CN110263758A - A kind of entity shops switch gate detection method and system - Google Patents
A kind of entity shops switch gate detection method and system Download PDFInfo
- Publication number
- CN110263758A CN110263758A CN201910584462.5A CN201910584462A CN110263758A CN 110263758 A CN110263758 A CN 110263758A CN 201910584462 A CN201910584462 A CN 201910584462A CN 110263758 A CN110263758 A CN 110263758A
- Authority
- CN
- China
- Prior art keywords
- entity shops
- scene image
- door
- shop
- shops
- 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.)
- Granted
Links
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
Abstract
The invention discloses a kind of entity shops business status detection method and systems.Pass through disclosed entity shops business status detection method, using AI detection model to timing acquiring to shop door scene image detect, determine the switch door state of entity shops, switch door state is compared again and whether preset switches door state corresponding to acquisition time is consistent, determine whether the business status of entity shops is normal, to achieve the purpose that supervise all entity shops automatically.
Description
Technical field
The present invention relates to field of communication technology, specially a kind of entity shops switch gate detection method and system.
Background technique
As the retail shop of same brand more and more, small-scale, dispersion, that manage similar commodity and service goes out
It is existing.The unified management person of retail shop can supervise the business shape of StoreFront by the IP Camera that all StoreFronts under are installed
State.
Conventionally, as the substantial amounts of chain store, so that the IP Camera of all shops is collected
Shop door scene image substantial amounts, and manager can not judge the business of all shops according to the shop door scene image of all shops
State, so that it cannot supervising to all shops, therefore, manager is in the network shooting overhead pass for facing a large amount of shops
Image/video when, the image/video of part shops can only be spot-check, then by spot-check image/video judge this part
Whether shops is in normal business status, and eventually by sampling, obtained abnormal business status estimates that all shops are not normal
The shops of business status, to achieve the purpose that the supervision to shops.
But due to the substantial amounts of chain store and spot-check with randomness, it will lead to nothing by the way of the prior art
Method reaches the problem of being monitored to shops's business status comprehensively.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of entity shops business status detection method and systems, by right
Timing acquiring to shop door scene image detected, and according to testing result determine shops business status it is whether normal, from
And achieve the purpose that supervise shops automatically.
To achieve the above object, the embodiment of the present invention provides the following technical solutions:
First aspect present invention discloses a kind of entity shops business status detection method, comprising:
The shop door scene image of all entity shops of timing acquiring obtains the shop door scene image letter of the entity shops
It ceases, the acquisition time comprising the acquisition shop door scene image in the shop door scene image information;
The shop door scene image information is detected based on the AI detection model pre-established, determines the shop door scene image
The switch door state of entity shops corresponding to information;
It compares the switch door state and whether preset switches door state corresponding to the acquisition time is consistent;
If consistent, it is determined that the entity shops is in normal business status;
If inconsistent, it is determined that the entity shops is in improper business status.
Preferably, the AI detection model establishment process, comprising:
By training sample input initial network training pattern be trained, obtain the switch gate of the entity shops as a result,
Switch gate result switch gate result corresponding with the training sample is compared, the training sample is predetermined
The shop door scene image information of the entity shops of switch gate result;
If comparison result is identical, the weight of the initial network training pattern is obtained, AI detection is established based on the weight
Model;
If comparison result is not identical, the weight of the initial network training pattern is adjusted, is continued based on the training sample
New network training model is trained, until comparison result is identical, obtains the weight of current network training model;
AI detection model is established based on the weight.
Preferably, described that the shop door scene image information is detected based on the AI detection model that pre-establishes, determine described in
The switch door state of entity shops corresponding to shop door scene image information, comprising:
The shop door scene image is input to the AI detection model pre-established;
The AI detection model obtains all characteristics of image in the shop door scene image information, to all features into
Row analysis determines that entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state is door opening state, determines
It is shutdown state that characteristics of image, which characterizes entity shops corresponding to the shop door scene image of shutdown state,.
Preferably, further includes:
Business status corresponding to all entity shops is obtained, by all entity shops by normal business status and non-
Normal business status is classified, and classification results are saved;
Classification results are counted by the preset time, generate the business status report of all entity shops.
Preferably, described that classification results are counted by the preset time, generate the business of all entity shops
After Status Reporting, further includes:
The business status report is sent in preset mailbox.
Second aspect of the present invention discloses a kind of entity shops switch gate detection system, comprising:
Image collecting device obtains the entity shops for the shop door scene image of all entity shops of timing acquiring
Shop door scene image information, comprising acquiring the acquisition time of the shop door scene image in the shop door scene image information;
Detection module determines institute for detecting the shop door scene image information based on the AI detection model pre-established
State the switch door state of entity shops corresponding to shop door scene image information;
Comparison module is for comparing preset switches door state corresponding to the switch door state and the acquisition time
It is no consistent;If consistent, it is determined that the entity shops is in normal business status;If inconsistent, it is determined that the entity shops
In improper business status.
Preferably, further includes:
Training module obtains the entity shops for training sample input initial network training pattern to be trained
Switch gate as a result, switch gate result switch gate result corresponding with the training sample is compared, the training
Sample is the shop door scene image information for predefining the entity shops of switch gate result;
Module is obtained, if identical for comparison result, obtains the weight of the initial network training pattern, is based on the power
Re-establish AI detection model;
Module is adjusted, if not identical for comparison result, the weight of the initial network training pattern adjusted, continues to be based on
The training sample is trained new network training model, until comparison result is identical, obtains current network training mould
The weight of type;
Module is established, for establishing AI detection model based on the weight.
Preferably, the detection module, comprising:
Input unit, for the shop door scene image to be input to the AI detection model pre-established;
Detection unit obtains all features in the shop door scene image information for the AI detection model, to described
All features are analyzed, and determine entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state to open the door
State determines that entity shops corresponding to the shop door scene image of characteristics of image characterization shutdown state is shutdown state.
Preferably, further includes:
Categorization module, for obtaining business status corresponding to all entity shops, by all entity shops by just
Normal business status and improper business status are classified, and classification results are saved;
Statistical module generates the battalion of all entity shops for counting by the preset time to classification results
Industry Status Reporting.
Preferably, further includes:
Sending module, for the business status report to be sent in preset mailbox.
As shown in the above, this application discloses a kind of entity shops business status detection method and systems, by fixed
When acquire the shop door scene images of all entity shops, obtain the shop door scene image information of the entity shops;Based on preparatory
The AI detection model of foundation detects the shop door scene image information, determines entity corresponding to the shop door scene image information
The switch door state of shops;Compare preset switches door state corresponding to the switch door state and the acquisition time whether one
It causes;If consistent, it is determined that the entity shops is in normal business status;If inconsistent, it is determined that the entity shops is in
Improper business status.By entity shops business status detection method disclosed above, using AI detection model to periodically adopting
The shop door scene image collected is detected, so that it is determined that the switch door state of entity shops, then compare the switch door state
It is whether consistent with preset switches door state corresponding to the acquisition time, so that it is determined that just whether the business status of entity shops
Often, achieve the purpose that supervise all entity shops automatically.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 a is a kind of entity shops business status detection system structure provided in an embodiment of the present invention;
Fig. 1 b is a kind of entity shops business status detection method flow chart provided in an embodiment of the present invention;
Fig. 2 is another entity shops provided in an embodiment of the present invention business status detection method flow chart;
Fig. 3 is another entity shops provided in an embodiment of the present invention business status detection method flow chart;
Fig. 4 is another entity shops provided in an embodiment of the present invention business status detection method flow chart;
Fig. 5 is another entity shops provided in an embodiment of the present invention business status detection method flow chart;
Fig. 6 is another entity shops provided in an embodiment of the present invention business status detection system structure;
Fig. 7 is another entity shops provided in an embodiment of the present invention business status detection system structure;
Fig. 8 is another entity shops provided in an embodiment of the present invention business status detection system structure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In this application, the terms "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion,
So that the process, method, article or equipment for including a series of elements not only includes those elements, but also including not having
The other element being expressly recited, or further include for elements inherent to such a process, method, article, or device.Do not having
There is the element limited in the case where more limiting by sentence "including a ...", it is not excluded that in the mistake including the element
There is also other identical elements in journey, method, article or equipment.
This application provides a kind of entity shops switch gate detection method and systems, are able to solve manager in face of a large amount of
When entity shops, accomplish the supervision to each entity shops, and improves manager to shops's efficiency of management.This application discloses
A kind of entity shops switch gate detection system as shown in Figure 1a, the system are made of camera 100 and processor 101.
Wherein, camera 100 is mounted on each entity shops, and processor 101 is then arranged in backstage or cloud, processor
101 are connected by network with camera 100.Camera 100 can acquire the shop door scene image information of entity shops, and will adopt
The shop door scene image information collected is sent to processor 101 by network.Processor 101 then to shop door scene image information into
Row processing finally obtains the switch door state of entity shops corresponding to shop door scene image information, and based on switch door state
Determine whether the business status of entity shops is normal, achievees the purpose that supervise all shops automatically.
A kind of reality is provided based on entity shops switch gate detection system, the specific embodiment of the present invention disclosed in above-mentioned Fig. 1 a
Body shops switch gate detection method, referring to Fig. 1 b, the above method includes at least following steps:
Step S101: the shop door scene image of all entity shops of timing acquiring obtains the shop door field of the entity shops
Scape image information.
It should be noted that including when acquiring the acquisition of the shop door scene image in the shop door scene image information
Between.
In step s101, the shop door scene image of all entity shops of the timing acquiring refer to image capture device by
The shop door scene image of all entity shops is acquired according to the preset time, image capture device here is equivalent in Fig. 1 a
The camera 100 shown, it is not limited to this, which can also be web camera or monitor etc.
Monitoring device with image photographic function.In this application, preset shop door scene image acquisition time is according to not
It is set with shops's business hours in area.
For example, the business hours of a certain geographic entity shops are 19 points to afternoon in 9 points of morning, then preset shop door
Scene image acquisition time should be then scheduled in the business hours, which can be
9 points, or it's 1 minute are past 9 points, entity shops shop door in the afternoon scene image acquisition time can for 18 points 58 minutes, can also be with
For 19 o'clock sharps.
Step S102: the shop door scene image information is detected based on the AI detection model pre-established, determines the shop
The switch door state of entity shops corresponding to door scene image information.
In step s 102, the switch door state includes door opening state and shutdown state.
The AI detection model is that have very strong image characteristic analysis ability, and being can be according to the shop door scene image
The model that characteristics of image in information obtains entity shops switch gate state corresponding to shop door scene image information therefore can
To detect by AI inspection model to the picture characteristics in shop door scene image information, shop door field is determined according to testing result
The switch door state of entity shops corresponding to scape image information.
During executing step S102, as shown in Fig. 2, specific implementation procedure the following steps are included:
Step S201: the shop door scene image is input to the AI detection model pre-established.
Step S202: the AI detection model obtains all features in the shop door scene image information, to described all
Feature is analyzed, and determines that entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state is enabling shape
State determines that entity shops corresponding to the shop door scene image of characteristics of image characterization shutdown state is shutdown state.
It should be noted that the image information in shop door scene image information is made of multiple images feature, AI detection
Model is summarized after obtaining multiple images feature, obtains summarizing characteristics of image, and judged based on characteristics of image is summarized, really
Surely the entity shops for summarizing image characteristic present is in door opening state or shutdown state.
It should be noted that specifically judging to be completed by class probability announcer, specific implementation procedure are as follows: be based on summary view
As feature progress switch gate state classification analysis, summarizing characteristics of image category the characteristics of image probability of determining characterization door opening state more
Classify in door opening state, determines that the characteristics of image that summarizes more than the characteristics of image probability of characterization shutdown state belongs to shutdown state point
Class.
Step S103: compare preset switches door state corresponding to the switch door state and the acquisition time whether one
It causes, if unanimously, thening follow the steps S104, if inconsistent, thens follow the steps S105.
It should be noted that since solid door shop door can be divided into door opening state and shutdown state, it can will preset
Switch door state is set as door opening state, can also be using preset switches door state as shutdown state.
Preferably, in this application, preset switches door state is door opening state.
Step S104: determine that the entity shops is in normal business status.
Step S105: determine that the entity shops is in improper business status.
In order to make it easy to understand, being exemplified below.
Preset switches door state is door opening state, and is made as example with 24 hour time, when acquisition time is 8, solid door
Shop A is door opening state, and the switch door state of entity shops A is consistent with preset switches door state, then illustrates that entity shops A is on time
What enabling was carried on the business, i.e. entity shops A is in normal business status.
Entity shops B is to close the door, and the switch door state and preset switches door state of entity shops B is inconsistent, then illustrates reality
Body shops B does not open the door on time to carry on the business, i.e. entity shops B is in improper business status.
If acquisition time is 20, entity shops C is shutdown state, and the switch door state of entity shops C is opened with default
Shutdown state is inconsistent, then illustrates that entity shops C is to close shop door to come off duty in advance, i.e. entity shops C is in improper business
State.
It should be noted that above-mentioned steps S102 to step S105 can be realized by the processor in Fig. 1 a.
The application passes through the shop door scene image of all entity shops of timing acquiring, obtains the shop door field of the entity shops
Scape image information;The shop door scene image information is detected based on the AI detection model pre-established, determines the shop door scene
The switch door state of entity shops corresponding to image information;It compares corresponding to the switch door state and the acquisition time
Whether preset switches door state is consistent;If consistent, it is determined that the entity shops is in normal business status;If inconsistent,
Determine that the entity shops is in improper business status.By shops's business status detection method disclosed above, AI is used
Detection model to timing acquiring to shop door scene image detect, so that it is determined that the switch door state of entity shops, then compare
It is whether consistent to the switch door state and preset switches door state corresponding to the acquisition time, determine the battalion of entity shops
Whether industry state is normal, to achieve the purpose that supervise all entity shops automatically.
Disclosed entity shops business status detection method based on the above embodiment, it is described in step S102 shown in fig. 1
The establishment process of AI detection model is established, as shown in figure 3, specifically includes the following steps:
Step S301: training sample input initial network training pattern is trained, opening for the entity shops is obtained
It closes the door as a result, switch gate result switch gate result corresponding with the training sample is compared, if comparison result phase
Together, S302 is thened follow the steps, if comparison result is not identical, thens follow the steps S303.
The training sample is the shop door scene image information for predefining the entity shops of switch gate result.
In step S301, training sample input initial network training pattern is trained, i.e., by shop door scene image
Information input initial network training pattern is trained, and initial network training pattern is special according to the image of shop door scene image information
Sign is handled, and the switch gate result of entity shops corresponding to the shop door scene image information is finally obtained.
It should be noted that the initial network training pattern is the primary network training with characteristics of image recognition capability
Model, by selecting the primary network training pattern with characteristics of image recognition capability to be trained training sample, Ke Yijia
Fast AI detection model establishes efficiency.
Optionally, the network structure of characteristics of image recognition capability can also is trained training sample by choosing,
To achieve the purpose that establish AI detection model.
Step S302: obtaining the weight of the initial network training pattern, establishes AI detection model based on the weight.
It should be noted that since the corresponding switch gate result of training sample and training sample pass through initial network training mould
The result that type is trained is consistent, then after illustrating that the initial network training pattern is trained training sample, obtained instruction
Practicing result is expectation as a result, therefore, directly acquiring the weight of initial network training pattern, and establish AI inspection based on obtained weight
Survey model.
Step S303: the weight of the initial network training pattern is adjusted, is continued based on the training sample to new net
Network training pattern is trained, and until comparison result is identical, obtains the weight of current network training model.
Step S304: AI detection model is established based on the weight.
During executing step S303 to step S304, the weight is that one of initial network training pattern is adjustable
Parameter is saved, the output result of training sample input initial network training pattern can be changed by adjusting weight.
It should be noted that the corresponding switch gate result of training sample and training sample by initial network training pattern into
The result of row training is inconsistent, then illustrates that initial network training pattern can not it is expected according to the characteristics of image in training sample
Effect, therefore, it is necessary to adjust the weight parameter of initial network training pattern repeatedly, by adjusting weight parameter, Zhi Daoxun repeatedly
It is consistent with the result that training sample is trained by network training model to practice the corresponding switch gate result of sample.
Based on the implementation procedure of S301 through the above steps to S304, in the AI detection model established, including one point
Class probability announcer, the class probability announcer are based on summarizing image spy for being summarized after obtaining multiple images feature
Sign carries out switch gate state classification analysis, determines that the characteristics of image that summarizes more than the characteristics of image probability of characterization door opening state belongs to out
Door state classification determines that the characteristics of image that summarizes more than the characteristics of image probability of characterization shutdown state belongs to shutdown state classification.
With reference to Fig. 4, for another entity shops provided by the embodiments of the present application business status detection method, the solid door
Shop business status detection method, comprising the following steps:
Step S401: the shop door scene image of all entity shops of timing acquiring obtains the shop door field of the entity shops
Scape image information.
Step S402: the shop door scene image information is detected based on the AI detection model pre-established, determines the shop
The switch door state of entity shops corresponding to door scene image information.
Step S403: compare preset switches door state corresponding to the switch door state and the acquisition time whether one
It causes, if unanimously, thening follow the steps S404, if inconsistent, thens follow the steps S405.
Step S404: determine that the entity shops is in normal business status.
Step S405: determine that the entity shops is in improper business status.
It should be noted that the implementation principle and specific implementation procedure of step S401 to step S405 and as shown in Figure 1
The implementation principle and specific implementation procedure of step S101 to step S105 is identical, reference can be made to above-mentioned corresponding description, no longer superfluous here
It states.
Step S406: obtaining business status corresponding to all entity shops, by all entity shops by normal battalion
Industry state and improper business status are classified, and classification results are saved.
It should be noted that all entity shops are carried out by normal business status and improper business status here
Classification, when temporarily checking the business status of all entity shops for the ease of manager, convenient for counting and checking.
Step S407: counting classification results by the preset time, generates the business shape of all entity shops
State report.
In step S 407, the preset time refers to that certain interval of time counts classification results, this
In the preset time can be each week Monday, or No. 1 of every month, or manager is logical
The preset time changed in sometime section manually is spent, in this application, and without limitation.
Classification processing result is counted, by the way that classification processing result is counted and is calculated, by count results and
Calculated result is saved into table, that is, generates the business status report of entity shops.
The business status report can be form, or picture and text combining form, it is in this application, not right
The form of business status report is defined.
In order to make it easy to understand, being exemplified below.
The business status of upper one month all entity shops is counted with monthly No. 1, when June 1, to May
The business status of 100 shops in 31 days is counted, at 8 points, have 10 entity shops have in this 31 days at least 1 time not
It opens to business on time, at 22 points, there are 5 entity shops to have in this 31 days at least 1 time and close the door in advance.
Step S408: the business status report is sent in preset mailbox.
In step S408, preset mailbox can be one, or multiple.
It should be noted that business status report is sent in preset mailbox, administrative staff can be made direct
The business status of shops is understood by the business status report checked in mailbox, in order to manage, also improves manager's management
The efficiency of multiple solid shop/brick and mortar store.
It should be noted that above-mentioned steps S406 to step S408 can be realized by the processor 101 in Fig. 1 a.
The application passes through the shop door scene image of all entity shops of timing acquiring, obtains the shop door field of the entity shops
Scape image information;The shop door scene image information is detected based on the AI detection model pre-established, determines the shop door scene
The switch door state of entity shops corresponding to image information;It compares corresponding to the switch door state and the acquisition time
Whether preset switches door state is consistent;If consistent, it is determined that the entity shops is in normal business status;If inconsistent,
Determine that the entity shops is in improper business status;Business status corresponding to all entity shops is obtained, by the institute
There is entity shops to classify by normal business status and improper business status, and classification results are saved;By presetting
Time classification results are counted, generate the business status report of all entity shops;The business status report is sent out
It send into preset mailbox.By entity shops business status detection method disclosed above, AI detection model pair is used
Timing acquiring to shop door scene image detected, so that it is determined that the switch door state of entity shops, then compare the switch
Whether door state and preset switches door state corresponding to the acquisition time are consistent, whether just to determine the business status of shops
Often, and to the business status of all entity shops statistic of classification is carried out, manager is finally sent to by mail with report form,
The effect to the supervision of all entity shops has not been only reached, also improves the efficiency of manager's management entity shops business status.
Corresponding with entity shops business status detection method disclosed in the embodiments of the present invention, the embodiment of the present invention mentions
A kind of entity shops business status detection system is supplied, as shown in figure 5, the entity shops business status detection system includes:
Image capture device 501 obtains the solid door for the shop door scene image of all entity shops of timing acquiring
The shop door scene image information in shop, when acquisition in the shop door scene image information comprising acquiring the shop door scene image
Between.
Detection module 502 is determined for detecting the shop door scene image information based on the AI detection model pre-established
The switch door state of entity shops corresponding to the shop door scene image information.
Comparison module 503, for comparing preset switches gate-shaped corresponding to the switch door state and the acquisition time
Whether state is consistent;If consistent, it is determined that the entity shops is in normal business status;If inconsistent, it is determined that the entity
Shops is in improper business status.
Preferably, as shown in fig. 6, the entity shops business status detection system, further includes:
Training module 601 obtains the solid door for training sample input initial network training pattern to be trained
The switch gate in shop is as a result, switch gate result switch gate result corresponding with the training sample is compared, the instruction
Practicing sample is the shop door scene image information for predefining the entity shops of switch gate result.
Module 602 is obtained, if identical for comparison result, obtains the weight of the initial network training pattern, is based on institute
It states weight and establishes AI detection model.
Module 603 is adjusted, if not identical for comparison result, adjusts the weight of the initial network training pattern, is continued
New network training model is trained based on the training sample, until comparison result is identical, obtains current network instruction
Practice the weight of model.
Module 604 is established, for establishing AI detection model based on the weight.
Preferably, as shown in fig. 7, the detection module 502, comprising:
Input unit 701, for the shop door scene image to be input to the AI detection model pre-established;
Detection unit 702 obtains all features in the shop door scene image information for the AI detection model, to institute
It states all features to be analyzed, determines that entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state is to open
Door state determines that entity shops corresponding to the shop door scene image of characteristics of image characterization shutdown state is shutdown state.
Preferably, as shown in figure 8, the entity shops business status detection system, further includes:
Categorization module 801 presses all entity shops for obtaining business status corresponding to all entity shops
Normal business status and improper business status are classified, and classification results are saved.
Statistical module 802 generates all entity shops for counting by the preset time to classification results
Business status report.
Preferably, the entity shops business status detection system, further includes:
Sending module, for the business status report to be sent in preset mailbox.
It should be noted that each mould in a kind of shops's business status detection system disclosed in the embodiments of the present invention
The specific implementation procedure and implementation principle of block and unit, reference can be made in data processing method disclosed in the above embodiment of the present invention
There is the corresponding portion of shops's business status detection method, here with regard to no longer being repeated.
The application passes through the shop door scene image of all entity shops of image collecting device timing acquiring, obtains the entity
The shop door scene image information of shops;Detection module detects the shop door scene image letter based on the AI detection model pre-established
Breath, determines the switch door state of entity shops corresponding to the shop door scene image information;It is switched described in comparison module
Whether door state and preset switches door state corresponding to the acquisition time are consistent;If consistent, it is determined that the entity shops
In normal business status;If inconsistent, it is determined that the entity shops is in improper business status.By disclosed above
Entity shops business status detection system, using AI detection model to timing acquiring to shop door scene image detect, from
And determine the switch door state of entity shops, then compare preset switches corresponding to the switch door state and the acquisition time
Whether door state is consistent, determines whether the business status of entity shops is normal, carries out automatically to reach to all entity shops
The purpose of supervision.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system or
For system embodiment, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to method
The part of embodiment illustrates.System and system embodiment described above is only schematical, wherein the conduct
The unit of separate part description may or may not be physically separated, component shown as a unit can be or
Person may not be physical unit, it can and it is in one place, or may be distributed over multiple network units.It can root
According to actual need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill
Personnel can understand and implement without creative efforts.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. a kind of entity shops switch gate detection method characterized by comprising
The shop door scene image of all entity shops of timing acquiring obtains the shop door scene image information of the entity shops, institute
State the acquisition time in shop door scene image information comprising acquiring the shop door scene image;
The shop door scene image information is detected based on the AI detection model pre-established, determines the shop door scene image information
The switch door state of corresponding entity shops;
It compares the switch door state and whether preset switches door state corresponding to the acquisition time is consistent;
If consistent, it is determined that the entity shops is in normal business status;
If inconsistent, it is determined that the entity shops is in improper business status.
2. the method according to claim 1, wherein the AI detection model establishment process, comprising:
Training sample input initial network training pattern is trained, obtains the switch gate of the entity shops as a result, by institute
It states switch gate result switch gate result corresponding with the training sample to be compared, the training sample is to predefine switch
The shop door scene image information of the entity shops of door result;
If comparison result is identical, the weight of the initial network training pattern is obtained, AI detection model is established based on the weight;
If comparison result is not identical, the weight of the initial network training pattern is adjusted, is continued based on the training sample to new
Network training model be trained, until comparison result is identical, obtain the weight of current network training model;
AI detection model is established based on the weight.
3. the method according to claim 1, wherein described based on described in the AI detection model pre-established detection
Shop door scene image information determines the switch door state of entity shops corresponding to the shop door scene image information, comprising:
The shop door scene image is input to the AI detection model pre-established;
The AI detection model obtains all characteristics of image in the shop door scene image information, divides all features
Analysis determines that entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state is door opening state, determines image
Entity shops corresponding to the shop door scene image of characteristic present shutdown state is shutdown state.
4. the method according to claim 1, wherein further include:
Business status corresponding to all entity shops is obtained, by all entity shops by normal business status and improper
Business status is classified, and classification results are saved;
Classification results are counted by the preset time, generate the business status report of all entity shops.
5. according to the method described in claim 4, it is characterized in that, described unite to classification results by the preset time
It counts, after the business status report for generating all entity shops, further includes:
The business status report is sent in preset mailbox.
6. a kind of entity shops switch gate detection system characterized by comprising
Image collecting device obtains the shop of the entity shops for the shop door scene image of all entity shops of timing acquiring
Door scene image information, comprising acquiring the acquisition time of the shop door scene image in the shop door scene image information;
Detection module determines the shop for detecting the shop door scene image information based on the AI detection model pre-established
The switch door state of entity shops corresponding to door scene image information;
Comparison module, for compare preset switches door state corresponding to the switch door state and the acquisition time whether one
It causes;If consistent, it is determined that the entity shops is in normal business status;If inconsistent, it is determined that the entity shops is in
Improper business status.
7. system according to claim 6, which is characterized in that further include:
Training module obtains opening for the entity shops for training sample input initial network training pattern to be trained
It closes the door as a result, switch gate result switch gate result corresponding with the training sample is compared, the training sample
For the shop door scene image information of the entity shops of predetermined switch gate result;
Module is obtained, if identical for comparison result, the weight of the initial network training pattern obtained, is built based on the weight
Vertical AI detection model;
Module is adjusted, if not identical for comparison result, adjusts the weight of the initial network training pattern, is continued based on described
Training sample is trained new network training model, until comparison result is identical, obtains current network training model
Weight;
Module is established, for establishing AI detection model based on the weight.
8. system according to claim 6, which is characterized in that the detection module, comprising:
Input unit, for the shop door scene image to be input to the AI detection model pre-established;
Detection unit obtains all features in the shop door scene image information for the AI detection model, to described all
Feature is analyzed, and determines that entity shops corresponding to the shop door scene image of characteristics of image characterization door opening state is enabling shape
State determines that entity shops corresponding to the shop door scene image of characteristics of image characterization shutdown state is shutdown state.
9. system according to claim 6, which is characterized in that further include:
Categorization module, for obtaining business status corresponding to all entity shops, by all entity shops by normal battalion
Industry state and improper business status are classified, and classification results are saved;
Statistical module generates the business shape of all entity shops for counting by the preset time to classification results
State report.
10. system according to claim 9, which is characterized in that further include:
Sending module, for the business status report to be sent in 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 true CN110263758A (en) | 2019-09-20 |
CN110263758B 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) |
Cited By (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 |
CN115601606A (en) * | 2022-12-15 | 2023-01-13 | 苏州万店掌软件技术有限公司(Cn) | Store state detection method, device, equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050173200A1 (en) * | 2002-05-14 | 2005-08-11 | Cook Brett E. | Neural network detection of obstructions within and motion toward elevator doors |
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 |
-
2019
- 2019-07-01 CN CN201910584462.5A patent/CN110263758B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050173200A1 (en) * | 2002-05-14 | 2005-08-11 | Cook Brett E. | Neural network detection of obstructions within and motion toward elevator doors |
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 |
Cited By (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 |
CN115601606A (en) * | 2022-12-15 | 2023-01-13 | 苏州万店掌软件技术有限公司(Cn) | Store state detection method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110263758B (en) | 2021-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105100732B (en) | A kind of server in machine room long-distance monitoring method and system | |
CN106878064A (en) | Data monitoring method and device | |
CN105554471B (en) | Video sequence adjustment method and device based on event statistics | |
CN100573375C (en) | Electrostatic real-time monitoring system and data acquisition thereof and analytical approach | |
CN107332715B (en) | Network application system with active performance test and passive shunt control and implementation method thereof | |
CN110263758A (en) | A kind of entity shops switch gate detection method and system | |
CN102882745A (en) | Method and device for monitoring business server | |
CN107733905A (en) | A kind of detection method of industry control network unit exception flow | |
US10750126B2 (en) | Systems and methods of measuring quality of video surveillance infrastructure | |
US20170353363A1 (en) | Systems and methods for managing network operations | |
US20140227992A1 (en) | Method of and An Operating Support System for Providing Performance Management in a Mobile Telecommunications System | |
CN105955876A (en) | Data monitoring processing method and apparatus | |
CN105205591A (en) | Intelligent monitoring method and system | |
CN107465652B (en) | Operation behavior detection method, server and system | |
CN109615817A (en) | Monitoring system combines alarm method, device, electronic equipment and storage medium | |
CN117093822A (en) | Industrial brain data analysis platform based on industrial knowledge graph | |
CN110533328A (en) | A kind of project scene traffic control method, apparatus, medium and terminal device | |
WO2016076841A1 (en) | Systems and methods of measuring quality of video surveillance infrastructure | |
CN112216073B (en) | Ladder violation operation warning method and device | |
CN110175747A (en) | A kind of Information Collection System based on working process | |
CN105007278A (en) | Automatic real-time acquisition system and acquisition method for network safety log | |
CN111178113A (en) | Information processing method, device and storage medium | |
CN114463873A (en) | Patrol system for community | |
WO2015028714A1 (en) | Fault anticipating service monitoring in a communications network | |
CN113177883A (en) | Data queue-based arrangement transmission system |
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 | ||
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. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |