Article index library updating method and device, electronic equipment and computer readable medium

Abstract

Embodiments of the present disclosure disclose an article index library updating method and device, electronic equipment and a computer readable medium. A specific implementation of the method comprises: obtaining a first article image feature set; in response to determining that there is an article image feature set satisfying a first preset condition in an article image feature library, merging the article image feature set satisfying the first preset condition and the first article image feature set to obtain a second article image feature set, and performing deletion processing on the article image feature sets in the article image feature library; performing denoising processing on the second article image feature set to obtain a third article image feature set; adding the third article image feature set to the article image feature library; determining an article index corresponding to the third article image feature set, and updating an article index library using the article index corresponding to the third article image feature set. The implementation realizes automatic and real-time updating of the article index library.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of computer technology, and more specifically to methods, apparatus, electronic devices, and computer-readable media for updating an item index database. Background Technology

[0002] The item index in the item index library is an identifier set for items to improve the speed of item retrieval. Currently, the common method for generating an item index library is as follows: directly build the item index library using existing item images, and then continuously collect item images during the use of the built item index library. The collected item images are manually reviewed and labeled. When the number of collected item images reaches a preset number or the collection time reaches a preset time, the item index library is manually rebuilt using newly collected item images and existing item images, thereby achieving the purpose of updating the item index library.

[0003] However, when building and updating the item index using the above method, the following technical problems often arise:

[0004] First, the construction and updating of the item index database requires manual review and annotation of the collected item images, and manual construction and updating of the item index database, which increases the workload of personnel and cannot achieve automatic and real-time updates of the item index database.

[0005] Second, the collected data may contain dirty data, which will reduce the matching degree between the item index and the item, thereby reducing the accuracy of item retrieval. Summary of the Invention

[0006] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0007] Some embodiments of this disclosure provide methods, apparatus, electronic devices, and computer-readable media for updating an item index database to address one or more of the technical problems mentioned in the background section above.

[0008] In a first aspect, some embodiments of this disclosure provide a method for updating an item index library. The method includes: acquiring a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group; in response to determining that an item image feature group satisfying a first preset condition exists in the item image feature library, merging the item image feature group satisfying the first preset condition with the first item image feature group to obtain a second item image feature group; and deleting the item image feature group in the item image feature library, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group; performing noise reduction processing on the second item image feature group to obtain a third item image feature group; adding the third item image feature group to the item image feature library; determining the item index corresponding to the third item image feature group, and updating the item index library using the item index corresponding to the third item image feature group.

[0009] Secondly, some embodiments of this disclosure provide an item index database update apparatus, the apparatus comprising: an acquisition unit configured to acquire a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group; a merging unit configured to, in response to determining that there exists an item image feature group in the item image feature database that satisfies a first preset condition, merge the item image feature group that satisfies the first preset condition and the first item image feature group to obtain a second item image feature group, and delete the item image feature group in the item image feature database, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group; a denoising unit configured to denoise the second item image feature group to obtain a third item image feature group; an adding unit configured to add the third item image feature group to the item image feature database; and an updating unit configured to determine the item index corresponding to the third item image feature group and update the item index database using the item index corresponding to the third item image feature group.

[0010] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0011] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0012] The above embodiments of this disclosure have the following beneficial effects: the item index library update method of some embodiments of this disclosure can realize automatic and real-time updates of the item index library without manual intervention. Specifically, the reason why the item index library cannot be automatically and in real-time updated is that the construction and updating of the item index library requires manual review and annotation of the collected item images, and manual construction and updating of the item index library. Based on this, the item index library update method of some embodiments of this disclosure firstly obtains a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group. Then, in response to determining that there is an item image feature group in the item image feature library that meets a first preset condition, the item image feature group that meets the first preset condition and the first item image feature group are merged to obtain a second item image feature group, and the item image feature group in the item image feature library is deleted, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group. Therefore, by merging the newly acquired image feature groups with the existing image feature groups, the image feature groups can be updated in real time. Then, the second item image feature group is denoised to obtain the third item image feature group. Thus, by denoising the real-time updated image feature group, dirty data in the updated image feature group can be removed, improving the matching degree between the image feature group and the item, thereby improving the accuracy of item recognition. Next, the third item image feature group is added to the item image feature library. This enables real-time and automatic updates to the item image feature library. Finally, the item index corresponding to the third item image feature group is determined, and the item index library is updated using the item index corresponding to the third item image feature group. Thus, by updating the item index library with the item index corresponding to the denoised image feature group, automatic and real-time updates to the item index library are achieved. Because the construction and updating of the item index library does not require manual intervention, automatic and real-time updates to the item index library are realized. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0014] Figure 1 This is a schematic diagram illustrating an application scenario of the item index database update method according to some embodiments of this disclosure;

[0015] Figure 2 This is a flowchart of some embodiments of the item index database update method according to this disclosure;

[0016] Figure 3 This is a schematic diagram of the structure of some embodiments of the item index database update apparatus according to the present disclosure;

[0017] Figure 4 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0018] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0019] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0020] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0021] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0022] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0023] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] Figure 1 This is a schematic diagram illustrating an application scenario of the item index database update method according to some embodiments of this disclosure.

[0025] exist Figure 1 In the application scenario, firstly, the computing device 101 can acquire a first item image feature group 102, wherein the first item image feature group 102 corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group 102. Then, in response to determining that there is an item image feature group in the item image feature library 103 that satisfies a first preset condition, the computing device 101 can merge the item image feature group that satisfies the first preset condition with the first item image feature group 102 to obtain a second item image feature group 104, and perform deletion processing on the item image feature group in the item image feature library 103, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group 102. Next, the computing device 101 can perform noise reduction processing on the second item image feature group 104 to obtain a third item image feature group 105. Afterwards, the computing device 101 can add the third item image feature group 105 to the item image feature library 103. Finally, the computing device 101 can determine the item index corresponding to the third item image feature group 105 and update the item index library 106 using the item index corresponding to the third item image feature group 105.

[0026] It should be noted that the aforementioned computing device 101 can be either hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed within the hardware devices listed above. It can be implemented as, for example, multiple software programs or software modules used to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0027] It should be understood that Figure 1 The number of computing devices shown is merely illustrative. Any number of computing devices can be used depending on implementation needs.

[0028] Continue to refer to Figure 2 The diagram illustrates a flow 200 of some embodiments of an item index database update method according to the present disclosure. This item index database update method includes the following steps:

[0029] Step 201: Obtain the first item image feature group.

[0030] In some embodiments, the execution body of the item index update method (such as...) Figure 1The computing device 101 shown can acquire a first set of image features for an item. This first set of image features can correspond to a category label. The category label can be the category label of the item corresponding to the first set of image features. For example, the executing entity can acquire the first set of image features for an item using a deep learning model. The deep learning model can include, but is not limited to, at least one of the following: CNN (Convolutional Neural Networks) or VGG (Visual Graphics Generator).

[0031] Optionally, before obtaining the first set of item image features, the aforementioned execution entity can extract image features of the items in the target image to obtain the first set of item image features. The target image can be an image of the acquired item or an image of the item captured in real time. The first set of item image features is a collection of item image features extracted from the target image. For example, the aforementioned execution entity can automatically extract image features of the items in the target image using HOG (Histogram of Oriented Gradients) and automatically label the extracted image features to obtain the first set of item image features.

[0032] Step 202: In response to determining that there is an item image feature group in the item image feature library that meets the first preset condition, the item image feature group that meets the first preset condition and the first item image feature group are merged to obtain the second item image feature group, and the item image feature group in the item image feature library is deleted.

[0033] In some embodiments, the executing entity may, in response to determining that an item image feature group satisfying a first preset condition exists in the item image feature library, merge the item image feature group satisfying the first preset condition with the first item image feature group to obtain a second item image feature group, and delete the item image feature group in the item image feature library. The first preset condition may be that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group. For example, the executing entity may, in response to determining that an item image feature group satisfying the first preset condition exists in the item image feature library, add the item image feature group satisfying the first preset condition to the first item image feature group to obtain a second item image feature group, and delete the item image feature group satisfying the first preset condition from the item image feature library.

[0034] Therefore, by merging the new image feature sets acquired in real time with the existing image feature sets, the image feature sets can be updated in real time.

[0035] Optionally, the aforementioned executing entity may, in response to determining that there are no item image features in the item image feature library that satisfy the first preset condition, determine the first item image feature group as the second item image feature group.

[0036] Step 203: Denoise the second item image feature group to obtain the third item image feature group.

[0037] In some embodiments, the execution entity may perform denoising processing on the second item image feature group to obtain a third item image feature group. For example, binning can be used to remove dirty data from the second item image feature group to obtain the third item image feature group. The binning method may include, but is not limited to, at least one of the following: equal-frequency binning, equal-distance binning, or decision tree binning.

[0038] Therefore, by denoising the image feature set updated in real time, dirty data in the updated image feature set can be removed, improving the matching degree between the image feature set and the object, and thus improving the accuracy of object recognition.

[0039] Optionally, the execution entity may perform denoising processing on the second item image feature group to obtain the third item image feature group, which may include the following steps:

[0040] The first step is to perform clustering processing on the second item image features in the above-mentioned second item image feature group to obtain a set of second item image feature clusters.

[0041] Clustering algorithms can be used to cluster the features of the second item image in the aforementioned second item image feature group, resulting in a set of second item image feature clusters. These clustering algorithms can include K-means clustering or DBSCAN (Density-Based Spatial Clustering of Applications with Noise). The number of clusters in the aforementioned set of second item image feature clusters can be determined using the following formula:

[0042]

[0043] Where z represents the number of clusters. Z represents a preset value. For example, Z can be 20 or 30. n represents the number of second item image features in the second item image feature group mentioned above. c is a preset value. For example, c can be 3 or 5. This indicates rounding down. min() indicates taking the minimum value.

[0044] Therefore, by performing clustering processing on the second item image features in the above-mentioned second item image feature group, second item image features with high similarity can be divided into the same cluster, and second item image features with low similarity can be divided into different clusters, thereby achieving the initial division of the above-mentioned second item image feature group.

[0045] The second step involves performing the following merging steps on the second set of image feature clusters for the item:

[0046] The first sub-step is to determine the similarity between each feature cluster of the second item image in the second item image feature cluster set, and obtain the similarity set.

[0047] The similarity between the aforementioned feature clusters of the second item images refers to the similarity between the cluster centers of any two feature clusters of the second item images within the set of feature clusters. The similarity between two feature clusters of the second item images can be generated using the following formula:

[0048]

[0049] Where i represents the sequence number, j represents the sequence number, and k represents the sequence number. S represents the similarity between two second item image feature clusters in the above set of second item image feature clusters. i，j Let X represent the similarity between the i-th and j-th second item image feature clusters in the aforementioned set of second item image feature clusters. Let X represent the set of cluster centers corresponding to the aforementioned set of second item image feature clusters. i，k X represents the k-th component of the i-th cluster center in the above set of cluster centers. j，k Let represent the k-th component of the j-th cluster center in the above set of cluster centers. d represents the dimension of the cluster center.

[0050] The aforementioned executing entity may determine the cluster center set corresponding to the aforementioned second item image feature cluster set by including the following steps:

[0051] The first step is to determine the mean of each second item image feature cluster in the aforementioned set of second item image feature clusters.

[0052] The second step is to standardize the mean of each second item image feature cluster in the above-mentioned second item image feature cluster set to obtain the cluster center set corresponding to the above-mentioned second item image feature cluster set.

[0053] The mean of each second item image feature cluster in the above set of second item image feature clusters can be standardized according to the following formula:

[0054]

[0055] Where x represents the cluster center. represents the mean of the feature cluster of the second item image. |||| represents the norm.

[0056] The second sub-step is to merge the two second item image feature clusters corresponding to the largest similarity in the similarity set that are greater than the first preset threshold, in response to the determination that the largest similarity in the similarity set is greater than the first preset threshold, so as to obtain the merged second item image feature cluster.

[0057] The aforementioned first preset threshold can be a pre-set value. Here, there is no limitation on the setting of the first preset threshold.

[0058] The third sub-step involves updating the second item image feature cluster set using the merged second item image feature clusters to obtain the updated second item image feature cluster set, and using the updated second item image feature cluster set as the second item image feature cluster set.

[0059] The aforementioned execution entity can delete the two second item image feature clusters corresponding to the highest similarity in the similarity set from the second item image feature cluster set, and add the merged second item image feature clusters to the second item image feature cluster set to obtain the updated second item image feature cluster set.

[0060] Therefore, by merging the second item image feature clusters with high similarity in the above-mentioned second item image feature cluster set, it is possible to integrate the second item image feature clusters that represent the same item.

[0061] The third step is to determine that the largest similarity in the similarity set is less than or equal to the first preset threshold or that the second item image feature cluster set contains only one second item image feature cluster, and to determine that the above merging step has been completed, and to generate the weight of each second item image feature cluster in the second item image feature cluster set, thereby obtaining the cluster weight set.

[0062] The aforementioned cluster weight set can be the ratio of the number of second item image features in each second item image feature cluster in the second item image feature cluster set to the sum of the number of second item image features included in each second item image feature cluster in the second item image feature cluster set.

[0063] Therefore, by using preset similarity conditions, the cluster division of the above-mentioned second item image feature group can be finally determined, and by determining the proportion of each second item image feature cluster in the second item image feature cluster set, data support can be provided for filtering out dirty data in the second item image feature group.

[0064] Optionally, the aforementioned executing entity may continue to perform the aforementioned merging steps in response to determining that the second item image feature cluster set includes at least two second item image feature clusters.

[0065] Optionally, the execution entity may perform denoising processing on the second item image feature group to obtain the third item image feature group, which may include the following steps:

[0066] The first step is to delete the second item image feature clusters corresponding to the cluster weights less than or equal to the second preset threshold from the above set of cluster weights, thereby obtaining the deleted set of second item image feature clusters.

[0067] The aforementioned second preset threshold can be a pre-set value. Here, there is no limitation on the setting of the second preset threshold.

[0068] The second step involves merging the second item image feature clusters from the previously deleted second item image feature cluster set to obtain the third item image feature group. Since item recognition depends on the proportion of item image features that correctly represent the item within the item image feature group, a higher proportion results in higher accuracy. However, the similarity between unmerged second item image feature clusters is relatively large, leading to a low probability of them representing the same item, thus affecting item recognition. Therefore, by setting a second preset threshold, second item image feature clusters that do not meet the threshold are deleted as dirty data, thereby improving the matching degree between item image feature groups and items, and ultimately improving the accuracy of item retrieval.

[0069] Step 204: Add the third item image feature group to the item image feature library.

[0070] In some embodiments, the execution entity may add the third item image feature group to the item image feature library.

[0071] This enables real-time and automatic updates to the object image feature library.

[0072] Step 205: Determine the item index corresponding to the third item image feature group, and update the item index database using the item index corresponding to the third item image feature group.

[0073] In some embodiments, the execution entity may determine the item index corresponding to the third item image feature group and update the item index database using the item index corresponding to the third item image feature group. The item index may be a unique identifier corresponding to the item.

[0074] Therefore, by updating the item index corresponding to the denoised image feature group to the item index library, the item index library can be automatically and in real time updated.

[0075] Optionally, the execution entity may determine the item index corresponding to the third item image feature group by including the following steps:

[0076] The first step is to determine the mean value of the third item image feature group mentioned above.

[0077] The arithmetic mean of the image feature group of the third item mentioned above can be determined.

[0078] The second step is to standardize the above mean values ​​to determine the item index corresponding to the third item image feature group.

[0079] The above mean can be standardized using the following formula:

[0080]

[0081] Where y represents the index of the above item. represents the mean above. |||| represents the norm.

[0082] Optionally, the execution entity may update the item index database using the item index corresponding to the third item image feature group, which may include the following steps:

[0083] The first step is to delete the item index that meets the second preset condition from the item index database after determining that there is an item index in the above item index database.

[0084] The second preset condition is that the category label corresponding to the item index in the item index library is the same as the category label corresponding to the first item image feature group.

[0085] The second step is to add the item index corresponding to the third item image feature group to the item index library.

[0086] Therefore, by pre-setting the conditions for generating the item index and updating the item index database, it is possible to automatically generate the item index and update it to the item index database in real time without human intervention.

[0087] The above embodiments of this disclosure have the following beneficial effects: the item index library update method of some embodiments of this disclosure can realize automatic and real-time updates of the item index library without manual intervention. Specifically, the reason why the item index library cannot be automatically and in real-time updated is that the construction and updating of the item index library requires manual review and annotation of the collected item images, and manual construction and updating of the item index library. Based on this, the item index library update method of some embodiments of this disclosure firstly obtains a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group. Then, in response to determining that there is an item image feature group in the item image feature library that meets a first preset condition, the item image feature group that meets the first preset condition and the first item image feature group are merged to obtain a second item image feature group, and the item image feature group in the item image feature library is deleted, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group. Therefore, by merging the newly acquired image feature groups with the existing image feature groups, the image feature groups can be updated in real time. Then, the second item image feature group is denoised to obtain the third item image feature group. Thus, by denoising the real-time updated image feature group, dirty data in the updated image feature group can be removed, improving the matching degree between the image feature group and the item, thereby improving the accuracy of item recognition. Next, the third item image feature group is added to the item image feature library. This enables real-time and automatic updates to the item image feature library. Finally, the item index corresponding to the third item image feature group is determined, and the item index library is updated using the item index corresponding to the third item image feature group. Thus, by updating the item index library with the item index corresponding to the denoised image feature group, automatic and real-time updates to the item index library are achieved. Because the construction and updating of the item index library does not require manual intervention, automatic and real-time updates to the item index library are realized.

[0088] Further reference Figure 3 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of an item index database update apparatus, which are similar to... Figure 2 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0089] like Figure 3As shown, the item index library update device 300 in some embodiments includes: an acquisition unit 301, a merging unit 302, a noise reduction unit 303, an adding unit 304, and an updating unit 305. The acquisition unit 301 is configured to acquire a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group; the merging unit 302 is configured to, in response to determining that there is an item image feature group in the item image feature library that meets a first preset condition, merge the item image feature group that meets the first preset condition and the first item image feature group to obtain a second item image feature group, and delete the item image feature group in the item image feature library, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group; the denoising unit 303 is configured to denoise the second item image feature group to obtain a third item image feature group; the adding unit 304 is configured to add the third item image feature group to the item image feature library; and the updating unit 305 is configured to determine the item index corresponding to the third item image feature group and update the item index library using the item index corresponding to the third item image feature group.

[0090] It is understandable that the units described in the device 300 are related to the reference. Figure 2 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 300 and the units contained therein, and will not be repeated here.

[0091] The following is for reference. Figure 4 It illustrates electronic devices suitable for implementing some embodiments of the present disclosure (such as...). Figure 1 The diagram shows the structure of the computing device 101)400. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0092] like Figure 4 As shown, electronic device 400 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 401, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 402 or a program loaded from storage device 408 into random access memory (RAM) 403. RAM 403 also stores various programs and data required for the operation of electronic device 400. Processing device 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0093] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic device 400 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 An electronic device 400 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 4 Each box shown can represent a device or multiple devices as needed.

[0094] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 409, or installed from storage device 408, or installed from ROM 402. When the computer program is executed by processing device 401, it performs the functions defined above in the methods of some embodiments of this disclosure.

[0095] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0096] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0097] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire a first set of item image features, wherein the first set of item image features corresponds to a category label, and the category label is the category label of the item corresponding to the first set of item image features; in response to determining that an item image feature set satisfying a first preset condition exists in the item image feature library, merge the item image feature set satisfying the first preset condition with the first item image feature set to obtain a second set of item image features, and delete the item image feature set in the item image feature library, wherein the first preset condition is that the category label corresponding to the item image feature set is the same as the category label corresponding to the first item image feature set; perform noise reduction processing on the second set of item image features to obtain a third set of item image features; add the third set of item image features to the item image feature library; determine the item index corresponding to the third set of item image features, and update the item index library using the item index corresponding to the third set of item image features.

[0098] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0099] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0100] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an acquisition unit, a merging unit, a denoising unit, an adding unit, and an updating unit. The names of these units do not necessarily limit the specific unit; for example, the acquisition unit may also be described as "a unit that acquires a first set of image features of an object."

[0101] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0102] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for updating an item index database, comprising: Obtain a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group; In response to determining that there is an item image feature group in the item image feature library that meets a first preset condition, the item image feature group that meets the first preset condition and the first item image feature group are merged to obtain a second item image feature group, and the item image feature group in the item image feature library is deleted, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group. The second item image feature group is denoised to obtain the third item image feature group; Add the third item image feature group to the item image feature library; Determine the item index corresponding to the third item image feature group, and update the item index database using the item index corresponding to the third item image feature group; The step of determining the item index corresponding to the third item image feature group includes: Determine the mean value of the third item image feature group; The mean is standardized to determine the item index corresponding to the third item image feature group.

2. The method according to claim 1, wherein, Before acquiring the first set of image features for the item, the method further includes: Extract the image features of the objects in the target image to obtain the first object image feature group.

3. The method according to claim 1, wherein, The step of denoising the second item image feature group to obtain the third item image feature group includes: The second item image features in the second item image feature group are clustered to obtain a set of second item image feature clusters. Perform the following merging steps on the second set of image feature clusters: Determine the similarity between each feature cluster of the second item image in the second item image feature cluster set to obtain a similarity set; In response to the determination that the largest similarity in the similarity set is greater than a first preset threshold, the two second item image feature clusters corresponding to the largest similarity are merged to obtain the merged second item image feature cluster; The second item image feature cluster set is updated using the merged second item image feature cluster to obtain the updated second item image feature cluster set, and the updated second item image feature cluster set is used as the second item image feature cluster set; In response to determining that the largest similarity in the similarity set is less than or equal to the first preset threshold or that the second item image feature cluster set contains only one second item image feature cluster, the merging step is determined to be completed, and the weight of each second item image feature cluster in the second item image feature cluster set is generated to obtain the cluster weight set.

4. The method according to claim 3, wherein, The step of denoising the second item image feature group to obtain the third item image feature group further includes: In response to determining that the set of second item image feature clusters includes at least two second item image feature clusters, the merging step continues.

5. The method according to claim 4, wherein, The step of denoising the second item image feature group to obtain the third item image feature group further includes: The second item image feature clusters corresponding to cluster proportions less than or equal to the second preset threshold in the cluster proportion set are deleted from the second item image feature cluster set to obtain the deleted second item image feature cluster set. The second item image feature clusters in the deleted second item image feature cluster set are merged to obtain the third item image feature group.

6. The method according to claim 5, wherein, The step of updating the item index library using the item index corresponding to the third item image feature group includes: In response to determining that there is an item index in the item index library that meets a second preset condition, the item index that meets the second preset condition is deleted from the item index library, wherein the second preset condition is that the category label corresponding to the item index in the item index library is the same as the category label corresponding to the first item image feature group; Add the item index corresponding to the third item image feature group to the item index library.

7. An item index database update device, comprising: The acquisition unit is configured to acquire a first item image feature group, wherein the first item image feature group corresponds to a category label, and the category label is the category label of the item corresponding to the first item image feature group; The merging unit is configured to, in response to determining that there is an item image feature group in the item image feature library that satisfies a first preset condition, merge the item image feature group that satisfies the first preset condition with the first item image feature group to obtain a second item image feature group, and delete the item image feature group in the item image feature library, wherein the first preset condition is that the category label corresponding to the item image feature group is the same as the category label corresponding to the first item image feature group. The denoising unit is configured to denoise the second item image feature group to obtain the third item image feature group; The adding unit is configured to add the third item image feature group to the item image feature library; The update unit is configured to determine the item index corresponding to the third item image feature group and update the item index library using the item index corresponding to the third item image feature group; wherein, determining the item index corresponding to the third item image feature group includes: determining the mean of the third item image feature group; standardizing the mean to determine the item index corresponding to the third item image feature group.

8. An electronic device, comprising: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 6.

Images