item information corresponding to the target image based on the confidence level.

The confidence coefficient is used for representing the reliability of the article class output by the article classification model, optionally, the higher the confidence coefficient is, the higher the matching degree of the article class output by the article classification model and the target article is, otherwise, the article class output by the article classification model is not matched, and the article class output by the article classification model cannot be used as the second article information.

Specifically, the target image may be input into an article classification model obtained through pre-training, and an article type of the target image and a confidence level of the article type may be determined based on the article classification model. According to the confidence of the item type, second item information of the target item, namely the item type of the target item, can be determined. The arrangement mode has the advantage that the accuracy and convenience for determining the second article information of the target article can be improved.

In this embodiment, the determining, based on the confidence, second item information corresponding to the target image includes: when the confidence value is higher than a preset confidence threshold value, taking the article class output by the article classification model as the second article information; and when the confidence value is lower than a preset confidence threshold value, the second article information is preset content.

The preset confidence threshold is predetermined, and optionally, the preset confidence threshold may be 95%.

Specifically, the item class of the target item and the confidence corresponding to the item class can be determined based on the item classification model. When the confidence value is higher than the preset confidence threshold value, the article class is credible, and the article class can be used as second article information; when the confidence coefficient is lower than the preset confidence coefficient threshold value, the article type output based on the article classification model and the target article have certain difference, and the article type cannot be used as the article type of the target article.

In an actual application process, if the confidence degree output based on the article classification model is lower than the preset confidence degree threshold, it is indicated that the article type output by the article classification model is greatly different from the article type of the target article, at this time, the second article information may be preset content, and optionally, the preset content may be empty, so as to determine the target article information of the target article based on the first article information and the second article information.

And S230, determining target item information of the target item based on the first item information and the second item information.

Optionally, if the second item information is the preset content, determining the target item information of the target item based on the first item information and the second item information, including: target item information for the target item is determined based on the first item information.

Specifically, based on the target image corresponding to the target object, the first object information of the target object may be determined, and based on the first object information, the target object information of the target object may be determined, and optionally, the object type, the object name, the object specification, the model number, and the information in the usage scenario of the target object.

According to the technical scheme of the embodiment of the invention, when the similar picture corresponding to the target image is not searched based on the picture searching technology or the second article information of the target article is not determined based on the picture searching technology, the second article information of the target article can be determined based on the article classification model obtained by pre-training, the target article information corresponding to the target article can be rapidly, accurately and conveniently determined based on the first article information and the second article information, and the technical effects of determining the accuracy and convenience of the target article information are improved.

EXAMPLE III

Fig. 3 is another schematic diagram of a method for determining item information according to a third embodiment of the present invention. Before the technical solution of the embodiment of the present invention is introduced, each module and the corresponding function of each module in the schematic diagram are briefly introduced. The material discernment includes material discernment interface, and with the corresponding at least three call of material discernment interface, it is respectively: the method comprises the steps of image searching and calling, commodity identification and image-text identification and calling. The image search call is mainly to determine whether an image similar to the target image exists in the image library according to the target image corresponding to the target article, and further determine second article information of the target article. The commodity identification calling is mainly to process a target image based on an article type identification model obtained by pre-training and determine second article information corresponding to the target image; the image-text recognition call mainly recognizes character information on the target image, and determines first article information of the target article based on the text information. Target item information of the target item is determined by comprehensively considering the first item information and the second item information.

Specifically, referring to fig. 3, after the target image corresponding to the target object is sent to the material recognition interface, the target image is processed based on a graph search call, a commodity recognition call, and an OCR (image and text recognition) call corresponding to the material recognition interface. Wherein, processing the target image based on the graph search call may be: referring to the system architecture for image invocation in fig. 3, the image search system mainly includes two parts, which are an upper module of the image search system and a bottom module of the image search system, respectively, where the upper module mainly includes an interface layer, a central logic module, and a database. The interface layer mainly comprises image warehousing and searching, the image warehousing mainly refers to an interface for creating a picture library, and the searching mainly comprises the step of determining an interface of a picture similar to the target image from the picture library according to the target image. The picture verification is mainly used for verifying whether the format or the size of the picture meets a preset condition, and if the format or the size of the picture meets the preset condition, the target image is directly processed; if the preset condition is not met, the target image is preprocessed to enable the target image to meet the preset condition. And sending the processed target image meeting the preset conditions to a bottom layer to process the target image. And the bottom layer of the image search system carries out target detection, classification, feature extraction and other operations on the target image by a real-time processing module, and finally recalls the image with the similarity value higher than a preset similarity threshold value from the database based on neighbor search, sorts the recalled images according to the similarity value, and returns the images to the upper layer of the image search. And the image searching upper layer determines the associated information corresponding to the recalled images according to the recalled images, such as the article type corresponding to each image, and returns the similar images and the associated information corresponding to the similar images. That is, after the image search call processes the target image, a similar image corresponding to the target image, a similarity value, and associated information corresponding to the target image may be determined. If the similarity value is greater than or equal to the preset similarity threshold, the associated information corresponding to the similar image may be used as the second item information.

And if the similarity value is smaller than the preset similarity threshold value, processing the target image based on commodity calling. Specifically, a plurality of training sample data are obtained, the training sample data include a sample image, an article type corresponding to the sample image, and a confidence level, and since the article type marked in the sample data is consistent with the target article, the confidence level of the marked article type at this time is 1. And training the training sample data by adopting a classification algorithm to obtain a determined article classification model. And inputting the target image into an article classification model obtained by pre-selection training, so as to obtain the article class corresponding to the target article and the confidence coefficient of the article class. If the confidence coefficient is larger than or equal to the preset confidence coefficient threshold value, the article type output by the model can be used as second article information. If the confidence is smaller than the preset confidence threshold, the second item information may be marked as a preset content, and optionally, the preset content is empty.

And acquiring target text content on the target image based on an image-text recognition (OCR) call, and acquiring associated content corresponding to preset keywords, such as the name of the article, the rule of the article, the type of the article and the like, from the target content based on the preset keywords. The item information determined based on the text content may be used as the first item information.

According to the name, specification and model of the article in the first article information, the article type in the second article information can comprehensively determine the target article information corresponding to the target article.

Based on the above technical solution, fig. 4 is a schematic diagram of a framework for determining object information provided in a third embodiment of the present invention. The material identification platform supports multi-instance deployment and supports dynamic lateral extension deployment according to calling and concurrency, namely, if material identification can simultaneously receive a plurality of target images, and the target images are respectively processed based on the modes adopted by the first embodiment and the second embodiment to determine target article information of the target images. That is, target item information for the target item is determined based on the graph search call, the item recognition call, and the OCR call.

Example four

Fig. 5 is a schematic structural diagram of an apparatus for determining item information according to a fourth embodiment of the present invention, as shown in fig. 5, the apparatus includes: an extraction module 510, an information processing module 520, and an item information determination module 530.

The extracting module 510 is configured to obtain a target image corresponding to a target article, and extract target text content and target image features in the target image; an information processing module 520, configured to determine first item information of the target item based on the target text content, determine a target similar image corresponding to the target image, and determine second item information of the target item based on the target similar image; an item information determining module 530, configured to determine target item information of the target item based on the first item information and the second item information.

On the basis of the above technical solution, the extraction module is further configured to: acquiring target text content in a target image based on a graph-text recognition technology; and performing feature extraction on the target image based on a picture feature extraction model obtained by pre-training to obtain at least one target image feature of the target image.

On the basis of the above technical solutions, the information processing module is further configured to: determining a target key word included in the target text content according to a predetermined key word library; determining associated content associated with the target key vocabulary from the target text content; determining first item information of the target item according to the associated content; the first article information includes at least one item of specification, model, applicable scene and article name of the target article.

On the basis of the above technical solutions, the information processing module is further configured to determine at least one to-be-processed similar image including the target image feature from a pre-established image library; the image library comprises images and image characteristics matched with the images; determining a similarity value between the at least one to-be-processed similar image and the target image; determining a target similar image based on the similarity value, and determining second item information of the target item based on the target similar image.

On the basis of the above technical solutions, the information processing module is further configured to: acquiring a corresponding to-be-processed similar image when the similarity value is higher than a preset similarity threshold value, and taking the corresponding to-be-processed similar image as the target similar image; and determining second item information of the target item based on the item type corresponding to the target similar image.

On the basis of the above technical solutions, the apparatus further includes a second item information processing module, and is further configured to, if the similarity values are all smaller than a preset similarity threshold value, process the target image based on an item classification model obtained through pre-training to obtain second item information corresponding to the target image.

On the basis of the technical solutions, the device further includes a second article information processing module, which is further configured to input the target image into an article classification model obtained through pre-training, and output an article type of the target article and a confidence level of the article type; and determining second item information corresponding to the target image based on the confidence.

On the basis of the above technical solutions, the apparatus further includes a second item information processing module, which is further configured to, when the confidence value is higher than a preset confidence threshold, take the item type output by the item classification model as the second item information; and when the confidence value is lower than a preset confidence threshold value, the second article information is preset content.

On the basis of the above technical solutions, the article information determining module is further configured to determine target article information of the target article based on the first article information.

On the basis of the above technical solutions, the article information determining module is further configured to determine the target article information of the target article based on the article name, the specification, the model, and the applicable scene in the first article information and the article type in the second article information.

On the basis of the above technical solutions, the apparatus further includes: and the translation module is used for translating the target article information into a target language type. The device for determining the article information provided by the embodiment of the invention can execute the method for determining the article information provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

It should be noted that, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiment of the invention.

EXAMPLE five

Fig. 6 is a schematic structural diagram of a server according to a fifth embodiment of the present invention. FIG. 6 illustrates a block diagram of an exemplary server 60 suitable for use in implementing embodiments of the present invention. The server 60 shown in fig. 6 is only an example, and should not bring any limitation to the function and the scope of use of the embodiment of the present invention.

As shown in fig. 6, the server 60 is in the form of a general purpose computing server. The components of server 60 may include, but are not limited to: one or more processors or processing units 601, a system memory 602, and a bus 603 that couples various system components including the system memory 602 and the processing unit 601.

Bus 603 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

The server 60 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by server 60 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 602 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)604 and/or cache memory 605. The server 60 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 606 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 603 by one or more data media interfaces. Memory 602 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 608 having a set (at least one) of program modules 607 may be stored, for example, in memory 602, such program modules 607 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 607 generally perform the functions and/or methods of the described embodiments of the invention.

The server 60 may also communicate with one or more external devices 609 (e.g., keyboard, pointing server, display 610, etc.), with one or more servers that enable a user to interact with the server 60, and/or with any servers (e.g., network card, modem, etc.) that enable the server 60 to communicate with one or more other computing servers. Such communication may occur via an input/output (I/O) interface 611. Also, the server 60 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet) via a network adapter 612. As shown, the network adapter 612 communicates with the other modules of the server 60 via the bus 603. It should be appreciated that although not shown in FIG. 6, other hardware and/or software modules may be used in conjunction with the server 60, including but not limited to: microcode, server drives, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processing unit 601 executes various functional applications and data processing by executing programs stored in the system memory 602, for example, to implement the method for determining article information provided by the embodiment of the present invention.

EXAMPLE six

An embodiment of the present invention also provides a storage medium containing computer-executable instructions for performing a method of determining item information when executed by a computer processor.

The method comprises the following steps:

determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image according to target image characteristics, and determining second item information of the target item based on the target similar image;

Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having 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. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for embodiments of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method of determining item information, comprising:

determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image according to the target image characteristic, and determining second item information of the target item based on the target similar image;

2. The method of claim 1, wherein extracting target text content and target image features in the target image comprises:

acquiring target text content in a target image based on a graph-text recognition technology;

and performing feature extraction on the target image based on a picture feature extraction model obtained by pre-training to obtain at least one target image feature of the target image.

3. The method of claim 1, wherein the determining first item information for the target item based on the target textual content comprises:

determining a target key word included in the target text content according to a predetermined key word library;

determining associated content associated with the target key vocabulary from the target text content;

determining first item information of the target item according to the associated content; the first article information includes at least one item of specification, model, applicable scene and article name of the target article.

4. The method according to claim 1, wherein the determining a target similar image corresponding to the target image according to the target image feature and determining second item information of the target item based on the target similar image comprises:

determining at least one to-be-processed similar image including the target image characteristics from a pre-established image library; the image library comprises images and image characteristics matched with the images;

determining a similarity value between the at least one to-be-processed similar image and the target image;

determining a target similar image based on the similarity value, and determining second item information of the target item based on the target similar image.

5. The method of claim 4, wherein determining a target similar image based on the similarity value and determining second item information for the target item based on the target similar image comprises:

acquiring a corresponding to-be-processed similar image when the similarity value is higher than a preset similarity threshold value, and taking the corresponding to-be-processed similar image as the target similar image;

and determining second item information of the target item based on the item type corresponding to the target similar image.

6. The method of claim 4, further comprising:

and if the similarity values are all smaller than a preset similarity threshold value, processing the target image based on an article classification model obtained through pre-training to obtain second article information corresponding to the target image.

7. The method according to claim 6, wherein the processing the target image based on the pre-trained object classification model to obtain second object information corresponding to the target image comprises:

inputting the target image into an article classification model obtained by pre-training, and outputting the article type of the target article and the confidence coefficient of the article type;

and determining second item information corresponding to the target image based on the confidence.

8. The method of claim 7, wherein determining second item information corresponding to the target image based on the confidence level comprises:

when the confidence value is higher than a preset confidence threshold value, taking the article class output by the article classification model as the second article information;

and when the confidence value is lower than a preset confidence threshold value, the second article information is preset content.

9. The method according to claim 8, wherein the second item information is preset content, and the determining the target item information of the target item based on the first item information and the second item information comprises:

target item information of the target item is determined based on the first item information.

10. The method of claim 1, further comprising:

and translating the target article information into a target language type, and sending the target article information of the target language type to a client so as to display on a display interface of the client.

11. An apparatus for determining item information, comprising:

the information processing module is used for determining first item information of the target item based on the target text content, determining a target similar image corresponding to the target image according to the target image characteristic, and determining second item information of the target item based on the target similar image;

12. A server, characterized in that the server comprises:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement a method of determining item information as claimed in any one of claims 1-10.

13. A storage medium containing computer executable instructions for performing the method of determining item information of any one of claims 1-10 when executed by a computer processor.