CN110263758B - Method and system for detecting opening and closing of physical store - Google Patents
Method and system for detecting opening and closing of physical store Download PDFInfo
- Publication number
- CN110263758B CN110263758B CN201910584462.5A CN201910584462A CN110263758B CN 110263758 B CN110263758 B CN 110263758B CN 201910584462 A CN201910584462 A CN 201910584462A CN 110263758 B CN110263758 B CN 110263758B
- Authority
- CN
- China
- Prior art keywords
- physical
- business
- state
- door
- door opening
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/35—Categorising the entire scene, e.g. birthday party or wedding scene
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for detecting business states of physical stores. According to the disclosed method for detecting the business state of the physical store, an AI detection model is used for detecting the store scene image collected at regular time, the door opening and closing state of the physical store is determined, and then whether the door opening and closing state is consistent with the preset door opening and closing state corresponding to the collection time is compared, and whether the business state of the physical store is normal is determined, so that the aim of automatically supervising all the physical stores is fulfilled.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a method and a system for detecting opening and closing of a physical store.
Background
With the advent of more and more, small-scale, distributed, and same brand of retail stores that operate on the same type of goods and services. The unified manager of the retail store can supervise the business state of the storefronts through the network cameras installed in all the storefronts under the flags.
In the prior art, due to the fact that the quantity of chain stores is huge, the quantity of shop scene images collected by the network cameras of all the shops is huge, and a manager cannot judge the business states of all the shops according to the shop scene images of all the shops so as not to supervise all the shops, therefore, when facing a large quantity of image videos uploaded by the network cameras of the shops, the manager can only carry out spot check on the image videos of part of the shops, judge whether the part of the shops are in normal business states through the spot check image videos, and finally estimate the shops in abnormal business states of all the shops through the abnormal business states obtained by sampling, and therefore the purpose of supervising the shops is achieved.
However, the number of chain stores is huge and the spot check is random, so that the prior art method cannot comprehensively monitor the business state of the store.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and a system for detecting business states of physical stores, which detect store scene images collected at regular time and determine whether the business states of the stores are normal according to a detection result, so as to achieve the purpose of automatically monitoring the stores.
In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions:
the invention discloses a method for detecting the business state of an entity store in a first aspect, which comprises the following steps:
acquiring storefront scene images of all physical storefronts at fixed time to obtain storefront scene image information of the physical storefronts, wherein the storefront scene image information comprises acquisition time for acquiring the storefront scene images;
detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information;
comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time;
if the business state is consistent with the business state, determining that the physical store is in a normal business state;
and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state.
Preferably, the AI detection model establishing process includes:
inputting a training sample into an initial network training model for training to obtain a door opening and closing result of the physical store, and comparing the door opening and closing result with a door opening and closing result corresponding to the training sample, wherein the training sample is store door scene image information of the physical store with a predetermined door opening and closing result;
if the comparison results are the same, obtaining the weight of the initial network training model, and establishing an AI detection model based on the weight;
if the comparison results are different, the weight of the initial network training model is adjusted, a new network training model is continuously trained based on the training samples until the comparison results are the same, and the weight of the current network training model is obtained;
and establishing an AI detection model based on the weight.
Preferably, the detecting the shop front scene image information based on the pre-established AI detection model and determining the door opening and closing state of the physical shop corresponding to the shop front scene image information includes:
inputting the shop door scene image into a pre-established AI detection model;
the AI detection model acquires all image characteristics in the shop door scene image information, analyzes all the characteristics, determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door opening state is in the door opening state, and determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door closing state is in the door closing state.
Preferably, the method further comprises the following steps:
the method comprises the steps of obtaining business states corresponding to all physical stores, classifying the physical stores according to normal business states and abnormal business states, and storing classification results;
and counting the classification results according to preset time to generate business state reports of all physical stores.
Preferably, after the counting the classification results according to the preset time and generating the business state reports of all the physical stores, the method further includes:
and sending the business state report to a preset mailbox.
