CN116343116B - Multi-scene personnel flow monitoring system based on image analysis - Google Patents
Multi-scene personnel flow monitoring system based on image analysis Download PDFInfo
- Publication number
- CN116343116B CN116343116B CN202310273261.XA CN202310273261A CN116343116B CN 116343116 B CN116343116 B CN 116343116B CN 202310273261 A CN202310273261 A CN 202310273261A CN 116343116 B CN116343116 B CN 116343116B
- Authority
- CN
- China
- Prior art keywords
- scene
- population
- people
- scenic spot
- monitoring
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 116
- 238000010191 image analysis Methods 0.000 title claims abstract description 14
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000005457 optimization Methods 0.000 claims abstract description 9
- 230000011664 signaling Effects 0.000 claims description 36
- 230000004083 survival effect Effects 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 6
- 229920006395 saturated elastomer Polymers 0.000 claims description 6
- 230000014759 maintenance of location Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a multi-scene personnel flow monitoring system based on image analysis, which relates to the technical field of personnel monitoring systems and is used for solving the problems that the existing monitoring system is too single in monitoring the scene personnel flow, cannot adapt to different scenes, and has more complete functions, and the monitored personnel flow data and analysis are more complicated and have more excessive functions; the system comprises a scene acquisition module, a scene analysis module, a detection module, a mode switching module and a processing module, wherein the scene analysis module can be used for rapidly analyzing an erected monitoring area, and switching the system to a personnel flow monitoring mode matched with the system, so that different personnel flow data are acquired and analyzed in different scenes in a targeted manner, corresponding optimization effects are obtained, and different application scenes can be flexibly adapted.
Description
Technical Field
The invention relates to the technical field of personnel monitoring systems, in particular to a multi-scene personnel flow monitoring system based on image analysis.
Background
With the development of science and technology, people increasingly enter places with large people flow and mixed people, such as markets, public transportation and other areas, because of more mobile people and larger occupied area, people flow more easily in some places such as scenic spots, people flow overstocked in scenic spots and are easy to occur out of business, so that people in and out of the places need to be effectively subjected to data recording and monitoring;
however, the existing monitoring system is too single in monitoring the flow of scene personnel, cannot adapt to different scenes, and the monitoring system with complete functions has the conditions of complicated monitored personnel flow data and excessive analysis, and can bring a large burden to the system in severe cases, so that the multi-scene personnel flow monitoring system based on image analysis is designed.
In order to solve the above-mentioned defect, a technical scheme is provided.
Disclosure of Invention
The invention aims to solve the problems that the existing monitoring system is too single for monitoring the flow of scene personnel and cannot adapt to different scenes, and the monitoring system with complete functions has the conditions of complicated monitored personnel flow data and excessive analysis, and the complicated data or excessive data volume and calculation analysis volume bring great burden to the system, and provides the multi-scene personnel flow monitoring system based on image analysis.
The aim of the invention can be achieved by the following technical scheme:
a multi-scenario personnel flow monitoring system based on image analysis, comprising:
the scene acquisition module is used for acquiring scene videos of the monitored erection area through the erected monitoring equipment;
the scene analysis module is used for receiving the scene video acquired by the scene acquisition module and analyzing the scene acquired by the monitoring equipment erection, and generating a shopping center mode switching signaling when the scene of the monitoring erection area is judged to be a shopping center scene; when the scene of the monitoring erection area is judged to be a public transportation scene, a public transportation mode switching signaling is generated; generating scenic spot mode switching signaling when the scene of the monitored erection area is determined to be a scenic spot scene; when the scene of the monitoring erection area is judged to be a commodity scene, commodity mode switching signaling is generated;
the monitoring module is used for monitoring the people stream data of the scene through the monitoring equipment after the determination of the scene of the monitoring equipment which is determined to be erected is completed, and monitoring different people stream data according to the scene of the shopping center, the scene of public transportation, the scene of scenic spots and the scene of goods;
the mode switching module is used for performing mode switching on the mode switching signaling generated by the receiving scene analysis module and forming a corresponding monitoring mode;
and the processing module is used for processing and analyzing the received matched people stream data after the monitoring mode is determined to form corresponding optimization measures.
Further, the specific operation steps of the scene analysis module for receiving the scene video acquired by the scene acquisition module and analyzing and obtaining the scene obtained by erecting the monitoring equipment are as follows:
capturing a frame of image in a scene video acquired in a receiving scene acquisition module, setting the captured frame of image as a scene reality image, carrying out similarity analysis and judgment on the scene reality image and a scene model image in a preset scene analysis module, wherein a scene model in the preset analysis module consists of four scene categories of a shopping center scene, a public transportation scene, a scenic spot scene and a commodity scene, five common scene model images are preset in each scene category, namely twenty scene model images in total, carrying out one-by-one comparison on the scene reality image and the twenty scene model images, and respectively extracting sift feature points of the scene reality image and the twenty scene model images;
after the number of the characteristic points of the scene reality image and the compared scene model image is obtained, comparing the number of the characteristic points of the scene reality image and the compared scene model image with a preset value, and when the number of the characteristic points of the scene reality image and the compared scene model image is smaller than the preset value, selecting a method which does not depend on the number of the characteristic points, namely based on histogram similarity calculation, to detect the scene reality image and the compared scene model image; when the number of the characteristic points of the scene reality image and the number of the characteristic points of the compared scene model image are larger than a preset value, performing similarity analysis on the scene reality image and the compared scene model image by a characteristic point matching method;
after the similarity of twenty scene model images consisting of the scene reality image and four scene categories is obtained respectively, twenty similarity values are obtained and respectively calibrated as G 1 、G 2 、G 3 、G 4 、G 5 、J 1 、J 2 、J 3 、J 4 、J 5 、Q 1 、Q 2 、Q 3 、Q 4 、Q 5 S and S 1 、S 2 、S 3 、S 4 、S 5 Comparing the twenty obtained similarity values to obtain a highest value; when the highest value is at G 1 、G 2 、G 3 、G 4 、G 5 If the scene of the monitoring erection area is the shopping center scene, generating a shopping center mode switching signaling; when the highest value is at J 1 、J 2 、J 3 、J 4 、J 5 If the traffic is in the middle, judging that the scene of the monitored erection area is a public transportation scene, and generating a public transportation mode switching signaling; when the highest value is at Q 1 、Q 2 、Q 3 、Q 4 、Q 5 If the scene of the monitored erection area is the scenic spot scene, generating a scenic spot mode switching signaling; when the highest value is at S 1 、S 2 、S 3 、S 4 、S 5 And if the scene of the monitored erection area is judged to be the commodity scene, commodity mode switching signaling is generated.
Further, the people stream data monitored in the monitoring module consists of shopping center people stream data, public transportation people stream data, scenic spot people stream data and commodity people stream data; the shopping center people stream data comprise the number of business hall entrance and exit people, the number of business hall entrance and exit people and peak and leisure passenger flow; the public traffic people stream data comprises the number of people entering and exiting stations, the waiting number of people waiting for each road of vehicles and the Gao Pingfeng people stream number; scenic spot people stream data comprise the population number of scenic spots, population flow condition and population diversion number of branches; the commodity stream data comprises the population number of the regional in and out, commodity bag lifting rate and stream commodity station storage time, and the monitoring equipment in the monitoring module adopts a binocular camera.
Further, the specific operation steps of the mode switching module are as follows:
when a shopping center mode switching signaling is received, switching to a shopping center personnel flow monitoring mode, and pertinently grabbing shopping center personnel flow data collected by the monitoring module; when a public transportation mode switching signaling is received, switching to a public transportation personnel flow monitoring mode, and pertinently grabbing scenic spot people flow data collected by the monitoring module; when receiving the scenic spot mode switching signaling, switching to a scenic spot personnel flow monitoring mode, and pertinently grabbing scenic spot personnel flow data collected by the monitoring module; when receiving the commodity mode switching signaling, switching to a commodity personnel flow monitoring mode, and pertinently grabbing commodity personnel flow data collected by the monitoring module.
Further, the specific operation steps of the processing module for analyzing and optimizing the people stream data of the shopping center after determining the people stream data monitoring mode of the shopping center are as follows:
receiving the people stream data of the shopping center, and analyzing the people stream data of the shopping center to obtain the population number of the inlets and outlets of the mall, the data of the inlets and outlets of the mall, the peak and the leisure passenger flow; the acquired population quantity of the store entrance and population quantity of the store exit are subjected to difference to obtain the population quantity of the store in real time, the population quantity of the store in real time is set as an ordinate, a population retention line graph is established by taking the abscissa as a time point, the population quantity of the store in real time is intuitively and real-timely observed, a plurality of peak points and low peak points are selected from the population retention line graph, population quantities of the peak points and the low peak points and corresponding time are recorded, so that peak passenger flow quantity, matched peak passenger flow quantity, low peak passenger flow quantity and matched low peak time quantity are obtained, and staff quantity of the low peak time quantity in the store is reduced and filled up to the peak time;
the acquired population entering quantity of each shop and the acquired population exiting quantity of the shops are subjected to difference to obtain real-time survival people in each shop, the survival people at a plurality of time points in a single shop are utilized to obtain the survival people at different time points in the shops through mean value calculation to obtain the survival people at different time points in the shops, the survival people at different time points in the shops are sequenced from small to large, the survival people at different time points in the shops which are ranked in the front 20% and the rear 20% are selected respectively, the shops which are ranked in the front 20% and the rear 20% are marked, so that the shops with better and worse operation performance are obtained, the shops are classified according to business operation types, the sum of the survival people at different types of the shops is calculated and compared, the shops which are located in the first and last two business types are selected, the shops with corresponding types are marked, the proper and improper operation type shops or saturated type shops of the shops are judged, and the shops with different operation types and the proper operation type or the saturated operation type of the shops of the proper operation type and the operation type of the business type of the shop are arranged according to the shops.
Further, the specific operation steps of analyzing and optimizing the mass transit people stream data after the processing module determines the mass transit people stream data monitoring mode are as follows:
when public transportation people stream data are received, analyzing the public transportation people stream data to obtain the number of people entering and exiting a station, the waiting number of people waiting for each road of vehicles and Gao Pingfeng people stream number; the method comprises the steps of subtracting the number of the station exit persons from the number of the station entrance persons to obtain the number of temporary persons in the station, counting and recording the number of the temporary persons in the station at each preset time node, simultaneously capturing a single frame of image of a monitored video according to the preset time point to obtain a plurality of images matched with the preset time node of the counted number of the temporary persons in the station, analyzing each image, carrying out graying and noise removing pretreatment on the images to strengthen the edges of foreground objects, and analyzing to obtain the passenger flow of each area in the images, so as to obtain the passenger flow distribution condition of each appointed area in the whole monitored area;
then, a line graph is established for a plurality of preset time node station temporary population as an ordinate and preset time nodes as an abscissa, a plurality of high peak values and low peak values are selected, and the station temporary population with the high peak values, the station temporary population with the low peak values and time points are recorded, namely Gao Pingfeng people stream number;
analyzing the waiting number of the collected vehicle population, comparing the waiting number of the multipath vehicle population, sequentially arranging the compared numerical values from large to small, selecting numerical values of a plurality of front digits and a plurality of rear digits, marking the corresponding vehicle numbering road numbers, properly increasing the vehicle numbering road numbers corresponding to the numerical values of the front digits, and reducing the vehicle numbering road numbers of the rear digits in a matched manner to achieve a more balanced state.
Further, the specific operation steps of analyzing and optimizing the scenic spot people stream data after the processing module determines the scenic spot people stream data monitoring mode are as follows:
when receiving scenic spot people stream data, analyzing the scenic spot people stream data to obtain the population number of scenic spots, population flow condition and diversion population diversion number of branches; subtracting the scenic spot outgoing flow from the acquired scenic spot incoming flow to obtain scenic spot existing flow, respectively establishing a scenic spot existing population curve for the ordinate and the abscissa of the scenic spot existing population and setting a preset extremum of the scenic spot existing population, and stopping entering the population when the scenic spot existing population reaches or exceeds the preset extremum to form a state of only entering, so that the scenic spot people flow overload is avoided and accidents are avoided;
selecting a plurality of high peak points and low peak points in the established existing people number curve, recording corresponding time points, and correspondingly debugging the working time of staff in the scenic spot according to the high peak points and the local wind points of the existing people flow in the scenic spot;
summing the acquired scenic spot inflow and scenic spot outflow to obtain total traffic, matching the inflow population of the scenic spot with the action speed of the scenic spot outflow population, normalizing the three values of the total traffic of the scenic spot, the inflow population action speed of the scenic spot and the scenic spot outflow population action speed, obtaining a flow condition value by using a formula, establishing a histogram of the flow condition value according to time points, and controlling the scenic spot inflow speed to avoid scenic spot overload when the flow condition value is higher than a preset value;
and sorting the collected population diversion numbers of the plurality of branches from large to small to obtain the population diversion numbers of the branches at the later positions, marking the corresponding scenic spot branches, and performing playability or scenic optimization on the marked scenic spot branches to divert the branches with more population diversion numbers.
Further, the specific operation steps of the processing module for analyzing and optimizing the commodity people stream data after determining the commodity people stream data monitoring mode are as follows:
receiving commodity people stream data, analyzing the commodity people stream data to obtain the number of in-out population, commodity bag lifting rate and people stream commodity station storage time, calculating the obtained number of in-out population to obtain the existing people stream in a monitoring area, judging a high people stream time period and a low people stream time period of the monitoring area according to different existing people streams at different time points, and changing the loading and unloading time of a storage rack of the monitoring area according to the high people stream time period and the low people stream time period;
and substituting the normalized bag lifting rate of each commodity on the goods shelf and the time of each commodity people stream station into a formula to obtain the attention value of a single commodity, averaging the attention values of a plurality of commodities, calculating the average value by using the average value calculation, replacing and considering the commodity with the attention value lower than the average value, and improving the overall attention value of the plurality of commodities.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the scene analysis module is utilized to rapidly analyze the erected monitoring area, the system is switched to the personnel flow monitoring mode matched with the scene analysis module, different personnel flow data acquisition and analysis are performed on different scenes in a targeted manner, the corresponding optimization effect is obtained, and different preset application scenes can be flexibly adapted.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
fig. 1 is a general block diagram of the system of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the present disclosure is for the purpose of describing particular embodiments only, and is not intended to be limiting of the disclosure. As used in the specification and claims of this disclosure, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the term "and/or" as used in the present disclosure and claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As shown in fig. 1, a multi-scene personnel flow monitoring system based on image analysis comprises a scene acquisition module, a scene analysis module, a detection module, a mode switching module and a processing module;
the scene acquisition module is used for acquiring scene videos of a monitoring erection area through an erected monitoring device; wherein the monitoring erection area is a fixed scene or a variable scene after the erection is completed;
the scene analysis module is used for receiving the scene video acquired by the scene acquisition module and analyzing and obtaining a scene obtained by erecting the monitoring equipment; capturing a frame of image in a received scene video, setting the frame of image as a scene reality image, carrying out similarity analysis and judgment on the scene reality image and a scene model image preset in a scene analysis module, wherein a scene model preset in the analysis module consists of 4 scene categories of a shopping center scene, a public transportation scene, a scenic spot scene and a commodity scene, 5 common scene model images are preset in each scene category, the total number of the scene model images is 20, the scene reality image and the 20 scene model images are compared one by one, and sift feature points of the scene reality image and the 20 scene model images are respectively extracted, wherein sift is Scale InvariantFeature Transform, is an existing algorithm for detecting the local feature points of the image, keeps invariance to image rotation, scale scaling, brightness change and the like, and is a very stable image feature description; obtaining the number of the characteristic points of the scene reality image and the compared scene model image, and selecting a method which does not depend on the number of the characteristic points, namely based on histogram similarity calculation, to detect the scene reality image and the compared scene model image when the number of the characteristic points of the scene reality image and the compared scene model image is smaller than a preset value; when the number of the characteristic points of the scene reality image and the number of the characteristic points of the compared scene model image are larger than a preset value, performing similarity analysis on the scene reality image and the compared scene model image by a characteristic point matching method; the two different methods can be used for adapting to the monitoring scene with simpler image picture and single color information, can also adapt to the monitoring scene with complex image picture and rich color information, and have better compatibility;
after the similarity of 20 scene model images consisting of the scene reality image and 4 scene categories is obtained respectively, 20 similarity values are obtained and respectively calibrated as G 1 、G 2 、G 3 、G 4 、G 5 、J 1 、J 2 、J 3 、J 4 、J 5 、Q 1 、Q 2 、Q 3 、Q 4 、Q 5 S and S 1 、S 2 、S 3 、S 4 、S 5 Comparing the obtained 20 similarity values to obtain a highest value; when the highest value is at G 1 、G 2 、G 3 、G 4 、G 5 If the scene of the monitoring erection area is the shopping center scene, generating a shopping center mode switching signaling; when the highest value is at J 1 、J 2 、J 3 、J 4 、J 5 If the traffic is in the middle, judging that the scene of the monitored erection area is a public transportation scene, and generating a public transportation mode switching signaling; when the highest value is at Q 1 、Q 2 、Q 3 、Q 4 、Q 5 If the scene of the monitored erection area is the scenic spot scene, generating a scenic spot mode switching signaling; when the highest value is at S 1 、S 2 、S 3 、S 4 、S 5 If the scene of the monitored erection area is the commodity scene, commodity mode switching signaling is generated;
the monitoring module is used for monitoring the people stream data of the scene through the monitoring equipment; the people stream data consists of shopping center people stream data, public transportation people stream data, scenic spot people stream data and commodity people stream data; the shopping center people stream data comprise the number of business hall entrance and exit population, peaks and leisure passenger flow; the public traffic people stream data comprises the number of people entering and exiting stations, the waiting number of people waiting for each road of vehicles and the Gao Pingfeng people stream number; scenic spot people stream data comprise the population number of scenic spots, population flow condition and population diversion number of branches; the commodity stream data comprises the population number of the regional in and out, commodity bag lifting rate and stream commodity station time; the shopping center people stream data comprise the number of business hall entrance and exit population, peaks and leisure passenger flow; the public traffic people stream data comprises the number of people entering and exiting stations, the waiting number of people waiting for each road of vehicles and the Gao Pingfeng people stream number; scenic spot people stream data comprise the population number of scenic spots, population flow condition and population diversion number of branches; the commodity stream data comprises the population number of the regional in and out, commodity bag lifting rate and stream commodity station time; the monitoring equipment in the monitoring module adopts the binocular camera, the binocular camera adopts the parallax technology, the defect of monocular is avoided, and the influence of disordered environments such as shot-light shining, western sun shine, shadow influence, carpet, pedestrian posture, pedestrian clothes color, double or multiple people side by side can be coped with.
The mode switching module is used for receiving the mode switching signaling generated by the scene analysis module and changing the corresponding personnel flow monitoring mode for the corresponding mode switching signaling; when a shopping center mode switching signaling is received, switching to a shopping center personnel flow monitoring mode, and pertinently grabbing shopping center personnel flow data collected by the monitoring module; when a public transportation mode switching signaling is received, switching to a public transportation personnel flow monitoring mode, and pertinently grabbing scenic spot people flow data collected by the monitoring module; when receiving the scenic spot mode switching signaling, switching to a scenic spot personnel flow monitoring mode, and pertinently grabbing scenic spot personnel flow data collected by the monitoring module; when receiving the commodity mode switching signaling, switching to a commodity personnel flow monitoring mode, and pertinently grabbing commodity personnel flow data collected by the monitoring module;
the processing module is used for receiving the people stream data and processing the data; when receiving the people stream data of the shopping center, analyzing the people stream data of the shopping center to obtain the population number of the inlets and outlets of the shopping malls, the data of the inlets and outlets of the shopping malls, peaks and leisure passenger flow; the acquired population entering quantity and market population exiting quantity of the market are respectively calibrated into JR and CQ, and the formula is utilized: the method comprises the steps of obtaining the number of real-time reserve population SR in a market by SR=JR-CQ, setting the number of real-time reserve population SR in the market as an ordinate, setting a reserve people line graph for time points, and conveniently observing the reserve people in the market in real time, wherein the time point interval can be set to 10min, 30min or 1h, selecting a plurality of peak points and low peak points from the reserve people line graph, recording the population numbers of the peak points and the low peak points and corresponding time, obtaining peak passenger flow volume, matched peak passenger flow volume, low peak passenger flow volume and matched low peak time volume, reducing the number of staff in the low peak time period in the market to the peak time period, so as to achieve balance of staff and passenger flow volume in the market, and improving customer experience; the acquired population entering quantity of each shop is differed from the population exiting quantity of the shop to obtain the real-time number of the survivors in each shop, and the number of the survivors at a plurality of time points in a single shop is respectively marked as CL 1 、CL 2 、CL 3 、......、CL s Where s is the number of selected time points, using the formula:obtaining the average value CJ of the number of people remaining at different time points in the shops, and respectively calibrating the average value of the number of people remaining at different time points in a plurality of shops as CJ 1 、CJ 2 、CJ 3 、......、CJ p Wherein p is the number of shops monitored by the monitoring equipment, and the average CJ of the number of people remaining in a plurality of shops at different time points is calculated 1 、CJ 2 、CJ 3 、......、CJ p Comparing and connecting CJ 1 、CJ 2 、CJ 3 、......、CJ p Sequencing from small to large in sequence, respectivelySelecting average values of the reserve people of the shops ranked in the front 20% and the rear 20% at different time points, marking the shops ranked in the front 20% and the rear 20% to obtain shops with better and worse operation performance, classifying the shops according to business operation types, calculating the sum of the average values of the reserve people of the shops of different types, comparing the average values of the reserve people of the shops with different types, selecting the types of the shops at the first and last, marking the shops with corresponding types, analyzing and judging the shops with proper and improper operation types or saturated operation types, and carrying out different types of shop layout according to the shops with proper and improper operation types or saturated operation types;
when public transportation people stream data are received, analyzing the public transportation people stream data to obtain the number of people entering and exiting a station, the waiting number of people waiting for each road of vehicles and Gao Pingfeng people stream number; the method comprises the steps of subtracting the number of the station exit persons from the number of the station entrance persons to obtain the number of temporary persons in the station, counting and recording the number of the temporary persons in the station at each preset time node, simultaneously capturing a single frame of image of a monitored video according to the preset time point to obtain a plurality of images matched with the preset time node of the counted number of the temporary persons in the station, analyzing each image, carrying out graying and noise removing pretreatment on the images to strengthen the edges of foreground objects, and analyzing to obtain the passenger flow of each area in the images, so as to obtain the passenger flow distribution condition of each appointed area in the whole monitored area; meanwhile, a line graph is established for a plurality of preset time node station temporary storage numbers serving as an ordinate and preset time nodes serving as an abscissa, a plurality of high peak values and low peak values are selected, and the station temporary storage numbers of the high peak values, the station temporary storage numbers of the low peak values and time points are recorded, namely Gao Pingfeng people stream numbers; the method is also used for analyzing the collected population waiting quantity of each vehicle, and respectively calibrating the population waiting quantity of multiple paths of vehicles as DD 1 、DD 2 、DD 3 、......、DD m Where m is the number of vehicle number passes, waiting for a number DD for multiple vehicle populations 1 、DD 2 、DD 3 、......、DD m Comparing and taking the comparison value fromThe number values of the first and the last digits are sequentially arranged from large to small, the corresponding number of the vehicle numbering paths is marked, the number of the vehicle numbering paths corresponding to the number values of the first digits is properly increased, the number of the vehicle numbering paths of the last digits is reduced in a matched manner, so that an equilibrium state is achieved, good riding conditions are provided for passengers, and the layout and the use of the scheduled vehicles are economically and reasonably carried out;
when receiving scenic spot people stream data, analyzing the scenic spot people stream data to obtain the population number of scenic spots, population flow condition and diversion population diversion number of branches; subtracting the scenic spot outgoing flow from the acquired scenic spot incoming flow to obtain scenic spot existing flow, respectively establishing a scenic spot existing population curve for the ordinate and the abscissa of the scenic spot existing population and setting a preset extremum of the scenic spot existing population, and stopping entering the population when the scenic spot existing population reaches or exceeds the preset extremum to form a state of only entering, so that the scenic spot people flow overload is avoided and accidents are avoided; meanwhile, a plurality of high peak points and low peak points in the established existing people number curve are selected, corresponding time points are recorded, and the working time of staff in the scenic spot is debugged according to the high peak points and the ground wind points of the existing people in the scenic spot, so that the emergency capacity of the scenic spot and the experience of tourists are improved; the collected sum of the inflow people and the outflow people of the scenic spot is calibrated to be ZR, meanwhile, the inflow population of the scenic spot and the movement speed of the outflow population of the scenic spot are collected and respectively calibrated to be JS and CS, and the three numerical values of the total inflow people ZR of the scenic spot, the inflow population movement speed JS of the scenic spot and the movement speed CS of the outflow population of the scenic spot are normalized and substituted into a formula:obtaining a mood value LQ, wherein +.>Beta and χ are preset weight coefficients of total people flow of the scenic spot, moving speed of the entrance population of the scenic spot and moving speed of the exit population of the scenic spot respectively, and the flow condition value LQ is subjected to histogram according to time pointsWhen the flow condition value LQ is higher than a preset value, controlling the scenic spot entering speed to avoid the occurrence of scenic spot overload; the collected diversion population diversion quantity is respectively calibrated as RL 1 、RL 2 、RL 3 、......、RL u Where u is the number of branches of the scenic spot, and the population of branches is divided into a number RL 1 、RL 2 、RL 3 、......、RL u Sorting from large to small to obtain population diversion numbers of a plurality of later branches, marking corresponding scenic spot branches, and performing playability or scenic optimization on the marked scenic spot branches to divert branches with a large population diversion number;
when commodity people stream data are received, analyzing the commodity people stream data to obtain the number of in-out population, commodity bag lifting rate and people stream commodity station storage time, calculating the obtained number of in-out population to obtain the existing people stream in a monitoring area, judging a high people stream time period and a low people stream time period of the monitoring area according to different existing people stream at different time points, and changing the loading and unloading time of a goods shelf in the monitoring area according to the high people stream time period and the low people stream time period to avoid influence of customers on commodity selection; calibrating the bag lifting rate of each commodity on the goods shelf to TD respectively, calibrating the standing time of each commodity to ZJ respectively, normalizing the bag lifting rate TD of the commodity and the standing time of the commodity to ZJ, and substituting the normalized bag lifting rate TD and the normalized standing time of the commodity into the formula:obtaining a concerned value GZ of a single commodity, wherein Γ is an influence factor, the value is 0.63, and the concerned values of a plurality of commodities are respectively calibrated as GZ 1 、GZ 2 、GZ 3 、......、GZ g And g is the number of commodities, the average value is calculated by using an average value calculation formula, and the commodities with the attention value GZ lower than the average value of the plurality of commodities are replaced and considered, so that the overall attention value of the plurality of commodities is improved, the good commodities are preferentially displayed to customers, and the customers can conveniently purchase and improve the bag lifting rate.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (7)
1. A multi-scenario personnel flow monitoring system based on image analysis, comprising:
the scene acquisition module is used for acquiring scene videos of the monitored erection area through the erected monitoring equipment;
the scene analysis module is used for receiving the scene video acquired by the scene acquisition module and analyzing and obtaining the scene obtained by the erection of the monitoring equipment, and the specific operation steps are as follows:
capturing a frame of image in a scene video acquired in a receiving scene acquisition module, setting the captured frame of image as a scene reality image, carrying out similarity analysis and judgment on the scene reality image and a scene model image in a preset scene analysis module, wherein a scene model in the preset analysis module consists of four scene categories of a shopping center scene, a public transportation scene, a scenic spot scene and a commodity scene, five common scene model images are preset in each scene category, namely twenty scene model images in total, carrying out one-by-one comparison on the scene reality image and the twenty scene model images, and respectively extracting sift feature points of the scene reality image and the twenty scene model images;
after the number of the characteristic points of the scene reality image and the compared scene model image is obtained, comparing the number of the characteristic points of the scene reality image and the compared scene model image with a preset value, and when the number of the characteristic points of the scene reality image and the compared scene model image is smaller than the preset value, selecting a method which does not depend on the number of the characteristic points, namely based on histogram similarity calculation, to detect the scene reality image and the compared scene model image; when the number of the characteristic points of the scene reality image and the number of the characteristic points of the compared scene model image are larger than a preset value, performing similarity analysis on the scene reality image and the compared scene model image by a characteristic point matching method;
after the similarity of twenty scene model images consisting of the scene reality image and four scene categories is obtained respectively, twenty similarity values are obtained and respectively calibrated as、/>、Is->Comparing the twenty obtained similarity values to obtain a highest value; when the highest value is at +.>If the scene of the monitoring erection area is the shopping center scene, generating a shopping center mode switching signaling; when the highest value is at +.>If the traffic is in the middle, judging that the scene of the monitored erection area is a public transportation scene, and generating a public transportation mode switching signaling; when the highest value is atIf the scene of the monitored erection area is the scenic spot scene, generating a scenic spot mode switching signaling; when the highest value is at +.>If the scene of the monitored erection area is the commodity scene, commodity mode switching signaling is generated;
when the scene of the monitoring erection area is judged to be a shopping center scene, generating a shopping center mode switching signaling; when the scene of the monitoring erection area is judged to be a public transportation scene, a public transportation mode switching signaling is generated; generating scenic spot mode switching signaling when the scene of the monitored erection area is determined to be a scenic spot scene; when the scene of the monitoring erection area is judged to be a commodity scene, commodity mode switching signaling is generated;
the monitoring module is used for monitoring the people stream data of the scene through the monitoring equipment after the determination of the scene of the monitoring equipment which is determined to be erected is completed, and monitoring different people stream data according to the scene of the shopping center, the scene of public transportation, the scene of scenic spots and the scene of goods;
the mode switching module is used for performing mode switching on the mode switching signaling generated by the receiving scene analysis module and forming a corresponding monitoring mode;
and the processing module is used for processing and analyzing the received matched people stream data after the monitoring mode is determined to form corresponding optimization measures.
2. The system of claim 1, wherein the people stream data monitored in the monitoring module is composed of shopping center people stream data, public transportation people stream data, scenic spot people stream data and commodity people stream data; the shopping center people stream data comprise the number of business hall entrance and exit people, the number of business hall entrance and exit people and peak and leisure passenger flow; the public traffic people stream data comprises the number of people entering and exiting stations, the waiting number of people waiting for each road of vehicles and the Gao Pingfeng people stream number; scenic spot people stream data comprise the population number of scenic spots, population flow condition and population diversion number of branches; the commodity stream data comprises the population number of the regional in and out, commodity bag lifting rate and stream commodity station storage time, and the monitoring equipment in the monitoring module adopts a binocular camera.
3. The multi-scene personnel flow monitoring system based on image analysis according to claim 1, wherein the specific operation steps of the mode switching module are as follows:
when a shopping center mode switching signaling is received, switching to a shopping center personnel flow monitoring mode, and pertinently grabbing shopping center personnel flow data collected by the monitoring module; when a public transportation mode switching signaling is received, switching to a public transportation personnel flow monitoring mode, and pertinently grabbing scenic spot people flow data collected by the monitoring module; when receiving the scenic spot mode switching signaling, switching to a scenic spot personnel flow monitoring mode, and pertinently grabbing scenic spot personnel flow data collected by the monitoring module; when receiving the commodity mode switching signaling, switching to a commodity personnel flow monitoring mode, and pertinently grabbing commodity personnel flow data collected by the monitoring module.
4. The multi-scenario personnel flow monitoring system based on image analysis according to claim 1, wherein the specific operation steps of analysis and optimization of shopping center personnel flow data after the processing module determines the shopping center personnel flow data monitoring mode are as follows:
receiving the people stream data of the shopping center, and analyzing the people stream data of the shopping center to obtain the population number of the inlets and outlets of the mall, the data of the inlets and outlets of the mall, the peak and the leisure passenger flow; the acquired population quantity of the store entrance and population quantity of the store exit are subjected to difference to obtain the population quantity of the store in real time, the population quantity of the store in real time is set as an ordinate, a population retention line graph is established by taking the abscissa as a time point, the population quantity of the store in real time is intuitively and real-timely observed, a plurality of peak points and low peak points are selected from the population retention line graph, population quantities of the peak points and the low peak points and corresponding time are recorded, so that peak passenger flow quantity, matched peak passenger flow quantity, low peak passenger flow quantity and matched low peak time quantity are obtained, and staff quantity of the low peak time quantity in the store is reduced and filled up to the peak time;
the acquired population entering quantity of each shop and the acquired population exiting quantity of the shops are subjected to difference to obtain real-time survival people in each shop, the survival people at a plurality of time points in a single shop are utilized to obtain the survival people at different time points in the shops through mean value calculation to obtain the survival people at different time points in the shops, the survival people at different time points in the shops are sequenced from small to large, the survival people at different time points in the shops which are ranked in the front 20% and the rear 20% are selected respectively, the shops which are ranked in the front 20% and the rear 20% are marked, so that the shops with better and worse operation performance are obtained, the shops are classified according to business operation types, the sum of the survival people at different types of the shops is calculated and compared, the shops which are located in the first and last two business types are selected, the shops with corresponding types are marked, the proper and improper operation type shops or saturated type shops of the shops are judged, and the shops with different operation types and the proper operation type or the saturated operation type of the shops of the proper operation type and the operation type of the business type of the shop are arranged according to the shops.
5. The multi-scene personnel flow monitoring system based on image analysis according to claim 1, wherein the specific operation steps of analyzing and optimizing the public transportation personnel flow data after the processing module determines the public transportation personnel flow data monitoring mode are as follows:
when public transportation people stream data are received, analyzing the public transportation people stream data to obtain the number of people entering and exiting a station, the waiting number of people waiting for each road of vehicles and Gao Pingfeng people stream number; the method comprises the steps of subtracting the number of the station exit persons from the number of the station entrance persons to obtain the number of temporary persons in the station, counting and recording the number of the temporary persons in the station at each preset time node, simultaneously capturing a single frame of image of a monitored video according to the preset time point to obtain a plurality of images matched with the preset time node of the counted number of the temporary persons in the station, analyzing each image, carrying out graying and noise removing pretreatment on the images to strengthen the edges of foreground objects, and analyzing to obtain the passenger flow of each area in the images, so as to obtain the passenger flow distribution condition of each appointed area in the whole monitored area;
then, a line graph is established for a plurality of preset time node station temporary population as an ordinate and preset time nodes as an abscissa, a plurality of high peak values and low peak values are selected, and the station temporary population with the high peak values, the station temporary population with the low peak values and time points are recorded, namely Gao Pingfeng people stream number;
analyzing the waiting number of the collected vehicle population, comparing the waiting number of the multipath vehicle population, sequentially arranging the compared numerical values from large to small, selecting numerical values of a plurality of front digits and a plurality of rear digits, marking the corresponding vehicle numbering road numbers, properly increasing the vehicle numbering road numbers corresponding to the numerical values of the front digits, and reducing the vehicle numbering road numbers of the rear digits in a matched manner to achieve a more balanced state.
6. The multi-scene personnel flow monitoring system based on image analysis according to claim 1, wherein the specific operation steps of analyzing and optimizing the scene personnel flow data after the processing module determines the scene personnel flow data monitoring mode are as follows:
when receiving scenic spot people stream data, analyzing the scenic spot people stream data to obtain the population number of scenic spots, population flow condition and diversion population diversion number of branches; subtracting the scenic spot outgoing flow from the acquired scenic spot incoming flow to obtain scenic spot existing flow, respectively establishing a scenic spot existing population curve for the ordinate and the abscissa of the scenic spot existing population and setting a preset extremum of the scenic spot existing population, and stopping entering the population when the scenic spot existing population reaches or exceeds the preset extremum to form a state of only entering, so that the scenic spot people flow overload is avoided and accidents are avoided;
selecting a plurality of high peak points and low peak points in the established existing people number curve, recording corresponding time points, and correspondingly debugging the working time of staff in the scenic spot according to the high peak points and the local wind points of the existing people flow in the scenic spot;
summing the acquired scenic spot inflow and scenic spot outflow to obtain total traffic, matching the inflow population of the scenic spot with the action speed of the scenic spot outflow population, normalizing the three values of the total traffic of the scenic spot, the inflow population action speed of the scenic spot and the scenic spot outflow population action speed, obtaining a flow condition value by using a formula, establishing a histogram of the flow condition value according to time points, and controlling the scenic spot inflow speed to avoid scenic spot overload when the flow condition value is higher than a preset value;
and sorting the collected population diversion numbers of the plurality of branches from large to small to obtain the population diversion numbers of the branches at the later positions, marking the corresponding scenic spot branches, and performing playability or scenic optimization on the marked scenic spot branches to divert the branches with more population diversion numbers.
7. The multi-scene personnel flow monitoring system based on image analysis according to claim 1, wherein the specific operation steps of analysis and optimization of commodity people flow data after the processing module determines the commodity people flow data monitoring mode are as follows:
receiving commodity people stream data, analyzing the commodity people stream data to obtain the number of in-out population, commodity bag lifting rate and people stream commodity station storage time, calculating the obtained number of in-out population to obtain the existing people stream in a monitoring area, judging a high people stream time period and a low people stream time period of the monitoring area according to different existing people streams at different time points, and changing the loading and unloading time of a storage rack of the monitoring area according to the high people stream time period and the low people stream time period;
and substituting the normalized bag lifting rate of each commodity on the goods shelf and the time of each commodity people stream station into a formula to obtain the attention value of a single commodity, averaging the attention values of a plurality of commodities, calculating the average value by using the average value calculation, replacing and considering the commodity with the attention value lower than the average value, and improving the overall attention value of the plurality of commodities.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273261.XA CN116343116B (en) | 2023-03-20 | 2023-03-20 | Multi-scene personnel flow monitoring system based on image analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273261.XA CN116343116B (en) | 2023-03-20 | 2023-03-20 | Multi-scene personnel flow monitoring system based on image analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116343116A CN116343116A (en) | 2023-06-27 |
| CN116343116B true CN116343116B (en) | 2023-09-01 |
Family
ID=86894342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310273261.XA Active CN116343116B (en) | 2023-03-20 | 2023-03-20 | Multi-scene personnel flow monitoring system based on image analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116343116B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256462A (en) * | 2018-01-12 | 2018-07-06 | 北京航空航天大学 | A kind of demographic method in market monitor video |
| CN112347814A (en) * | 2019-08-07 | 2021-02-09 | 中兴通讯股份有限公司 | Passenger flow estimation and display method, system and computer readable storage medium |
| CN115081762A (en) * | 2022-08-24 | 2022-09-20 | 北京交通大学 | Passenger integrated intelligent travel method and system based on urban rail transit |
-
2023
- 2023-03-20 CN CN202310273261.XA patent/CN116343116B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256462A (en) * | 2018-01-12 | 2018-07-06 | 北京航空航天大学 | A kind of demographic method in market monitor video |
| CN112347814A (en) * | 2019-08-07 | 2021-02-09 | 中兴通讯股份有限公司 | Passenger flow estimation and display method, system and computer readable storage medium |
| CN115081762A (en) * | 2022-08-24 | 2022-09-20 | 北京交通大学 | Passenger integrated intelligent travel method and system based on urban rail transit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116343116A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108388888B (en) | Vehicle identification method and device and storage medium | |
| CN108596277B (en) | Vehicle identity recognition method and device and storage medium | |
| US9251425B2 (en) | Object retrieval in video data using complementary detectors | |
| CN107169415A (en) | Human motion recognition method based on convolutional neural networks feature coding | |
| CN109636795B (en) | Real-time non-tracking monitoring video remnant detection method | |
| CN104992177A (en) | Internet porn image detection method based on deep convolution nerve network | |
| CN107609485A (en) | The recognition methods of traffic sign, storage medium, processing equipment | |
| CN105069816B (en) | A kind of method and system of inlet and outlet people flow rate statistical | |
| CN104573625A (en) | Facial detection | |
| CN101711393A (en) | System and method based on the fire detection of video | |
| CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
| CN104298969A (en) | Crowd scale statistical method based on color and HAAR feature fusion | |
| CN109935080A (en) | The monitoring system and method that a kind of vehicle flowrate on traffic route calculates in real time | |
| CN108198202A (en) | A kind of video content detection method based on light stream and neural network | |
| CN107145819A (en) | A kind of bus crowding determines method and apparatus | |
| CN103489012B (en) | Crowd density detecting method and system based on support vector machine | |
| CN106845620A (en) | A kind of passenger flow counting method based on quene state analysis | |
| CN107403451A (en) | Adaptive binary feature monocular vision odometer method and computer, robot | |
| CN112598003B (en) | Real-time semantic segmentation method based on data expansion and full-supervision preprocessing | |
| CN105095837B (en) | A kind of TV station symbol recognition method and system | |
| CN112906590A (en) | FairMOT-based multi-target tracking pedestrian flow monitoring method | |
| CN110837786A (en) | Density map generation method and device based on spatial channel, electronic terminal and medium | |
| CN116343116B (en) | Multi-scene personnel flow monitoring system based on image analysis | |
| CN106779043A (en) | A kind of method of counting based on number of people detection | |
| CN113920585A (en) | Behavior recognition method and device, equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |