CN115222986A

CN115222986A - Method, device, equipment and medium for updating article display information

Info

Publication number: CN115222986A
Application number: CN202210823997.5A
Authority: CN
Inventors: 李朋芯; 王传正; 李想; 王迎春
Original assignee: Yantai Chuangyi Software Co ltd
Current assignee: Yantai Chuangyi Software Co ltd
Priority date: 2022-07-13
Filing date: 2022-07-13
Publication date: 2022-10-21

Abstract

The invention discloses a method, a device, equipment and a medium for updating article display information. The method comprises the following steps: acquiring a target shelf image, and detecting the target shelf image based on a target detection model to obtain at least one article single body detection result; determining at least one group of article single type detection results according to the at least one article single type detection result; and updating the article display information of the target shelf according to the article single detection result and the article single type detection result. The technical scheme solves the problem of low updating efficiency of the article display information, can effectively improve the updating efficiency of the article display information and reduce the management cost while ensuring the article identification accuracy.

Description

Method, device, equipment and medium for updating article display information

Technical Field

The invention relates to the technical field of computer vision, in particular to a method, a device, equipment and a medium for updating article display information.

Background

Places such as supermarkets, convenience stores, warehouses and the like often need goods shelves to display various articles. In order to facilitate the management of articles, it is generally necessary for a manager to know the display condition of the articles on the shelf and to perform operations such as replenishment and arrangement in time.

At present, goods shelves in places such as supermarkets and the like usually rely on manual work to complete tasks such as classification, recording and the like, so as to obtain goods display information, and further effectively manage goods according to the goods display information.

However, because of the variety of articles, the labor is used to identify and classify a large number of articles, which results in high economic and time costs and an uncertain article identification accuracy. Once the items on the shelves change, the updating of the item display information is inefficient.

Disclosure of Invention

The invention provides a method, a device, equipment and a medium for updating article display information, which are used for solving the problem of low updating efficiency of the article display information, effectively improving the updating efficiency of the article display information and reducing the management cost while ensuring the accuracy rate of article identification.

According to an aspect of the present invention, there is provided an update method of article display information, the method comprising:

acquiring a target shelf image, and detecting the target shelf image based on a target detection model to obtain at least one article single body detection result;

determining at least one group of article single type detection results according to at least one article single type detection result;

and updating the article display information of the target shelf according to the article single detection result and the article single type detection result.

According to another aspect of the present invention, there is provided an apparatus for updating information on display of an article, the apparatus comprising:

the single detection result determining module is used for acquiring a target shelf image and detecting the target shelf image based on a target detection model to obtain at least one article single detection result;

the single-type detection result determining module is used for determining at least one group of single-type detection results of the articles according to the single-type detection result of the at least one article;

and the article display information updating module is used for updating the article display information of the target shelf according to the article single detection result and the article single type detection result.

According to another aspect of the present invention, there is provided an electronic apparatus including:

at least one processor; and

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

the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to enable the at least one processor to perform a method of updating item display information according to any embodiment of the invention.

According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the method for updating item display information according to any one of the embodiments of the present invention when the computer instructions are executed.

According to the technical scheme of the embodiment of the invention, the target shelf image is obtained, and the target shelf image is detected based on the target detection model, so that at least one article single body detection result is obtained; determining at least one group of article single type detection results according to at least one article single type detection result; and updating the article display information of the target shelf according to the article single detection result and the article single type detection result. The scheme can solve the problem of low updating efficiency of the article display information, effectively improves the updating efficiency of the article display information and reduces the management cost while ensuring the article identification accuracy.

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

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1A is a flowchart of a method for updating information on an article display according to an embodiment of the present invention;

FIG. 1B is a schematic illustration of an item display shelf provided in accordance with an embodiment of the present invention;

fig. 2 is a flowchart of an updating method of the article display information according to a second embodiment of the present invention;

fig. 3 is a schematic structural diagram of an apparatus for updating article display information according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device implementing the method for updating article display information according to the embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. According to the technical scheme, the data acquisition, storage, use, processing and the like meet relevant regulations of national laws and regulations.

Example one

Fig. 1A is a flowchart of an embodiment of the present invention, which provides a method for updating article display information, where the embodiment is applicable to a case of updating article display information, and the method may be performed by an article display information updating apparatus, which may be implemented in a form of hardware and/or software, and the apparatus may be configured in an electronic device. As shown in fig. 1A, the method includes:

s110, obtaining a target shelf image, and detecting the target shelf image based on a target detection model to obtain at least one article single body detection result.

The scheme can be executed by an updating platform, and the updating platform can respond to an updating request of the article display information initiated by a user to acquire the target shelf image. The update request of the article display information may be a store patrol request initiated by the user through the user terminal, or may be that the user uploads the target shelf image. The updating platform can detect the target shelf image based on the target detection model according to request information of the user terminal or user registration information and the like.

The target detection model may be obtained by training based on a public article data set, or may be obtained by training based on a self-built article data set. The target detection model can be a model constructed based on a one-stage target detection algorithm such as YOLO, SSD and the like, and can also be a model constructed based on a two-stage target detection algorithm such as Faster R-CNN and the like. It should be noted that the target detection model may implement dense target detection, and may detect each item in the target shelf image.

After detecting the item singles in the target shelf image, the update platform may generate an item singles detection result. The article single body detection result can comprise each single body position, single body type, each single body detection frame coordinate, each single body detection accuracy rate and the like. Fig. 1B is a schematic diagram of an article display shelf provided according to an embodiment of the present invention, where the article display shelf in fig. 1B is used as a target shelf image, a single detection box may be as shown in the detection box 1 in fig. 1B, and an update platform may construct a coordinate system according to the target shelf image, and record a center coordinate and/or each vertex coordinate of the single detection box to describe a location of a single article. The updating platform can intercept the target shelf image according to the single detection frame so as to obtain each single image.

And S120, determining at least one group of item single type detection results according to at least one item single type detection result.

It can be understood that in the scenes of supermarkets, warehouses, vending machines and the like, in order to facilitate management, management personnel usually need to know the distribution situation of the single articles of the same category. According to the detection result of each single article in the target shelf image, the updating platform can obtain the detection result of each single article according to the detection result of the single article corresponding to the single article in the same category. For example, the update platform may map the individual article detection results of the individual articles in the same category to the individual article detection results according to the individual category. Similarly, the item single type detection result may include an item type, a distribution position of like items, an item single type detection frame, and the like.

And S130, updating the article display information of the target shelf according to the article single detection result and the article single type detection result.

According to the detection result of the single article, the distribution condition of the single article on the target shelf can be obtained by the updating platform. Similarly, according to the detection result of the item single type, the update platform can determine the distribution condition of each type of item on the target shelf. The update platform can update the item display information of the target shelf according to the distribution condition of the single item and the distribution condition of each category of items. Wherein the article display information may be a shed cut chart or a shed cut table of the target shelf.

Specifically, the update platform may determine the number of layers of the target shelf where the article single body is located according to the position of the single body image in the article single type detection result. As shown in fig. 1B, the positions of the articles on the same layer on the article display shelf should be within the same coordinate range, and the update platform may perform cluster analysis on the center coordinates of each individual article detection frame, so as to obtain the layer number distribution of the articles on the target shelf.

In this embodiment, optionally, the article single body detection result includes a single body image;

the determining at least one group of item single type detection results according to at least one item single type detection result comprises:

extracting the characteristics of each single image based on the characteristic extraction model;

and determining at least one group of item single-type detection results according to the feature extraction results of the single images and the similarity comparison results of the features in the pre-acquired item feature set.

As can be easily understood, in order to realize the refined identification of the single image, the updating platform may perform feature extraction on the single image captured from the target shelf image based on the feature extraction model. The feature extraction model can be a feature extraction model constructed based on deep learning algorithms such as a convolutional neural network and the like, and can also be a feature extraction model constructed based on traditional graphics algorithms such as scale-invariant feature transformation and the like. After the feature extraction is performed on the single image, the updating platform may obtain a feature extraction result of the single image, which may be a feature vector, for example.

The update platform may pre-build an item feature set, which may include a plurality of features to cover all item features required by the application scenario. The updating platform can compare the feature extraction result of each single image with each feature in the article feature set to determine the category to which each single image belongs, and further determine at least one group of article single detection results according to each single category. Specifically, the updating platform may sequentially calculate the feature extraction result of the single image and the euclidean distance, the cosine distance, and the like of the features in the article feature set to represent the similarity between the two. And comparing the similarity of the same single image, and selecting the article classification associated with the feature with the highest similarity as the target classification of the single image.

Assuming that 3 single objects are detected from the target shelf image, 3 single object detection results are obtained, wherein the object feature set comprises 4 features, and each feature is associated with one object classification. The updating platform can calculate the similarity of the feature vector of the single image 1 with the feature 1, the feature 2, the feature 3 and the feature 4 to obtain the similarity 1-1, the similarity 1-2, the similarity 1-3 and the similarity 1-4. Similarly, the update platform may calculate the similarity of the feature vectors of the individual images 2 and 3 to the features 1, 2, 3, and 4. The updating platform can determine the target classification of the single image 1 by comparing the similarity 1-1, the similarity 1-2, the similarity 1-3 and the similarity 1-4. Assuming that the similarity 1-1, the similarity 1-2, the similarity 1-3 and the similarity 1-4 are 0.1, 0.4,0.3 and 0.9, respectively, the article classification associated with the feature 4 is taken as the target classification of the single image 1.

According to the technical scheme, a target shelf image is obtained, and the target shelf image is detected based on a target detection model, so that at least one article single body detection result is obtained; determining at least one group of article single type detection results according to at least one article single type detection result; and updating the article display information of the target shelf according to the article single detection result and the article single type detection result. The scheme can solve the problem of low updating efficiency of the article display information, effectively improves the updating efficiency of the article display information and reduces the management cost while ensuring the article identification accuracy.

Example two

Fig. 2 is a flowchart of an updating method of article display information according to a second embodiment of the present invention, which is detailed based on the above-mentioned embodiments. As shown in fig. 2, the method includes:

s210, obtaining a target shelf image, and detecting the target shelf image based on a target detection model to obtain at least one article single body detection result; the article single body detection result comprises a single body image and a single body detection frame.

In this embodiment, the item type detection result may include item type information and type detection frame information.

And S220, extracting the features of the single images based on the feature extraction model.

And S230, respectively determining the similarity between the feature extraction result of each single image and the features in the article feature set, and respectively sequencing the similarity associated with each single image to obtain a similarity sequencing result.

Suppose that 3 single objects are detected from the target shelf image to obtain 3 single object detection results, the object feature set comprises 4 features, and each feature is associated with one object classification. The updating platform can calculate the similarity of the feature vector of the single image 1 with the feature 1, the feature 2, the feature 3 and the feature 4 to obtain the similarity 1-1, the similarity 1-2, the similarity 1-3 and the similarity 1-4. Similarly, the update platform can obtain the similarity 2-1, the similarity 2-2, the similarity 2-3 and the similarity 2-4, and also can obtain the similarity 3-1, the similarity 3-2, the similarity 3-3 and the similarity 3-4. The updating platform can obtain the similarity arrangement result of the single image 1 by comparing the similarity 1-1, the similarity 1-2, the similarity 1-3 and the similarity 1-4. Assuming that the similarity 1-1, the similarity 1-2, the similarity 1-3, and the similarity 1-4 are 0.1, 0.4,0.3, and 0.9, respectively, the similarity ranking result of the single image 1 may be: 0.9,0.4,0.3,0.1.

And S240, determining the article type information of each single image according to the similarity sorting result.

The updating platform can select the article category associated with the feature with the largest similarity from the article feature set as the classification result of the single image according to the similarity ranking result. The updating platform can also select the article category associated with part of the features in the article feature set as the classification result of the single image according to the similarity sorting result. Under the condition that hardware conditions such as a server and a storage device allow, the updating platform can also keep the article categories associated with all the features as the classification results of the single images according to the similarity ranking results.

As an example in S230, the classification result of the single image 1 may be an article classification corresponding to the feature 4, an article classification corresponding to the feature 4 may be a target classification, an article classification corresponding to the feature 2 may be a candidate classification, an article classification corresponding to the feature 4 may be a target classification, and article classifications corresponding to the features 2, 3, and 1 are used as candidate classification lists. The updating platform can generate corresponding article category information according to the classification result.

And S250, merging the single detection frames corresponding to the single images with the same article type information into a single detection frame, and generating single detection frame information.

On the article display shelves, the same kind of articles are usually placed in adjacent positions for easy management. The updating platform can compare the article type information of each single image, and if the article type information is the same, the single images can be determined to be of the same type. The update platform may merge the single detection frames of the same category to obtain a single detection frame, and the single detection frame may be shown as a detection frame 2, a detection frame 3, and a detection frame 4 in fig. 1B. According to the single-type detection frame, the updating platform can generate single-type detection frame information. The single-type detection frame information may include information such as a position of a single-type detection frame and a number of merged single detection frames.

And S260, updating the article display information of the target shelf according to the article type information, the single type detection frame information and the article single body detection result.

In a possible solution, the determining the item category information of each single image according to the similarity ranking result includes:

determining the target classification and the target classification confidence of each single image according to the similarity sorting result;

and determining the article category information of each single image according to the target classification and the target classification confidence.

In this embodiment, the target classification confidence may be determined based on the highest similarity, and still taking the assumption in S230 as an example, the target classification of the single image 1 is the category a corresponding to the feature 4, and the target classification confidence may be 0.9. The confidence of the target classification may also be obtained based on the similarity ranking result, for example, the update platform may calculate the similarity proportion of the target classification according to the similarity ranking result

The update platform may generate item class information for the monoscopic image based on the target classification and the target classification confidence.

According to the scheme, the target classification confidence can be determined, the classification result can be optimized according to the reliability evaluation of the target classification, and the characteristic extraction model can be trained pertinently according to the target classification confidence, so that the performance of the characteristic extraction model is improved.

On the basis of the scheme, the single detection frame information comprises the number of the merged single detection frames;

after determining at least one group of item singles detection results, the method further comprises:

if the number of the merged single detection frames is 1, determining at least one similarity between a single image in the single detection frame and a target single image in an adjacent single detection frame;

if at least one similarity is higher than a preset similarity threshold and the target classification confidence of the single image is lower than a preset confidence threshold, merging the single detection frame into an adjacent single detection frame with the highest similarity, and determining article type correction information;

and updating the item feature set according to the item type correction information.

If the number of the combined single detection boxes is 1, the single detection boxes only have one single, and the single existence is probably classified by mistake. The updating platform can determine whether the single body and the adjacent single body belong to the same category by calculating the similarity of the single body image in the single type detection frame and the target single body image in the adjacent single type detection frame. If at least one similarity is higher than a preset similarity threshold value and the target classification confidence coefficient of the single image is lower than a preset confidence coefficient threshold value, it is indicated that the original target classification confidence coefficient of the single image is low, the similarity between the single image and an adjacent single image is high, and the single image is misclassified. Therefore, the updating platform can combine the single-type detection frame into the adjacent single-type detection frame with the highest similarity and determine the item type correction information.

The item type correction information may include a single image to be corrected, a correction reason, a correction type, a correction target reference image, and the like. The updating platform can add the item type correction information to the item feature set and associate the item type correction information with the target classification, so that the classification accuracy is improved in the subsequent using process. In order to ensure the updating accuracy, the updating platform can send the item type correction information to the user terminal, and determine whether to update the item feature set according to the checking result of the user terminal.

The scheme can effectively identify and correct the misclassification condition, and is favorable for ensuring the accuracy of the article classification result.

In another possible solution, the determining the item category information of each single image according to the similarity ranking result includes:

determining a target classification and at least one candidate classification of each single image according to the similarity sorting result, and determining a target classification confidence coefficient and each candidate classification confidence coefficient;

and determining the article category information of each single image according to the target classification, the target classification confidence coefficient, each candidate classification and each candidate classification confidence coefficient.

It is to be understood that, similar to the target classification confidence, the candidate classification confidence may be determined based on the similarity between the features associated with each candidate classification and the single image or may be obtained based on the similarity ranking result of the single image. The update platform may generate a corresponding classification list for each single image according to the similarity ranking result, where table 1 below is a classification list of the single image 1.

Table 1:

sort ordering	Categories of articles	Confidence level
			Object classification	Class A	0.9
Candidate classification 1	Class B	0.4
			Candidate classification 2	Class C	0.3
Candidate classification 3	Class D	0.1

And according to the target classification, the candidate classification, the target classification confidence coefficient and the candidate classification confidence coefficient information in the classification list, the updating platform can generate the article class information of the single image. The article type information may include information such as unit price information and specification attributes of each type of article.

According to the scheme, at least one candidate classification and candidate classification confidence coefficient can be determined according to the similarity sorting result, and the target classification information can be corrected according to the candidate classification information, so that the classification reliability is guaranteed.

On the basis of the scheme, the single-type detection frame information further comprises a unit price for identification; the identification unit price is determined based on the character identification result of the price tag image in the single-class detection frame;

if the identification unit price is not matched with the unit price of the object classified article, determining article type correction information according to the unit price of the candidate classified article and the confidence coefficient of the candidate classification;

As shown in fig. 1B, when detecting the target shelf image, the price tag on the shelf may be detected, and the price tag image may be obtained according to the detection result. In the process of merging the single detection frames of the same type of single images, the single detection frames can also merge price tag images. And performing character recognition on the price tag image through a character recognition model, and obtaining the recognition unit price of the article type in the single-type detection frame by the updating platform.

If the unit price of the identified item is inconsistent with the unit price of the object classified or the difference is larger than a certain price threshold value, the identified item is possibly a new item or the single image classification of the item is wrong. The update platform may compare the item unit price of the candidate classification with the identification unit price. And the update platform selects the candidate classification with the highest confidence coefficient and the smallest difference between the unit price of the article and the unit price of the identification as the replacement classification of the target classification, and generates article class correction information.

The updating platform can send the item type correction information to the user terminal, and determine whether to update the item feature set according to the proofreading result of the user terminal. If the item is a new item, the update platform may add new item features to the set of item features. If the classification of the single image in the category is wrong, the updating platform can correct the features in the article feature set.

The scheme can check the conditions of error classification, new product shelving and the like, can avoid operation loss as much as possible, and is favorable for continuously perfecting the characteristic set of the articles so as to improve the classification accuracy.

It should be noted that the update platform may perform statistical analysis on all item type correction information to determine correction data of each item type. The correction data may include information such as the number of corrections and the probability of correction. According to the correction data, the updating platform can expand the single image to be corrected to the data set of the feature extraction model, so that the specific diagnosis extraction model can be trained specifically, and the feature extraction effectiveness is improved.

EXAMPLE III

Fig. 3 is a schematic structural diagram of an apparatus for updating information on display of an article according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes:

the single detection result determining module 310 is configured to obtain a target shelf image, and detect the target shelf image based on a target detection model to obtain at least one article single detection result;

the single-type detection result determining module 320 is configured to determine at least one group of single-type detection results of the article according to at least one single-article detection result;

and an article display information updating module 330, configured to update the article display information of the target shelf according to the article single detection result and the article single type detection result.

the single-class detection result determining module 320 includes:

the characteristic extraction unit is used for extracting the characteristics of each single image based on the characteristic extraction model;

and the single-type detection result determining unit is used for determining at least one group of single-type detection results of the articles according to the feature extraction result of each single image and the similarity comparison result of the features in the pre-acquired article feature set.

On the basis of the above scheme, optionally, the article single body detection result further includes a single body detection frame; the article single-type detection result comprises article type information and single-type detection frame information;

the single-type detection result determining unit comprises:

the similarity sequencing result determining subunit is used for respectively determining the similarity between the feature extraction result of each single image and the features in the article feature set, and respectively sequencing the similarity associated with each single image to obtain a similarity sequencing result;

the article type information determining subunit is used for determining the article type information of each single image according to the similarity sorting result;

and the single-type detection frame information generation subunit is used for merging the single detection frames corresponding to the single images with the same article type information into a single-type detection frame and generating single-type detection frame information.

In a possible solution, the item category information determining subunit is specifically configured to:

the device further comprises:

the similarity determining module is used for determining at least one similarity between a single image in the single detection frame and a target single image in an adjacent single detection frame if the number of the merged single detection frames is 1;

the first correction information determining module is used for merging the single-type detection frame into an adjacent single-type detection frame with the highest similarity and determining article type correction information if at least one similarity is higher than a preset similarity threshold and the target classification confidence of the single image is lower than a preset confidence threshold;

and the first item feature set updating module is used for updating the item feature set according to the item type correction information.

In another possible solution, the item category information determining subunit is specifically configured to:

On the basis of the scheme, the single-type detection frame information further comprises a unit price for identification; the identification unit price is determined based on the character identification result of the price tag image in the single-type detection frame;

the device further comprises:

a second correction information determining module, configured to determine item type correction information according to the item unit price of each candidate classification and each candidate classification confidence if the identification unit price does not match the item unit price of the target classification;

and the second article characteristic set updating module is used for updating the article characteristic set according to the article type correction information.

The updating device of the article display information provided by the embodiment of the invention can execute the updating method of the article display information provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

Example four

FIG. 4 shows a schematic block diagram of an electronic device 410 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 4, the electronic device 410 includes at least one processor 411, and a memory communicatively connected to the at least one processor 411, such as a Read Only Memory (ROM) 412, a Random Access Memory (RAM) 413, and the like, wherein the memory stores computer programs executable by the at least one processor, and the processor 411 may perform various appropriate actions and processes according to the computer programs stored in the Read Only Memory (ROM) 412 or the computer programs loaded from the storage unit 418 into the Random Access Memory (RAM) 413. In the RAM 413, various programs and data necessary for the operation of the electronic device 410 can also be stored. The processor 411, ROM 412, and RAM 413 are connected to each other by a bus 414. An input/output (I/O) interface 415 is also connected to bus 414.

Various components in the electronic device 410 are connected to the I/O interface 415, including: an input unit 416 such as a keyboard, a mouse, or the like; an output unit 417 such as various types of displays, speakers, and the like; a storage unit 418, such as a magnetic disk, optical disk, or the like; and a communication unit 419 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 419 allows the electronic device 410 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

Processor 411 can be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The processor 411 performs the various methods and processes described above, such as the update method of the item display information.

In some embodiments, the method of updating item display information may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 418. In some embodiments, part or all of the computer program may be loaded and/or installed onto electronic device 410 via ROM 412 and/or communications unit 419. When loaded into RAM 413 and executed by processor 411, may perform one or more of the steps of the above-described method of updating item display information. Alternatively, in other embodiments, the processor 411 may be configured by any other suitable means (e.g., by way of firmware) to perform the update method of the item display information.

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

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

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

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

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

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

It should be understood that various forms of the flows shown above, reordering, adding or deleting steps, may be used. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired result of the technical solution of the present invention can be achieved.

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

Claims

1. A method of updating article display information, the method comprising:

2. The method of claim 1, wherein the item cell detection result comprises a cell image;

the determining at least one group of article single type detection results according to at least one article single type detection result comprises:

and determining at least one group of item single-type detection results according to the feature extraction result of each single image and the similarity comparison result of the features in the pre-acquired item feature set.

3. The method of claim 2, wherein the item cell detection result further comprises a cell detection box; the item single type detection result comprises item type information and single type detection frame information;

the determining at least one group of item single-type detection results according to the feature extraction results of the single images and the similarity comparison results of the features in the pre-acquired item feature set comprises the following steps:

respectively determining the similarity between the feature extraction result of each single image and the features in the article feature set, and respectively sequencing the similarity associated with each single image to obtain a similarity sequencing result;

determining the article category information of each single image according to the similarity sorting result;

and merging the single detection frames corresponding to the single images with the same article type information into a single detection frame, and generating single detection frame information.

4. The method according to claim 3, wherein the determining the item category information of each single image according to the similarity ranking result comprises:

5. The method of claim 3, wherein the single type of frame information comprises a number of merged single frames;

if at least one similarity is higher than a preset similarity threshold value and the target classification confidence of the single image is lower than a preset confidence threshold value, merging the single detection frame into an adjacent single detection frame with the highest similarity, and determining article type correction information;

6. The method according to claim 3, wherein the determining the item category information of each single image according to the similarity ranking result comprises:

7. The method according to claim 6, wherein the one-class detection frame information further includes an identification unit price; the identification unit price is determined based on the character identification result of the price tag image in the single-type detection frame;

8. An apparatus for updating information on an article display, the apparatus comprising:

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of updating item display information of any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions for causing a processor to perform the method of updating item display information of any one of claims 1-7 when executed.