The second aspect of the present invention discloses a door opening and closing detection system for physical stores, comprising:
the system comprises an image acquisition device, a processing device and a display device, wherein the image acquisition device is used for acquiring storeroom scene images of all physical storerooms at regular time to obtain storeroom scene image information of the physical storerooms, and the storeroom scene image information comprises acquisition time for acquiring the storeroom scene images;
the detection module is used for detecting the shop door scene image information based on a pre-established AI detection model and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information;
the comparison module is used for comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state.
Preferably, the method further comprises the following steps:
the training module is used for inputting a training sample into an initial network training model for training to obtain a door opening and closing result of the physical store, and comparing the door opening and closing result with a door opening and closing result corresponding to the training sample, wherein the training sample is store door scene image information of the physical store with a predetermined door opening and closing result;
the acquisition module is used for acquiring the weight of the initial network training model if the comparison results are the same, and establishing an AI (artificial intelligence) detection model based on the weight;
the adjusting module is used for adjusting the weight of the initial network training model if the comparison results are different, continuing to train a new network training model based on the training samples until the comparison results are the same, and acquiring the weight of the current network training model;
and the establishing module is used for establishing an AI detection model based on the weight.
Preferably, the detection module includes:
the input unit is used for inputting the shop door scene image into a pre-established AI detection model;
and the detection unit is used for acquiring all the characteristics in the shop door scene image information by the AI detection model, analyzing all the characteristics, determining that the physical shop corresponding to the shop door scene image with the image characteristic representing the door opening state is in the door opening state, and determining that the physical shop corresponding to the shop door scene image with the image characteristic representing the door closing state is in the door closing state.
Preferably, the method further comprises the following steps:
the classification module is used for acquiring business states corresponding to all physical stores, classifying all the physical stores according to normal business states and abnormal business states, and storing classification results;
and the counting module is used for counting the classification results according to the preset time and generating business state reports of all the physical stores.
Preferably, the method further comprises the following steps:
and the sending module is used for sending the business state report to a preset mailbox.
From the above, the application discloses a method and a system for detecting business states of physical stores, wherein the method comprises the steps of acquiring store scene images of all the physical stores at regular time to obtain the store scene image information of the physical stores; detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information; comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state. According to the method for detecting the business state of the physical store, the AI detection model is used for detecting the store scene image collected at regular time, so that the door opening and closing state of the physical store is determined, and then the door opening and closing state is compared with the preset door opening and closing state corresponding to the collection time to determine whether the business state of the physical store is normal or not, and the purpose of automatically supervising all the physical stores is achieved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1a is a schematic structural diagram of a system for detecting business states of physical stores according to an embodiment of the present invention;
fig. 1b is a flowchart of a method for detecting business status of a physical store according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for detecting the business status of a physical store according to an embodiment of the present invention;
FIG. 3 is a flowchart of another method for detecting business status of physical stores according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating another method for detecting business status of a physical store according to an embodiment of the present invention;
FIG. 5 is a flowchart illustrating another method for detecting business status of physical stores according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of another physical store business status detection system according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of another physical store business status detection system according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of another physical store business status detection system according to an embodiment of the present invention.
Detailed Description
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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The application provides a method and a system for detecting opening and closing of physical stores, which can realize supervision on each physical store when a manager faces a large number of physical stores, and improve the management efficiency of the manager on the stores. The application discloses a physical store door opening and closing detection system shown in fig. 1a, which is composed of a camera 100 and a processor 101.
The camera 100 is installed in each physical store, the processor 101 is arranged in the background or cloud, and the processor 101 is connected with the camera 100 through a network. The camera 100 can acquire the store door scene image information of the physical store and transmit the acquired store door scene image information to the processor 101 through the network. The processor 101 processes the shop-door scene image information to finally obtain a door opening and closing state of the physical shop corresponding to the shop-door scene image information, and determines whether the business state of the physical shop is normal or not based on the door opening and closing state, so as to achieve the purpose of automatically supervising all shops.
Based on the physical store door opening and closing detection system disclosed in fig. 1a, a specific embodiment of the present invention provides a physical store door opening and closing detection method, referring to fig. 1b, the method at least includes the following steps:
step S101: and acquiring the storeroom scene images of all the physical storerooms at fixed time to obtain the storeroom scene image information of the physical storerooms.
It should be noted that the store door scene image information includes acquisition time for acquiring the store door scene image.
In step S101, the step of regularly acquiring the store scene images of all the physical stores means that an image acquisition device acquires the store scene images of all the physical stores according to a preset time, where the image acquisition device is equivalent to the camera 100 shown in fig. 1a, but is not limited thereto, and the image acquisition device may also be a monitoring device such as a web camera or a monitor with an image taking function. In the present application, the preset shop-door scene image capturing time is set according to the shop operating hours in different areas.
For example, if the business hours of a brick-and-mortar store in a certain region are from 9 am to 19 pm, the preset store scene image acquisition time should be determined within the business hours, the early store scene image acquisition time of the brick-and-mortar store may be 9 am, or 1 min after 9 am, or the late store scene image acquisition time of the brick-and-mortar store may be 18 pm, 58 min, or 19 pm.
Step S102: and detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information.
In step S102, the door opening and closing state includes a door opening state and a door closing state.
The AI detection model has strong image feature analysis capability and is a model capable of obtaining the door opening and closing state of the physical store corresponding to the shop-door scene image information according to the image features in the shop-door scene image information, so that the AI detection model can detect the image characteristics in the shop-door scene image information and determine the door opening and closing state of the physical store corresponding to the shop-door scene image information according to the detection result.
In the process of executing step S102, as shown in fig. 2, the specific execution process includes the following steps:
step S201: and inputting the shop door scene image into a pre-established AI detection model.
Step S202: the AI detection model acquires all the characteristics in the shop door scene image information, analyzes all the characteristics, determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door opening state is in the door opening state, and determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door closing state is in the door closing state.
It should be noted that the image information in the store door scene image information is composed of a plurality of image features, the AI detection model obtains the plurality of image features and then summarizes the image features to obtain a summarized image feature, and determines whether the physical store represented by the summarized image feature is in an open door state or a closed door state based on the summarized image feature.
It should be noted that, the specific determination is completed by the classification probability indicator, and the specific implementation process is as follows: and performing door opening and closing state classification analysis based on the summarized image features, determining that the summarized image features representing the door opening state and having high image feature probability belong to the door opening state classification, and determining that the summarized image features representing the door closing state and having high image feature probability belong to the door closing state classification.
Step S103: and comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time, if so, executing the step S104, and if not, executing the step S105.
It should be noted that, because the physical store door can be divided into a door opening state and a door closing state, the preset door opening/closing state can be set as the door opening state, and the preset door opening/closing state can also be set as the door closing state.
Preferably, in this application, the preset door opening and closing state is a door opening state.
Step S104: determining that the physical store is in a normal business state.
Step S105: determining that the physical store is in an abnormal business state.
For ease of understanding, the following examples are given.
The preset door opening and closing state is the door opening state, and taking 24-hour time as an example, when the acquisition time is 8 hours, the physical store a is the door opening state, and the door opening and closing state of the physical store a is consistent with the preset door opening and closing state, it is indicated that the physical store a opens the door on time for business, that is, the physical store a is in the normal business state.
The physical store B is closed, and the door opening and closing state of the physical store B is inconsistent with the preset door opening and closing state, which indicates that the physical store B is not opened on time for business, namely, the physical store B is in an abnormal business state.
If the acquisition time is 20 o' clock, the physical store C is in a door closing state, and the door opening and closing state of the physical store C is inconsistent with the preset door opening and closing state, it is indicated that the physical store C closes the store door to leave the office in advance, that is, the physical store C is in an abnormal business state.
It should be noted that, the steps S102 to S105 can be implemented by the processor in fig. 1 a.
The method comprises the steps of acquiring shop-door scene images of all physical shops at regular time to obtain shop-door scene image information of the physical shops; detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information; comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state. According to the shop business state detection method, the AI detection model is used for detecting the shop scene images collected at regular time, so that the door opening and closing state of the physical shop is determined, and then whether the door opening and closing state is consistent with the preset door opening and closing state corresponding to the collection time is compared, and whether the business state of the physical shop is normal is determined, so that the aim of automatically supervising all the physical shops is fulfilled.
Based on the method for detecting business states of physical stores disclosed in the foregoing embodiment, in step S102 shown in fig. 1, the process of establishing the AI detection model specifically includes the following steps, as shown in fig. 3:
step S301: inputting a training sample into an initial network training model for training to obtain a door opening and closing result of the physical store, comparing the door opening and closing result with a door opening and closing result corresponding to the training sample, if the comparison results are the same, executing a step S302, and if the comparison results are different, executing a step S303.
The training sample is shop door scene image information of the physical shop with a predetermined door opening and closing result.
In step S301, the training sample is input into the initial network training model for training, that is, the store-door scene image information is input into the initial network training model for training, and the initial network training model performs processing according to the image features of the store-door scene image information, so as to finally obtain the door opening and closing result of the physical store corresponding to the store-door scene image information.
It should be noted that the initial network training model is a primary network training model with an image feature recognition capability, and the establishment efficiency of the AI detection model can be accelerated by selecting the primary network training model with the image feature recognition capability to train the training sample.
Optionally, a network structure with image feature recognition capability may be selected to train the training sample, so as to achieve the purpose of establishing the AI detection model.
Step S302: and acquiring the weight of the initial network training model, and establishing an AI detection model based on the weight.
It should be noted that, because the door opening and closing result corresponding to the training sample is consistent with the result obtained by training the training sample through the initial network training model, it is indicated that the training result obtained after the initial network training model trains the training sample is an expected result, and therefore, the weight of the initial network training model is directly obtained, and the AI detection model is established based on the obtained weight.
Step S303: and adjusting the weight of the initial network training model, continuing training a new network training model based on the training samples until the comparison results are the same, and acquiring the weight of the current network training model.
Step S304: and establishing an AI detection model based on the weight.
In the process of executing steps S303 to S304, the weight is an adjustable parameter of the initial network training model, and the output result of the training sample input to the initial network training model can be changed by adjusting the weight.
It should be noted that the result of opening and closing the door corresponding to the training sample is not consistent with the result of training the training sample through the initial network training model, which indicates that the initial network training model cannot obtain the expected effect according to the image features in the training sample, and therefore, the weight parameter of the initial network training model needs to be adjusted repeatedly until the result of opening and closing the door corresponding to the training sample is consistent with the result of training the training sample through the network training model by adjusting the weight parameter repeatedly.
Based on the execution process of the steps S301 to S304, the established AI detection model includes a classification probability indicator, which is used to obtain a plurality of image features and then summarize the image features, and perform classification analysis on the open/close state based on the summarized image features, determine that the summarized image features with high probability of the image features representing the open-door state belong to the open-door state classification, and determine that the summarized image features with high probability of the image features representing the closed-door state belong to the closed-door state classification.
Referring to fig. 4, for another method for detecting business status of a physical store provided in the embodiment of the present application, the method for detecting business status of a physical store includes the following steps:
step S401: and acquiring the storeroom scene images of all the physical storerooms at fixed time to obtain the storeroom scene image information of the physical storerooms.
Step S402: and detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information.
Step S403: and comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time, if so, executing the step S404, and if not, executing the step S405.
Step S404: determining that the physical store is in a normal business state.
Step S405: determining that the physical store is in an abnormal business state.
It should be noted that the execution principle and the specific execution process of steps S401 to S405 are the same as the execution principle and the specific execution process of steps S101 to S105 shown in fig. 1, and reference may be made to the above corresponding description, which is not described herein again.
Step S406: and acquiring the business states corresponding to all the physical stores, classifying all the physical stores according to the normal business states and the abnormal business states, and storing the classification result.
It should be noted that, here, all the physical stores are classified according to normal business states and abnormal business states, which is convenient for a manager to count and check the business states of all the physical stores temporarily.
Step S407: and counting the classification results according to preset time to generate business state reports of all physical stores.
In step S407, the preset time refers to counting the classification results at intervals, where the preset time may be monday of each week, number 1 of each month, or manually changing the preset time in a certain time interval by the administrator, and is not limited in this application.
And counting the classification processing results, and storing the counting results and the calculation results into a table by counting and calculating the classification processing results, namely generating a business state report of the physical store.
The business state report can be in a form of a table, and can also be in a form of combining pictures and texts, and in the application, the form of the business state report is not limited.
For ease of understanding, the following examples are given.
The business states of all the physical stores in the previous month are counted by 1 monthly, when the business state is 1 monthly, the business states of 100 stores in 31 days of 5 months are counted, at 8, 10 physical stores have at least 1 non-opening-time business in the 31 days, and at 22, 5 physical stores have at least 1 closing-time in the 31 days.
Step S408: and sending the business state report to a preset mailbox.
In step S408, there may be one or more predetermined mailboxes.
It should be noted that, the business state report is sent to the preset mailbox, so that the manager can directly know the business state of the store by looking up the business state report in the mailbox, the management is facilitated, and the efficiency of managing a plurality of physical stores by the manager is also improved.
It should be noted that, steps S406 to S408 can be implemented by the processor 101 in fig. 1 a.
The method comprises the steps of acquiring shop-door scene images of all physical shops at regular time to obtain shop-door scene image information of the physical shops; detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information; comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; if not, determining that the physical store is in an abnormal business state; the method comprises the steps of obtaining business states corresponding to all physical stores, classifying the physical stores according to normal business states and abnormal business states, and storing classification results; counting the classification results according to preset time to generate business state reports of all physical stores; and sending the business state report to a preset mailbox. According to the method for detecting the business state of the physical store, the AI detection model is used for detecting the store scene image collected at regular time, so that the door opening and closing state of the physical store is determined, the door opening and closing state is compared with the preset door opening and closing state corresponding to the collection time, whether the business state of the store is normal is determined, the business states of all the physical stores are subjected to classified statistics, and finally the business states are sent to a manager in a report form through mails, so that the effect of supervising all the physical stores is achieved, and the efficiency of managing the business states of the physical stores by the manager is improved.
Corresponding to the method for detecting the business state of the physical store disclosed in the embodiment of the present invention, the embodiment of the present invention provides a system for detecting the business state of the physical store, as shown in fig. 5, the system for detecting the business state of the physical store comprises:
the image acquisition device 501 is configured to acquire store scene images of all physical stores at regular time to obtain store scene image information of the physical stores, where the store scene image information includes acquisition time for acquiring the store scene images.
The detection module 502 is configured to detect the storefront scene image information based on a pre-established AI detection model, and determine a door opening and closing state of an entity store corresponding to the storefront scene image information.
A comparison module 503, configured to compare whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state.
Preferably, as shown in fig. 6, the physical store business state detection system further includes:
the training module 601 is configured to input a training sample into an initial network training model for training, obtain a door opening and closing result of the physical store, compare the door opening and closing result with a door opening and closing result corresponding to the training sample, where the training sample is store door scene image information of the physical store with a predetermined door opening and closing result.
An obtaining module 602, configured to obtain a weight of the initial network training model if the comparison results are the same, and establish an AI detection model based on the weight.
And an adjusting module 603, configured to adjust the weight of the initial network training model if the comparison results are different, continue training a new network training model based on the training samples until the comparison results are the same, and obtain the weight of the current network training model.
An establishing module 604 for establishing an AI detection model based on the weights.
Preferably, as shown in fig. 7, the detecting module 502 includes:
an input unit 701, configured to input the storefront scene image to a pre-established AI detection model;
the detection unit 702 is configured to acquire all the features in the store door scene image information by the AI detection model, analyze all the features, determine that the physical store corresponding to the store door scene image with the image feature representing the door-open state is the door-open state, and determine that the physical store corresponding to the store door scene image with the image feature representing the door-closed state is the door-closed state.
Preferably, as shown in fig. 8, the physical store business state detection system further includes:
the classification module 801 is configured to obtain business states corresponding to all physical stores, classify all the physical stores according to a normal business state and an abnormal business state, and store a classification result.
And the counting module 802 is configured to count the classification results according to a preset time and generate business state reports of all the physical stores.
Preferably, the physical store business state detection system further includes:
and the sending module is used for sending the business state report to a preset mailbox.
It should be noted that, for the specific execution process and execution principle of each module and unit in the store business state detection system disclosed in the foregoing embodiment of the present invention, reference may be made to corresponding parts of the store business state detection method in the data processing method disclosed in the foregoing embodiment of the present invention, and details are not repeated here.
The method comprises the steps that the image acquisition device is used for acquiring shop-door scene images of all physical shop at regular time to obtain shop-door scene image information of the physical shop; the detection module detects the shop door scene image information based on a pre-established AI detection model and determines the door opening and closing state of the physical shop corresponding to the shop door scene image information; the comparison module compares whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state. Through the above-mentioned disclosed entity shop business state detection system, use AI detection model to detect the shop door scene image of regularly gathering to confirm the opening and closing state of entity shop, compare again the opening and closing state with gather whether the door state is unanimous with the preset opening and closing state that the time corresponds, confirm whether normal the business state of entity shop, thereby reach the purpose of carrying out the automatic supervision to all entity shops.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for detecting opening and closing of a physical store is characterized by comprising the following steps:
acquiring storefront scene images of all physical storefronts at fixed time to obtain storefront scene image information of the physical storefronts, wherein the storefront scene image information comprises acquisition time for acquiring the storefront scene images; the step of collecting the storefront scene images of all the physical storefronts at regular time refers to the step that the image collecting equipment collects the storefront scene images of all the physical storefronts according to preset time; the preset acquisition time is set according to the business hours of stores in different areas;
detecting the shop door scene image information based on a pre-established AI detection model, and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information;
comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time;
if the business state is consistent with the business state, determining that the physical store is in a normal business state;
and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state.
2. The method of claim 1, wherein the AI detection model building process comprises:
inputting a training sample into an initial network training model for training to obtain a door opening and closing result of the physical store, and comparing the door opening and closing result with a door opening and closing result corresponding to the training sample, wherein the training sample is store door scene image information of the physical store with a predetermined door opening and closing result;
if the comparison results are the same, obtaining the weight of the initial network training model, and establishing an AI detection model based on the weight;
if the comparison results are different, the weight of the initial network training model is adjusted, a new network training model is continuously trained based on the training samples until the comparison results are the same, and the weight of the current network training model is obtained;
and establishing an AI detection model based on the weight.
3. The method according to claim 1, wherein the detecting the storefront scene image information based on the pre-established AI detection model and determining the door opening and closing state of the physical storefront corresponding to the storefront scene image information comprises:
inputting the shop door scene image into a pre-established AI detection model;
the AI detection model acquires all image characteristics in the shop door scene image information, analyzes all the characteristics, determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door opening state is in the door opening state, and determines that the physical shop corresponding to the shop door scene image with the image characteristic representation in the door closing state is in the door closing state.
4. The method of claim 1, further comprising:
the method comprises the steps of obtaining business states corresponding to all physical stores, classifying the physical stores according to normal business states and abnormal business states, and storing classification results;
and counting the classification results according to preset time to generate business state reports of all physical stores.
5. The method as claimed in claim 4, wherein after counting the classification results according to the preset time and generating the business status report of all physical stores, the method further comprises:
and sending the business state report to a preset mailbox.
6. A physical store door opening and closing detection system is characterized by comprising:
the system comprises an image acquisition device, a processing device and a display device, wherein the image acquisition device is used for acquiring storeroom scene images of all physical storerooms at regular time to obtain storeroom scene image information of the physical storerooms, and the storeroom scene image information comprises acquisition time for acquiring the storeroom scene images; the step of collecting the storefront scene images of all the physical storefronts at regular time refers to the step that the image collecting equipment collects the storefront scene images of all the physical storefronts according to preset time; the preset acquisition time is set according to the business hours of stores in different areas;
the detection module is used for detecting the shop door scene image information based on a pre-established AI detection model and determining the door opening and closing state of the physical shop corresponding to the shop door scene image information;
the comparison module is used for comparing whether the door opening and closing state is consistent with a preset door opening and closing state corresponding to the acquisition time; if the business state is consistent with the business state, determining that the physical store is in a normal business state; and if the business information is inconsistent with the business information, determining that the physical store is in an abnormal business state.
7. The system of claim 6, further comprising:
the training module is used for inputting a training sample into an initial network training model for training to obtain a door opening and closing result of the physical store, and comparing the door opening and closing result with a door opening and closing result corresponding to the training sample, wherein the training sample is store door scene image information of the physical store with a predetermined door opening and closing result;
the acquisition module is used for acquiring the weight of the initial network training model if the comparison results are the same, and establishing an AI (artificial intelligence) detection model based on the weight;
the adjusting module is used for adjusting the weight of the initial network training model if the comparison results are different, continuing to train a new network training model based on the training samples until the comparison results are the same, and acquiring the weight of the current network training model;
and the establishing module is used for establishing an AI detection model based on the weight.
8. The system of claim 6, wherein the detection module comprises:
the input unit is used for inputting the shop door scene image into a pre-established AI detection model;
and the detection unit is used for acquiring all the characteristics in the shop door scene image information by the AI detection model, analyzing all the characteristics, determining that the physical shop corresponding to the shop door scene image with the image characteristic representing the door opening state is in the door opening state, and determining that the physical shop corresponding to the shop door scene image with the image characteristic representing the door closing state is in the door closing state.
9. The system of claim 6, further comprising:
the classification module is used for acquiring business states corresponding to all physical stores, classifying all the physical stores according to normal business states and abnormal business states, and storing classification results;
and the counting module is used for counting the classification results according to the preset time and generating business state reports of all the physical stores.
10. The system of claim 9, further comprising:
and the sending module is used for sending the business state report to a preset mailbox.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910584462.5A CN110263758B (en) | 2019-07-01 | 2019-07-01 | Method and system for detecting opening and closing of physical store |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910584462.5A CN110263758B (en) | 2019-07-01 | 2019-07-01 | Method and system for detecting opening and closing of physical store |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110263758A CN110263758A (en) | 2019-09-20 |
| CN110263758B true CN110263758B (en) | 2021-06-25 |
Family
ID=67923534
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910584462.5A Active CN110263758B (en) | 2019-07-01 | 2019-07-01 | Method and system for detecting opening and closing of physical store |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110263758B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461762A (en) * | 2020-03-05 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Merchant detection method and device and electronic equipment |
| CN115601606B (en) * | 2022-12-15 | 2023-03-03 | 苏州万店掌软件技术有限公司 | Store state detection method, device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147596A (en) * | 2010-02-04 | 2011-08-10 | 成都光码智能科技有限公司 | Automatic hotel room status monitoring system and method |
| CN105447931A (en) * | 2015-03-09 | 2016-03-30 | 北京天诚盛业科技有限公司 | Remote access control authorization method, device and system |
| CN106219367A (en) * | 2016-08-05 | 2016-12-14 | 沈阳聚德视频技术有限公司 | A kind of elevator O&M based on intelligent vision light curtain monitoring method |
| CN106658590A (en) * | 2016-12-28 | 2017-05-10 | 南京航空航天大学 | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information |
| CN106878423A (en) * | 2017-02-21 | 2017-06-20 | 苏州亮磊知识产权运营有限公司 | A kind of shop business information based on monitoring camera provides method and its device |
| CN107679475A (en) * | 2017-09-25 | 2018-02-09 | 平安科技（深圳）有限公司 | Shops's monitoring evaluation method, device and storage medium |
| CN109033993A (en) * | 2018-06-29 | 2018-12-18 | 南京行者易智能交通科技有限公司 | A kind of method and device of image recognition detection switch door |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7165655B2 (en) * | 2002-05-14 | 2007-01-23 | Otis Elevator Company | Neural network detection of obstructions within and motion toward elevator doors |
-
2019
- 2019-07-01 CN CN201910584462.5A patent/CN110263758B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147596A (en) * | 2010-02-04 | 2011-08-10 | 成都光码智能科技有限公司 | Automatic hotel room status monitoring system and method |
| CN105447931A (en) * | 2015-03-09 | 2016-03-30 | 北京天诚盛业科技有限公司 | Remote access control authorization method, device and system |
| CN106219367A (en) * | 2016-08-05 | 2016-12-14 | 沈阳聚德视频技术有限公司 | A kind of elevator O&M based on intelligent vision light curtain monitoring method |
| CN106658590A (en) * | 2016-12-28 | 2017-05-10 | 南京航空航天大学 | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information |
| CN106878423A (en) * | 2017-02-21 | 2017-06-20 | 苏州亮磊知识产权运营有限公司 | A kind of shop business information based on monitoring camera provides method and its device |
| CN107679475A (en) * | 2017-09-25 | 2018-02-09 | 平安科技（深圳）有限公司 | Shops's monitoring evaluation method, device and storage medium |
| CN109033993A (en) * | 2018-06-29 | 2018-12-18 | 南京行者易智能交通科技有限公司 | A kind of method and device of image recognition detection switch door |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110263758A (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108780418A (en) | For using flow point analysis to measure the system and method that effective client of network problem influences in real time | |
| CN110493348A (en) | A kind of intelligent monitoring and alarming system based on Internet of Things | |
| CN105246033B (en) | A kind of crowd state monitoring method and monitoring device based on terminal positioning | |
| US9679243B2 (en) | System and method for detecting platform anomalies through neural networks | |
| US8370193B2 (en) | Method, computer-readable media, and apparatus for determining risk scores and generating a risk scorecard | |
| CN110428522A (en) | A kind of intelligent safety and defence system of wisdom new city | |
| CN104243236B (en) | A kind of method, system and the server of monitoring system O＆M alarm data analysis | |
| CN110263758B (en) | Method and system for detecting opening and closing of physical store | |
| US10404525B2 (en) | Classification of detected network anomalies using additional data | |
| CN106992974B (en) | Live video information monitoring method, device and equipment | |
| CN104899948B (en) | A kind of closing space congestion monitoring method and system | |
| CN100573375C (en) | Electrostatic real-time monitoring system and data acquisition thereof and analytical approach | |
| CN110996259B (en) | Intelligent garbage monitoring and clearing method and device based on edge calculation | |
| US20080186876A1 (en) | Method for classifying applications and detecting network abnormality by statistical information of packets and apparatus therefor | |
| US10750126B2 (en) | Systems and methods of measuring quality of video surveillance infrastructure | |
| CN114743141A (en) | Dangerous behavior detection system for intelligent community | |
| CN106774565B (en) | A kind of intelligent management system and its management method of clinical trial room temperature humidity | |
| CN103945219A (en) | Network side video quality monitoring system | |
| CN113849774A (en) | Intelligent equipment maintenance platform based on life analysis and health monitoring | |
| CN107465652B (en) | Operation behavior detection method, server and system | |
| CN110309737A (en) | A kind of information processing method applied to cigarette sales counter, apparatus and system | |
| CN110362713B (en) | Video monitoring and early warning method and system based on Spark Streaming | |
| Kreuger et al. | Scalable in-network rate monitoring | |
| Chakraborty et al. | System Failure Prediction within Software 5G Core Networks using Time Series Forecasting | |
| Oliveira et al. | Measurement-driven mobile data traffic modeling in a large metropolitan area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201120 Address after: Room 011, first floor, no.2429, Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province (office only) Applicant after: CHUANGYOU digital technology (Guangdong) Co., Ltd Address before: 519000 -41072, 105 room 6, Baohua Road, Hengqin New District, Zhuhai, Guangdong (centralized office area) Applicant before: MINISO (HENGQIN) ENTERPRISE MANAGEMENT Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |