Remote inspection method and system for store

Abstract

The invention discloses a remote inspection method and system for stores, and relates to the technical field of remote inspection, and the method comprises the steps: determining the people flow state of each store according to the personnel distribution condition of a street sales area and the position of each store; the remote inspection system of the street sales area is determined according to the people flow state of each store, the position of each store and the area of the street sales area, and the accuracy of the remote inspection system of the street sales area is improved. In the remote inspection system, the abnormal area of the corresponding store is determined based on the remote inspection of each store, and the abnormal event of the store is determined according to the abnormal area and the store type of the corresponding store; the safety prevention and control area is determined based on the multiple abnormal events and the positions of the corresponding stores, and the corresponding safety prevention and control measures are determined based on the safety prevention and control area, the multiple abnormal events and the people flow states of the stores, so that the accuracy of the safety prevention and control measures is improved, and the remote inspection effect of the stores is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of remote inspection, and particularly relates to a method and a system for remotely inspecting stores. Background Art

[0002] With the development of technology, buildings are distributed in towns. Generally, the first floor of a building has streets and multiple stores. The multiple stores are arranged in sequence. The products sold in the multiple stores are different and have different customer groups. In the prior art, the multiple stores form the street sales area of the building. The street sales area presents each store and the corresponding pedestrian flow state. The inspection of the street sales area is usually a single inspection of the store, without considering the pedestrian flow state of each store and the location of each store, resulting in a low accuracy of the remote inspection system for the street sales area and unable to ensure the remote inspection effect of the store. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a method and a system for remotely inspecting stores.

[0004] An embodiment of the present invention provides a method for remotely inspecting stores, including: Collecting the locations of each store, and determining the corresponding street sales area according to the locations of each store and the store types; Determining the personnel distribution in the street sales area based on the personnel detection in the street sales area, and determining the pedestrian flow state of each store according to the personnel distribution in the street sales area and the locations of each store; Determining the remote inspection system for the street sales area according to the pedestrian flow state of each store, the locations of each store, and the area of the street sales area. The remote inspection system includes the inspection routes between each store; In the remote inspection system, determining the abnormal area of the corresponding store based on the remote inspection of each store, and determining the abnormal event of the store according to the abnormal area and the store type of the corresponding store; Determining the safety prevention and control area based on multiple abnormal events and the locations of the corresponding stores, and determining the corresponding safety prevention and control measures based on the safety prevention and control area, multiple abnormal events, and the pedestrian flow state of each store.

[0005] An embodiment of the present invention provides a remote inspection system for stores. The remote inspection system for stores is applied to the above method for remotely inspecting stores. The remote inspection system for stores includes: A street sales area module, configured to collect the locations of each store, and determine the corresponding street sales area according to the locations of each store and the store types; The pedestrian flow status module is used to determine the personnel distribution in the street sales area based on the personnel detection in the street sales area, and determine the pedestrian flow status of each store according to the personnel distribution in the street sales area and the locations of each store; The remote inspection system module is used to determine the remote inspection system of the street sales area according to the pedestrian flow status of each store, the locations of each store, and the area of the street sales area. The remote inspection system includes the inspection routes between each store; The abnormal event module is used to determine the abnormal area of the corresponding store based on the remote inspection of each store in the remote inspection system, and determine the abnormal event of the store according to the abnormal area and the store type of the corresponding store; The safety prevention and control measure module is used to determine the safety prevention and control area based on multiple abnormal events and the locations of the corresponding stores, and determine the corresponding safety prevention and control measures based on the safety prevention and control area, multiple abnormal events, and the pedestrian flow status of each store.

[0006] Compared with the prior art, the beneficial effects of the present invention are: In the embodiment of the present invention, by the method in the embodiment of the present invention, the locations of each store are collected, and the corresponding street sales area is determined according to the locations and store types of each store; the personnel distribution in the street sales area is determined based on the personnel detection in the street sales area, and the pedestrian flow status of each store is determined according to the personnel distribution in the street sales area and the locations of each store; the remote inspection system of the street sales area is determined according to the pedestrian flow status of each store, the locations of each store, and the area of the street sales area, which comprehensively considers the pedestrian flow status of each store, the locations of each store, and the area of the street sales area, and improves the accuracy of the remote inspection system of the street sales area.

[0007] Therefore, in the remote inspection system, the abnormal area of the corresponding store is determined based on the remote inspection of each store, and the abnormal event of the store is determined according to the abnormal area and the store type of the corresponding store; the safety prevention and control area is determined based on multiple abnormal events and the locations of the corresponding stores, and the corresponding safety prevention and control measures are determined based on the safety prevention and control area, multiple abnormal events, and the pedestrian flow status of each store. By introducing the abnormal events of the store, the overall consideration of the safety prevention and control area, multiple abnormal events, and the pedestrian flow status of each store is realized, the accuracy of the safety prevention and control measures is improved, and the remote inspection effect of the store is ensured. Description of the Drawings

[0008] Figure 1 is a schematic flowchart of the remote inspection method for stores in the embodiment of the present invention; Figure 2 is a schematic flowchart of step S11 in the remote inspection method for stores in the embodiment of the present invention; Figure 3 It is a schematic flowchart of step S12 in the remote inspection method for stores in the embodiments of the present invention; Figure 4 It is a schematic flowchart of step S13 in the remote inspection method for stores in the embodiments of the present invention; Figure 5 It is a schematic flowchart of step S14 in the remote inspection method for stores in the embodiments of the present invention; Figure 6 It is a schematic flowchart of step S15 in the remote inspection method for stores in the embodiments of the present invention; Figure 7 It is a schematic diagram of the structural composition of the remote inspection system for stores in the embodiments of the present invention. Detailed implementation manners

[0009] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0010] Please refer to Figures 1 to 7 , a remote inspection method for stores, which is applied to the remote inspection scenario of stores; the remote inspection method for stores includes: Step S11: Collect the locations of each store, and determine the corresponding street sales area according to the locations of each store and the store types; Step S12: Determine the personnel distribution in the street sales area based on the personnel detection in the street sales area, and determine the pedestrian flow state of each store according to the personnel distribution in the street sales area and the locations of each store; Step S13: Determine the remote inspection system for the street sales area according to the pedestrian flow state of each store, the locations of each store, and the area of the street sales area, and the remote inspection system includes the inspection routes between each store; Step S14: In the remote inspection system, determine the abnormal area of the corresponding store based on the remote inspection of each store, and determine the abnormal event of the store according to the abnormal area and the store type of the corresponding store; Step S15: Determine the safety prevention and control area based on multiple abnormal events and the locations of the corresponding stores, and determine the corresponding safety prevention and control measures based on the safety prevention and control area, multiple abnormal events, and the pedestrian flow state of each store; Refer to Figure 2 , in step S11, collect the locations of each store, and determine the corresponding street sales area according to the locations of each store and the store types; In the specific implementation process of the present invention, the specific steps are as follows: S111: Collect the name of the building, determine the distribution map of the building according to the name of the building and the building database, mark multiple stores in the distribution map of the building, and collect the positions of the multiple stores relative to the building. S112: In each store, trigger corresponding area detection based on the position of each store to collect the business area of each store, and determine the first sub-region according to the positions of the stores and the corresponding business areas of the stores. S113: Determine the second sub-region according to the positions of the stores and the store types, and determine the corresponding street sales region based on the first sub-region, the second sub-region and the street of the building. The street sales region presents the corresponding street characteristics.

[0011] In the embodiment of the present application, collecting the name of the building, determining the distribution map of the building according to the name of the building and the building database, marking multiple stores in the distribution map of the building, and collecting the positions of the multiple stores relative to the building, takes into account the overall consideration of the name of the building and the building database, and ensures the accuracy of the distribution map of the building.

[0012] At this time, collect the names of all important buildings in the target area. These buildings are landmark buildings, shopping centers, large office buildings or any structures that have a significant impact on the area. The collection methods include on-site surveys, consulting urban planning documents, using online map services (such as Google Maps, Baidu Maps), etc. At this time, on a commercial street, the names of buildings such as "XX Shopping Center", "YY Building", and "ZZ Bank" are collected.

[0013] Once the names of the buildings are collected, the next step is to use this information to match with the building database to obtain the geographical location information of the buildings. The building database contains detailed information such as the longitude and latitude coordinates, height, and usage of the buildings. Through the matching, draw the distribution map of the buildings on the map. At this time, use GIS software or an online map service, enter the collected building names into the search bar, find the corresponding geographical locations, and mark them on the map. For example, "XX Shopping Center" is located in the center of the commercial street, "YY Building" is adjacent to the north of the shopping center, and "ZZ Bank" is located at the east end of the commercial street.

[0014] After the distribution map of the buildings is completed, the next step is to mark the locations of the target stores on the map. These stores include various retail stores, restaurants, cafes, etc. When marking the locations, it is necessary to ensure accuracy for subsequent analysis. At the same time, it is also necessary to collect the location information of the stores relative to the surrounding buildings, such as distance, direction, etc. At this time, mark stores such as "Brand A Clothing Store" and "Cafe B" inside the "XX Shopping Center", and mark "Fast Food Restaurant C" on the first floor of the "YY Building". For each store, the relative location information relative to the entrance, elevator or main passage of the shopping center will also be recorded. For example, the "Brand A Clothing Store" is located on the right side of the atrium on the first floor of the shopping center, and "Cafe B" is adjacent to the left side of the atrium.

[0015] Furthermore, in each store, corresponding area detection is triggered based on the location of each store to collect the business area of each store. The first sub-region is determined according to the location of each store and the business area of the corresponding store, taking into account the overall consideration of the location of each store and the business area of the corresponding store, ensuring the accuracy of the first sub-region.

[0016] At this time, accurately measure or estimate the business area of each store. The area detection is carried out in various ways, including on-site actual measurement, remote sensing analysis using satellite images or aerial photos, or referring to the architectural design drawings of the store. The purpose of the area detection is to obtain the actual business area of each store, which is crucial for understanding the store scale, layout, and subsequent analysis of the flow state of people and formulation of safety prevention and control measures. At this time, on a commercial street, use a laser rangefinder or a tape measure to conduct on-site measurement of the "Starbucks Coffee Shop" in the "Shengshi Shopping Center" to obtain its business area. For the "KFC Fast Food Restaurant" on the first floor of the "Jinmao Building", if it is inconvenient to visit on-site, use high-resolution satellite images for area estimation, or contact the building management to obtain official area data.

[0017] After completing the area detection, the next step is to divide the first sub-region according to the location and business area information of the stores. The division of the first sub-region is based on factors such as the density of the stores, the size of the area, and the similarity of the business types. By dividing the sub-region, the distribution characteristics of the stores in different regions can be analyzed more carefully, providing a basis for subsequent safety management and formulation of marketing strategies. At this time, according to the business area and location information of the stores, the inside of the "Shengshi Shopping Center" is divided into the first sub-regions such as the "Large Retail Store Area" (such as the "ZARA Clothing Store") and the "Food and Beverage Service Area" (such as the "Starbucks Coffee Shop" and the "McDonald's Fast Food Restaurant"). For the first floor of the "Jinmao Building", it is divided into sub-regions such as the "Fast Food and Snack Area" (such as the "KFC Fast Food Restaurant") and the "Convenience Service Area" (such as the "FamilyMart Convenience Store").

[0018] Therefore, determine the second sub-region according to the locations and types of each store, and determine the corresponding street sales region based on the first sub-region, the second sub-region, and the street of the building. The street sales region presents the corresponding street characteristics, incorporates the overall considerations of the first sub-region, the second sub-region, and the street of the building, and ensures the accuracy of the corresponding street sales region.

[0019] At this time, conduct a more detailed classification of the stores based on their geographical locations and the types of goods or services they offer; the division of the second sub-region aims to identify groups of stores with similar business characteristics or complementarity, which helps to understand the commercial layout on the street, consumer behavior patterns, and potential market opportunities; for example, group stores that offer similar goods (such as multiple clothing stores) or complementary services (such as restaurants and cinemas) into the same sub-region; at this time, on the commercial street, it is found that inside the "Shengshi Shopping Center", in addition to the "Large Retail Store Area" and the "Food and Beverage Service Area", there is also a "Luxury Shopping Area" composed of multiple jewelry stores and watch stores; similarly, on the first floor of the "Jinmao Building", in addition to the "Fast Food and Snack Area" and the "Convenience Service Area", a "Beauty Product Area" composed of multiple beauty stores is identified. The division of these sub-regions is based on the types of stores and their relative positions within the building.

[0020] After determining the first sub-region and the second sub-region, the next step is to combine this information with the overall layout and characteristics of the street to generate the street sales region; the street sales region should reflect the uniqueness of different commercial areas on the street, pedestrian flow characteristics, and potential market demands; by integrating the first sub-region (based on area and location), the second sub-region (based on store types), and the characteristics of the street itself (such as traffic conditions, pedestrian flow density, cultural atmosphere, etc.), a more comprehensive understanding of the street's commercial ecosystem can be obtained, providing a basis for formulating subsequent business strategies; at this time, combine the "Large Retail Store Area", the "Food and Beverage Service Area", and the "Luxury Shopping Area" inside the "Shengshi Shopping Center" with other commercial facilities on the street (such as independent stores, outdoor billboards, etc.) to form a comprehensive street sales region, which is characterized by its diverse shopping options, high-quality goods and services, and active dining atmosphere; similarly, combine the "Fast Food and Snack Area", the "Convenience Service Area", and the "Beauty Product Area" on the first floor of the "Jinmao Building" with other relevant facilities on the street to form another street sales region with unique commercial characteristics.

[0021] In an embodiment of the present application, collect the sub-region matching table, and the sub-region matching table is shown in Table 1; Table 1 Sub-region Matching Table Store Number Store Name Store Type Location Description First Sub-region Second Sub-region 001 Starbucks Coffee Shop Food and Beverage 1st Floor of Shengshi Shopping Center Large Retail Store Area Food and Beverage Service Area 002 ZARA Clothing Store Retail 2nd Floor of Shengshi Shopping Center Large Retail Store Area Luxury Shopping Area 003 KFC Fast Food Restaurant Food and Beverage 1st Floor of Jinmao Tower Fast Food and Snack Area Fast Food and Snack Area 004 FamilyMart Convenience Store Retail 1st Floor of Jinmao Tower Convenience Service Area Convenience Service Area 005 Dior Cosmetics Store Retail (Beauty) 3rd Floor of Shengshi Shopping Center Luxury Shopping Area Beauty Product Area Combining the first sub-region (large retail store area, food and beverage service area, luxury shopping area) and the second sub-region (food and beverage service area, luxury shopping area), as well as street characteristics (such as high-end shopping atmosphere, diverse food and beverage options), it is determined that this area is a comprehensive high-end shopping and dining area; at the same time, combining the first sub-region (fast food and snack area, convenience service area) and the second sub-region (fast food and snack area, convenience service area), as well as street characteristics (such as convenient public transportation, high-density pedestrian flow), it is determined that this area is a fast food and convenience service area.

[0022] Reference Figure 3 , in step S12, based on the personnel detection in the street sales area, the personnel distribution in the street sales area is determined, and the pedestrian flow status of each store is determined according to the personnel distribution in the street sales area and the locations of each store; In the specific implementation process of the present invention, the specific steps are as follows: S121: Collect the street sales area, conduct personnel detection on the street sales area to present the markers of the people in the street sales area, and determine the personnel distribution in the street sales area according to the street sales area and the markers of the people; S122: Determine the personnel situation of each store based on the matching of the personnel distribution in the street sales area and the locations of each store; S123: Determine the personnel congestion coefficient of each store according to the personnel situation of each store and the area of each store, and determine the pedestrian flow status of each store based on the personnel congestion coefficient of each store, the location of each store, and the state mapping relationship.

[0023] In the embodiment of the present application, collecting the street sales area, conducting personnel detection on the street sales area to present the markers of the people in the street sales area, and determining the personnel distribution in the street sales area according to the street sales area and the markers of the people takes into account the overall consideration of the street sales area and the markers of the people, ensuring the accuracy of the personnel distribution in the street sales area.

[0024] At this time, collect the street sales area. At this time, on the commercial street, install multiple high-definition cameras, respectively covering key sales areas such as "Shengshi Shopping Center" and "Jinmao Tower". These cameras will capture pedestrians, vehicles and other activities on the street in real time, providing raw data for personnel detection.

[0025] Use computer vision techniques (such as deep learning models) to process the collected images or video data to identify and mark people in the street sales area. These technologies can automatically detect and track pedestrians, even in complex environments. Maintain high accuracy; at this time, use deep learning models such as YOLO (You Only Look Once) or SSD (Single Shot MultiBox Detector) to process the image data captured by the camera. These models will automatically detect and mark pedestrians in the image, while providing the location, size, and other relevant information of each pedestrian.

[0026] The detected people are visualized on the image or video in some form (such as marked boxes, heat maps, etc.), which helps to intuitively understand the distribution of people in the street sales area and provide an intuitive basis for subsequent analysis; at this time, the detected pedestrians are marked on the image in the form of marked boxes, or a heat map is used to show the density of people; heat maps usually use color depth to indicate the density of people in different areas, and the darker the color, the denser the population.

[0027] The marked people are counted and analyzed to determine the distribution of people in the street sales area, which includes calculating key indicators such as population density, flow direction, and residence time to fully understand the personnel activities in the street sales area; at this time, the number of people in each marked box is counted to calculate the population density in different areas; at the same time, the flow direction of pedestrians is also analyzed to understand from which direction they enter the sales area and their movement paths within the area; in addition, the residence time of pedestrians in a specific area is calculated to evaluate the attractiveness of the area.

[0028] Furthermore, the personnel situation of each store is determined based on the matching of the personnel distribution of the street sales area and the location of each store, which is compatible with the overall consideration of the matching of the personnel distribution of the street sales area and the location of each store, ensuring the accuracy of the personnel situation of each store.

[0029] At this point, the distribution of people in the street sales area is obtained from the previous personnel detection and analysis, which usually includes a personnel density map, a flow direction map, or a residence time map, etc., which provide detailed information about the activities of people on the street; at this point, a heat map is obtained from the camera data, which shows the density of people in different areas. This heat map will serve as the basis for subsequent analysis.

[0030] Collect and organize the precise location information of each store on the street, which is accomplished through Geographic Information System (GIS) or on-site surveys to ensure the accuracy of the location data of each store; meanwhile, obtain the precise coordinates and boundary information of "Shengshi Shopping Center", "Jinmao Building" and each store inside through the GIS system.

[0031] Match the personnel distribution in the street sales area with the location information of each store, which usually involves spatial analysis techniques such as buffer analysis or overlay analysis to determine the personnel density and activities around each store; at this time, use GIS software to create a buffer that expands a certain distance (such as 50 meters or 100 meters) around each store; then, conduct an overlay analysis of this buffer with the personnel density heat map to calculate the personnel density around each store.

[0032] Determine the number of people attracted by each store, the stay time and other relevant indicators, which will be used to evaluate the attractiveness of the store, the flow of people and potential business opportunities; meanwhile, it is found that the personnel density in the buffer around the "Starbucks Coffee Shop" is relatively high and the average stay time is relatively long, which indicates that the "Starbucks Coffee Shop" attracts a large number of people and customers tend to stay at this store for a long time; on the contrary, if the personnel density in the buffer around a certain store is relatively low or the stay time is relatively short, it indicates that the attractiveness of this store is insufficient or it needs to improve its marketing strategy.

[0033] Specifically, obtain the personnel distribution through the heat map and determine the location information of each store through the GIS system; next, use GIS software to create a 100-meter buffer for each store and conduct an overlay analysis of these buffers with the personnel density heat map; the analysis results show that the personnel density in the buffer around the "Starbucks Coffee Shop" is the highest and the average stay time is also the longest; in contrast, the personnel density in the buffer around a certain brand clothing store is relatively low and the average stay time is relatively short; based on these analysis results, draw the conclusion that the "Starbucks Coffee Shop" has a relatively high attractiveness in the "Shengshi Shopping Center" area, attracting a large number of people and maintaining a long customer stay time; while a certain brand clothing store needs to consider improving its marketing strategy or product portfolio to enhance its attractiveness and the flow of people.

[0034] Therefore, determine the personnel congestion coefficient of each store according to the personnel situation and the area of each store, and determine the flow state of each store based on the personnel congestion coefficient, the location and the state mapping relationship of each store. Considering the overall situation of the personnel congestion coefficient, the location and the state mapping relationship of each store, ensure the accuracy of the flow state of each store.

[0035] At this time, obtain the number of personnel (or density) of each store and the regional area of the store from the previous analysis; the number of personnel is obtained by counting the personnel in the buffer zone around the store, while the regional area is usually obtained through a GIS system or on-site measurement; at this time, in the "Prosperous Street" area of the "Flourishing Shopping Center", the number of personnel in the buffer zone around the "Starbucks Coffee Shop" and the regional area of this store have been obtained from the previous analysis.

[0036] Calculate the personnel congestion coefficient based on the number of personnel and the regional area of the store; the congestion coefficient is usually expressed as the number of personnel per square meter and is used to measure the congestion degree of the store; at this time, in the example of "Prosperous Street", assume that the regional area of the "Starbucks Coffee Shop" is 150 square meters and the number of personnel in the buffer zone around it is 50 people; then, the personnel congestion coefficient of the "Starbucks Coffee Shop" is 50 people / 150 square meters ≈ 0.33 people per square meter.

[0037] The state mapping relationship maps the personnel congestion coefficient to a specific pedestrian flow state, which usually involves setting a series of thresholds, and each threshold corresponds to a pedestrian flow state (such as unobstructed, slightly congested, congested, extremely congested, etc.); at this time, in the example of "Prosperous Street", the following state mapping relationship is set: congestion coefficient < 0.2 people per square meter: unobstructed; 0.2 people per square meter ≤ congestion coefficient < 0.5 people per square meter: slightly congested; 0.5 people per square meter ≤ congestion coefficient < 1 person per square meter: congested; congestion coefficient ≥ 1 person per square meter: extremely congested.

[0038] Compare the personnel congestion coefficient of each store with the state mapping relationship to determine its pedestrian flow state; at the same time, consider the location information of the store and further analyze the spatial distribution characteristics of the pedestrian flow state; at this time, in the example of "Prosperous Street", since the personnel congestion coefficient of the "Starbucks Coffee Shop" is 0.33 people per square meter, according to the state mapping relationship, its pedestrian flow state is determined to be "slightly congested"; at the same time, combined with the pedestrian flow states of other stores, analyze the pedestrian flow distribution characteristics of the entire "Flourishing Shopping Center" area, such as which areas have relatively dense pedestrian flow and which areas are relatively sparse.

[0039] Specifically, assume that in the "Prosperous Shopping Center" area of the "Bustling Street", the personnel congestion coefficients of each store have been calculated, and the state mapping relationship has been set; the following are the determination of the flow status of some specific stores: "Starbucks Coffee Shop": the personnel congestion coefficient is 0.33 people per square meter, and the flow status is "slightly congested"; "A certain brand clothing store": the personnel congestion coefficient is 0.15 people per square meter, and the flow status is "unobstructed"; "Food court": the personnel congestion coefficient is 0.6 people per square meter, and the flow status is "congested"; "Cinema": the personnel congestion coefficient is 0.08 people per square meter (considering the large internal space of the cinema, the congestion coefficient here is low, but the external queuing situation is not considered), and the flow status is "unobstructed"; combining this information, the flow distribution characteristics of the "Prosperous Shopping Center" area are obtained. For example, the flow of people around the "Starbucks Coffee Shop" and the "Food court" is relatively dense, while the flow of people around the "A certain brand clothing store" and the "Cinema" is relatively sparse. This helps the mall managers understand the flow of people, optimize the layout and marketing strategies.

[0040] Reference Figure 4 , in step S13, according to the flow status of each store, the location of each store, and the area of the street sales area, a remote inspection system for the street sales area is determined, and the remote inspection system includes the inspection routes between each store; In the specific implementation process of the present invention, the specific steps are as follows: S131: In the street sales area, determine the first training data based on the flow status of each store and the location of each store, and determine the second training data based on the flow status of each store and the area of the street sales area; S132: Collect the location of the street sales area, and determine the corresponding remote inspection training mode according to the location of the street sales area and the shape of the street sales area. Determine the remote inspection system of the street sales area based on the training of the remote inspection training mode, the first training data, and the second training data; S133: Detect the remote inspection system of the street sales area, and determine the inspection routes between each store according to the detection of the remote inspection system of the street sales area, and consider the influence of the flow status of each store.

[0041] In the embodiment of the present application, in the street sales area, determine the first training data based on the flow status of each store and the location of each store, and determine the second training data based on the flow status of each store and the area of the street sales area, which is compatible with the overall consideration of the flow status of each store and the area of the street sales area, and ensures the accuracy of the second training data.

[0042] At this time, the quantitative data (such as the number of people per square meter) or qualitative data (such as the level of the flow state) obtained through previous data analysis (such as in step S123) reflects the degree of crowding of people in each store during a specific period; the store location is the specific location of the store in the street sales area, and the precise coordinates are obtained through the Geographic Information System (GIS), which is also simplified to a relative location description (such as near the entrance, central area, corner, etc.); at the same time, the flow state and the store location are combined to form a set of training data, which is used to train the model in order to predict or optimize the store's operation strategy, personnel allocation, inspection route, etc. Optionally, collect the flow state data of each store during a specific period (such as weekend afternoons); obtain the precise location data of each store in the street sales area; combine the flow state data and the location data to form the first training data set.

[0043] Determine the second training data based on the flow state of the store and the area of the region. At this time, the area of the region is the total area of the street sales area, which is usually a fixed value, but may also vary due to different time periods (such as the area is expanded during the night market period due to the addition of temporary stalls); at the same time, the flow state and the area of the region are combined to form another set of training data, which is used to analyze the flow distribution of the entire sales area and the relationship between the flow density and the area of the region; Optionally, obtain the total area data of the street sales area; collect the flow state data of each store during a specific period; combine the flow state data and the area data of the region to form the second training data set.

[0044] Specifically, assume that in the street sales area of "Busy Street", there are the following stores: Store A: located in the center of the area, and the flow state on weekend afternoons is "crowded" (assuming 5 people per square meter); Store B: located at the edge of the area, and the flow state on weekend afternoons is "slightly crowded" (assuming 2 people per square meter); Store C: located in another corner, and the flow state on weekend afternoons is "uncongested" (assuming 0.5 people per square meter); the total area of the street sales area is 1000 square meters.

[0045] The first training data: Store A: location (center), flow state (crowded, 5 people / square meter); Store B: location (edge), flow state (slightly crowded, 2 people / square meter); Store C: location (corner), flow state (uncongested, 0.5 people / square meter). This set of data is used to train the model to predict the flow state of stores in different locations or optimize the operation strategy of the store. Second training data: Pedestrian flow density in the entire area: The total number of people on weekend afternoons is ((5 people per square meter × x square meters (area of Store A) + 2 people per square meter × y square meters (area of Store B) + 0.5 people per square meter × z square meters (area of Store C)) / 1000 square meters). Here, the specific area of each store needs to be known, or an average area can be assumed for estimation; assuming that the area of each store is the same, then the total number of people is (5 + 2 + 0.5) × average store area / 1000 = pedestrian flow density.

[0046] Furthermore, collect the location of the street sales area, and determine the corresponding remote patrol training mode based on the location and shape of the street sales area. The remote patrol system for the street sales area is determined through training based on this remote patrol training mode, the first training data, and the second training data. It takes into account the overall training of the remote patrol training mode, the first training data, and the second training data, ensuring the accuracy of the remote patrol system for the street sales area. At the same time, it takes into account the overall situation of the pedestrian flow status of each store, the location of each store, and the area of the street sales area, further improving the accuracy of the remote patrol system for the street sales area.

[0047] At this time, collect the location and shape information of the street sales area. At this time, the location information usually includes the longitude and latitude coordinates of the street sales area, the surrounding traffic conditions, the relative position to important landmarks (such as shopping malls, bus stops, etc.). This information helps to understand the specific location and surrounding environment of the sales area, providing a basis for the subsequent selection of the remote patrol training mode; the shape information includes the shape, size, store distribution, sidewalk width, etc. of the street sales area; the shape information is crucial for planning the patrol route and predicting the pedestrian flow distribution, etc. At this time, use GIS tools or on-site surveys to obtain the location information of the street sales area; collect the shape information of the sales area through maps, on-site photos or videos, etc.

[0048] Based on the location and shape information of the sales area, select a suitable remote patrol training mode, which includes a patrol mode based on pedestrian flow density, a patrol mode based on store importance, a patrol mode based on time period, etc.; at the same time, the basis for selecting the training mode usually includes the pedestrian flow characteristics of the sales area, store types, safety requirements, etc. For example, areas with dense pedestrian flow require more frequent patrols, while important stores (such as flagship stores, jewelry stores, etc.) require stricter patrol standards. Optionally, analyze the pedestrian flow data and store types in the sales area to determine the key points and difficulties of the patrol; according to the analysis results, select a suitable remote patrol training mode.

[0049] This includes the previously collected first training data (based on store footfall status and location) and second training data (based on store footfall status and area); combine the training data with the remote inspection training mode, and through machine learning or data analysis methods, construct a remote inspection system that should be able to guide inspectors on how to optimize the inspection routes and frequencies according to the footfall status in the sales area and store distribution; at the same time, use data analysis tools or machine learning frameworks to import and preprocess the training data; set the model parameters and training objectives according to the selected remote inspection training mode; conduct model training until satisfactory performance indicators (such as accuracy, recall, etc.) are achieved; use the trained model to generate a remote inspection system, including inspection routes, frequencies, key areas, etc.

[0050] Specifically, assume that in the sales area of "Busy Street", there are the following stores and the first and second training data have been collected: Store A (flagship store): located in the center of the area with dense footfall; Store B (restaurant): located at the edge of the area with moderate footfall; Store C (clothing store): located in another corner with less footfall; the total area of the sales area is 1000 square meters, approximately rectangular in shape, and the sidewalk width is moderate. Busy Street is located in the city center, surrounded by multiple shopping centers and bus stops, with convenient transportation; the sales area is rectangular in shape, the stores are evenly distributed, and the sidewalk width is moderate, facilitating inspections.

[0051] According to the footfall characteristics and store types in the sales area, select the inspection mode based on footfall density; areas with dense footfall (such as near Store A) require more frequent inspections, while areas with less footfall (such as near Store C) appropriately reduce the inspection frequency; at the same time, import the first training data (store footfall status and location) and second training data (area footfall density) into the data analysis tool; set the parameters and objectives of the machine learning model, such as using decision tree or random forest algorithms for classification and prediction; conduct model training until satisfactory performance indicators are achieved; use the trained model to generate a remote inspection system; for example, determine the inspection route to start from Store A and sequentially inspect Store B and Store C in a clockwise direction along the sidewalk; the inspection frequency is once an hour for Store A, and once every two hours for Store B and Store C; the key area is the area with dense footfall near Store A.

[0052] Therefore, the remote inspection system for the street sales area is tested, and based on the test of the remote inspection system for the street sales area, the inspection routes between each store are determined, and considering the influence of the footfall status of each store, the influence of the footfall status of each store is introduced.

[0053] At this time, ensure the effectiveness and accuracy of the remote inspection system, which includes verifying the rationality of the inspection route, the suitability of the inspection frequency, and the ability to respond promptly to changes in the passenger flow status of each store; at this time, it is carried out through methods such as simulated inspections, on-site tests, and data analysis; simulated inspections simulate the inspection situations at different time periods and passenger flow statuses to evaluate the reaction speed and accuracy of the system; on-site tests are to directly conduct inspections in the actual environment, collect data, and evaluate the effects; data analysis is to analyze the data collected during the inspection process to evaluate the performance of the system; optionally, design simulated inspection scenarios, including different time periods, passenger flow statuses, weather conditions, etc.; use the remote inspection system to conduct simulated inspections, record the inspection process, time, and results; analyze the results of the simulated inspections to evaluate the accuracy and efficiency of the system; according to the evaluation results, adjust and optimize the remote inspection system.

[0054] Based on the detection results of the remote inspection system and considering factors such as the location, passenger flow status, and importance of each store, plan the optimal inspection route, which should ensure that the inspection personnel can efficiently and accurately cover all stores while promptly responding to changes in the passenger flow status; at the same time, the inspection route should have a certain degree of flexibility to adapt to changes in different time periods and passenger flow statuses; for example, during peak passenger flow periods, it is necessary to increase the inspection frequency of crowded areas; while during periods with less passenger flow, the number of inspections can be appropriately reduced or the route can be adjusted; optionally, use GIS tools or map software to draw a map of the sales area and mark the locations of each store; based on the detection results of the remote inspection system and the passenger flow status of the stores, plan a preliminary inspection route; evaluate and adjust the preliminary route to ensure its rationality and effectiveness; during the actual inspection process, flexibly adjust the inspection route according to changes in the passenger flow status and actual situations.

[0055] Monitor the passenger flow status of each store in real-time or regularly, including passenger flow density, passenger flow speed, passenger flow direction, etc., which helps to promptly understand the passenger flow situation of the store and provide a basis for adjusting the inspection route; dynamically adjust the inspection route and frequency according to changes in the passenger flow status; for example, during peak passenger flow periods, increase the number of inspections of crowded areas; while during periods with less passenger flow, appropriately reduce the number of inspections or adjust to other areas; optionally, use passenger flow monitoring equipment or data analysis tools to collect the passenger flow status data of each store in real-time or regularly; analyze and predict the passenger flow status data to understand the passenger flow trend and changes of the store; according to changes in the passenger flow status, promptly adjust the inspection route and frequency; feedback the adjusted inspection route and frequency to the inspection personnel to ensure that they can promptly respond to changes in the passenger flow status.

[0056] Specifically, assume that in the sales area of "Busy Street", there are several stores as follows, and a remote inspection system has been established: Store A (flagship store): located in the center of the area with dense foot traffic; Store B (restaurant): located at the edge of the area with moderate foot traffic; Store C (clothing store): located in another corner with less foot traffic.

[0057] Design a simulated inspection scenario: Assume that the weekend afternoon is the peak foot traffic period, and there is dense foot traffic near Store A; while the weekday morning is the period with less foot traffic, and the foot traffic in each store is moderate; Use the remote inspection system for simulated inspections: On the weekend afternoon, the inspection system can accurately identify the areas with dense foot traffic near Store A and increase the inspection frequency; while on the weekday morning, it can maintain a moderate inspection frequency; Analyze the results of the simulated inspections: Evaluate the accuracy and efficiency of the system and find that the system can better adapt to the changes in different time periods and foot traffic conditions.

[0058] Based on the detection results of the remote inspection system and the store locations, plan a preliminary inspection route: Start from Store A and sequentially inspect Store B and Store C in a clockwise direction along the sidewalk; Evaluate and adjust the preliminary route: Considering the dense foot traffic near Store A on the weekend afternoon, increase the inspection frequency for this area; while on the weekday morning, maintain a moderate inspection frequency and route; At the same time, use foot traffic monitoring devices to collect the foot traffic status data of each store: It is found that the foot traffic density near Store A is relatively high on the weekend afternoon, while the foot traffic in Store B and Store C is moderate; The foot traffic in each store is moderate on the weekday morning; According to the changes in the foot traffic status, timely adjust the inspection route and frequency: On the weekend afternoon, increase the inspection frequency near Store A; while on the weekday morning, maintain a moderate inspection frequency and route; Feed back the adjusted inspection route and frequency to the inspection personnel: Ensure that they can respond to the changes in the foot traffic status in a timely manner and improve the inspection efficiency and accuracy.

[0059] In an embodiment of the present application, collect an inspection route matching table, and this inspection route matching table is shown in Table 2; Table 2 Inspection Route Matching Table Time Period Store A (Flagship Store) Store B (Food and Beverage Store) Store C (Clothing Store) Recommended Inspection Route Morning Moderate Foot Traffic Less Foot Traffic Less Foot Traffic A -> B -> C Noon Dense Foot Traffic Moderate Foot Traffic Moderate Foot Traffic A (Multiple Times) -> B -> C Afternoon Dense Foot Traffic Moderate Foot Traffic Less Foot Traffic A (Multiple Times) -> B -> (Optional C) Evening Moderate Foot Traffic Less Foot Traffic Less Foot Traffic A -> B -> C (Quickly) Reference Figure 5 , in step S14, in the remote inspection system, based on the remote inspections of each store, determine the abnormal areas of the corresponding stores, and determine the abnormal events of the stores according to the abnormal areas and the store types of the corresponding stores; In the specific implementation process of the present invention, the specific steps are as follows: S141: Monitor the remote inspection system in real time, and conduct remote inspections of each store along the inspection route in the remote inspection system to achieve remote inspections of each store; S142: In each store, determine multiple abnormal features based on the remote inspection of each store. The multiple abnormal features are in the same store. Determine the abnormal area of the corresponding store based on the multiple abnormal features and the sales area of the store. S143: Collect multiple abnormal areas, determine multiple abnormal contents based on the multiple abnormal areas and the locations of the corresponding stores, and determine the abnormal events of the store based on the multiple abnormal contents and the store types of the corresponding stores.

[0060] In the embodiments of the present application, the remote inspection system is monitored in real time, and each store is remotely inspected along the inspection route in the remote inspection system to implement the remote inspection of each store, and the remote inspection of each store is introduced.

[0061] At this time, the remote inspection system is monitored in real time, and remote inspection is carried out along the inspection route. At this time, according to factors such as the layout of the street sales area, store distribution, historical problems, etc., a reasonable inspection route is preset in advance. This is one or more routes to ensure that all stores can be covered; during the real-time monitoring process, according to the set inspection route, each store is remotely inspected in turn, which is realized through the interface operation of the monitoring software, such as switching the monitoring screen, adjusting the camera angle, etc. During the actual inspection process, if it is found that there are abnormalities or problems in some stores, the inspection route is adjusted temporarily, and these stores are given priority for in-depth inspection.

[0062] Ensure that the inspection route can cover all stores without omission; during the inspection process, the monitoring data of each store is analyzed in real time to timely discover potential problems or abnormalities; record the problems or abnormalities found during the inspection process and timely feedback them to relevant personnel for handling.

[0063] Specifically, assume that there are three stores, namely Store A, Store B, and Store C, on a bustling commercial street, which are engaged in clothing, catering, and electronic products respectively. These stores are all equipped with surveillance cameras and are connected to a remote inspection system; the staff in the monitoring center can see the monitoring screens of Store A, Store B, and Store C in real time through the remote monitoring software; the monitoring software can also collect the sensor data in the store in real time, such as the density of people flow, temperature, humidity, etc.

[0064] The preset inspection route is: Store A (clothing store) -> Store B (catering store) -> Store C (electronic product store); the staff follows this route and switches the monitoring screens in turn to remotely inspect each store; when inspecting Store B, it is found that the monitoring screen shows a dense flow of people, and there are people gathering near the cashier, with a problem of too long queues; so, the staff adjusts the inspection route temporarily and gives priority to in-depth inspection of Store B.

[0065] Through the patrol inspection, the staff ensured that all three stores, namely Store A, Store B, and Store C, were fully covered without omission; during the patrol inspection, problems existing in Store B were promptly discovered and recorded; subsequently, the staff reported the problems in Store B to the store management and suggested that they take measures to solve the problem of overly long queues, such as adding cash registers and optimizing the cash register process, etc.; through such an example, we clearly understood the specific operations and practical applications of each detail in Step S141; in actual operations, flexible adjustments and optimizations also need to be made according to specific circumstances to ensure the accuracy and effectiveness of the remote patrol inspection system.

[0066] Furthermore, in each store, multiple abnormal features are determined based on the remote patrol inspection of each store. The multiple abnormal features are in the same store. Based on the multiple abnormal features and the sales area of the store, the abnormal area of the corresponding store is determined, which takes into account the overall consideration of the multiple abnormal features and the sales area of the store, ensuring the accuracy of the abnormal area of the corresponding store.

[0067] At this time, collect the remote patrol inspection data of each store, which includes video streams, images, sensor readings, etc.; through intelligent analysis methods, identify multiple abnormal features from the patrol inspection data; the abnormal features include abnormal human behaviors (such as gathering, fighting), abnormal item states (such as missing, damaged), abnormal environmental parameters (such as too high temperature, too high humidity), etc.

[0068] Confirm whether the identified multiple abnormal features are in the same store, which is achieved by comparing the location information of the abnormal features with the layout diagram of the store; at the same time, it is also necessary to confirm that these abnormal features appear within the same time period to ensure their relevance.

[0069] Understand the sales area division of the store, including the locations of each shelf, cash register, rest area, etc.; compare the identified abnormal features with the sales area of the store to determine the specific area where these features are located, and this area is the abnormal area of the store.

[0070] Therefore, collect multiple abnormal areas, determine multiple abnormal contents based on the multiple abnormal areas and the locations of the corresponding stores, and determine the abnormal events of the store according to the multiple abnormal contents and the store types of the corresponding stores, which takes into account the overall consideration of the multiple abnormal contents and the store types of the corresponding stores, ensuring the accuracy of the abnormal events of the store.

[0071] At this time, after determining the multiple abnormal areas in the store, it is necessary to collect detailed data of these areas, which includes surveillance videos, images, sensor readings, customer feedback, etc.; organize the collected data to ensure the accuracy and integrity of the information; for video and image data, it is necessary to perform editing or screenshotting for subsequent analysis.

[0072] Analyze the specific locations of the abnormal areas within the store and their relationships with other areas of the store, which helps to understand the environment and background in which the abnormalities occur; combine the collected data to identify the specific content of the abnormal areas, including personnel behaviors (such as gathering, fighting), item statuses (such as missing, damaged), environmental parameters (such as too high temperature, too high humidity), etc.; at the same time, understand the type and main business of the store, which helps to understand the impact of the abnormal event on the operation of the store; combine the abnormal content and the store type to define specific abnormal events; the abnormal events can accurately reflect the nature, scope and impact of the abnormal situation.

[0073] Specifically, assume there is a coffee shop E located in the commercial area of the city center; coffee shop E is equipped with surveillance cameras and is connected to a remote inspection system; in previous inspections, several abnormal areas within coffee shop E have been identified; through the remote inspection system, surveillance videos and images of several abnormal areas within coffee shop E have been collected, and these areas include near the cashier desk, behind the bar counter, and a customer rest area; the collected videos and images are clipped and screenshot for subsequent analysis.

[0074] Analyze the locations of the abnormal areas and find that there are people gathering near the cashier desk and the queuing time is relatively long, there are employees busy behind the bar counter and some coffee machines are idle, and there are people arguing in the customer rest area; combine the surveillance videos and images to identify the abnormal content as too long queue at the cashier desk, uneven utilization rate of coffee machines, and customer disputes.

[0075] Understand that the type and main business of coffee shop E is to provide coffee and light food services; combine the abnormal content to define the following abnormal events: Too long queue at the cashier desk event: resulting in too long waiting time for customers, affecting customer satisfaction and the image of the coffee shop; Uneven utilization rate of coffee machines event: indicating low work efficiency of employees or unreasonable configuration of coffee machines, affecting the coffee making speed and quality; Customer disputes event: having a negative impact on the atmosphere of the coffee shop and even triggering more serious conflicts.

[0076] In an embodiment of the present application, collect an abnormal event matching table, and the abnormal event matching table is shown in Table 3: Table 3 Abnormal event matching table Store Type Abnormal Content Abnormal Event Coffee Shop People Gathering Near the Cashier Long Queue at the Cashier Event Coffee Shop Coffee Machine Idle Behind the Bar Uneven Utilization Rate of Coffee Machine Event Coffee Shop People Quarreling in the Customer Rest Area Customer Dispute Event Clothing Store Merchandise on Shelves Messily Arranged Improper Merchandise Management Event Clothing Store Dressing Room Door Lock Damaged Safety Hazard Event Electronics Store Products Missing in the Display Area Merchandise Loss Event Assume there is a coffee shop F, and the following abnormal areas are found in the remote inspection: Abnormal area 1: People gathering near the cashier desk; Abnormal area 2: Coffee machines idle behind the bar counter; According to the abnormal event matching table, it is determined that the abnormal event corresponding to abnormal area 1 is "Too long queue at the cashier desk event"; the abnormal event corresponding to abnormal area 2 is "Uneven utilization rate of coffee machines event".

[0077] Reference Figure 6, in step S15, determine the security prevention and control area based on multiple abnormal events and the locations of the corresponding stores, and determine the corresponding security prevention and control measures based on the security prevention and control area, multiple abnormal events, and the flow status of people in each store; In the specific implementation process of the present invention, the specific steps are as follows: S151: Collect multiple abnormal events and determine the corresponding first prevention and control area according to the multiple abnormal events; S152: Determine the second prevention and control area according to the multiple abnormal events and the locations of the corresponding stores, and determine the security prevention and control area based on the first prevention and control area, the second prevention and control area, and the street sales area; S153: Mark multiple abnormal events and the flow status of people in each store in the security prevention and control area, determine the first safety factor according to the shape of the security prevention and control area and the multiple abnormal events, determine the second safety factor according to the shape of the security prevention and control area and the flow status of people in each store, and determine the corresponding security prevention and control measures based on the first safety factor, the second safety factor, and the prevention and control measure mapping relationship.

[0078] In the embodiment of the present application, collecting multiple abnormal events and determining the corresponding first prevention and control area according to the multiple abnormal events takes into account the overall consideration of multiple abnormal events and ensures the accuracy of the corresponding first prevention and control area.

[0079] At this time, collect multiple abnormal events, analyze the nature of each abnormal event, such as whether it involves personal safety, property safety, service quality, etc.; evaluate the impact degree of the abnormal event on store operation and customer experience, including direct losses and potential risks; analyze whether there is a connection between abnormal events, such as whether they are caused by the same reason and whether they occur concentratedly within the same time period, etc.

[0080] Determine the first prevention and control area: According to the analysis results of the abnormal events, delimit one or more areas that need to be prioritized for prevention and control as the first prevention and control area. These areas are usually areas where abnormal events occur concentratedly, have a large impact range, or have high potential risks; clarify the characteristics of the first prevention and control area, such as the boundary, area, main facilities, and personnel flow situation, etc., in order to formulate targeted prevention and control measures in the follow-up.

[0081] Specifically, assume there is a large shopping mall G, which contains multiple floors and different types of stores, such as clothing stores, restaurants, electronic product stores, etc.; in the recent operation, multiple abnormal events have occurred in the shopping mall G, including customer property loss, disputes between employees and customers, product quality complaints, etc.

[0082] Collect data on multiple abnormal events from the shopping mall's monitoring system and customer service center, including the specific location, time, and value of lost customer property; the information of both parties, reasons for disputes, and handling results of employee-customer disputes; the types of goods, content of complaints, and progress of handling of product quality complaints; organize this data, remove duplicate information, classify similar events, and ensure the accuracy and integrity of the data.

[0083] Analysis reveals that customer property loss incidents mainly occur in the lower floors of the shopping mall, especially near clothing stores and restaurants; employee-customer dispute incidents are scattered across all floors, but are more frequent in electronics stores and restaurants; product quality complaints mainly occur in electronics stores and certain specific brand clothing stores; evaluate the impact of these events on the operation of the shopping mall and customer experience, and find that property loss incidents directly affect customers' sense of security and shopping experience, employee-customer disputes affect the image of the shopping mall and customer satisfaction, and product quality complaints damage the reputation and brand value of the shopping mall; analyze the correlations between events and find that some property loss incidents are related to poor employee management or monitoring blind spots, employee-customer disputes are related to inappropriate service attitudes or communication methods, and product quality complaints are related to loose supply chain management or product quality control.

[0084] Based on the analysis results of abnormal events, demarcate the area near clothing stores and restaurants on the lower floors of Shopping Mall G as the first prevention and control area, because these areas are high-incidence areas of property loss incidents and have a greater impact on customers' sense of security and shopping experience; clarify the boundary of the first prevention and control area as all clothing stores and restaurants on the lower floors and their surrounding passages, with an area of approximately 500 square meters. The main facilities include surveillance cameras, cash registers, fitting rooms, etc., and the personnel flow is relatively intensive.

[0085] Furthermore, determine the second prevention and control area based on the locations of multiple abnormal events and corresponding stores, and determine the safety prevention and control area based on the first prevention and control area, the second prevention and control area, and the street sales area, taking into account the overall situation of the first prevention and control area, the second prevention and control area, and the street sales area, to ensure the accuracy of the safety prevention and control area.

[0086] At this time, analyze the abnormal events in the first prevention and control area determined in S151 and their correlations with store locations; based on the distribution of abnormal events and store locations, identify areas with potential risks. These areas are adjacent to the first prevention and control area or have similar abnormal event characteristics to the first prevention and control area; demarcate the identified areas with potential risks as the second prevention and control area, and these areas need to cover abnormal event points more widely than the first prevention and control area to prevent the spread of abnormal events or the occurrence of new abnormal events.

[0087] Integrate the first prevention and control area and the second prevention and control area to form a larger prevention and control scope; compare the integrated prevention and control area with the street sales area to ensure that the safety prevention and control area not only covers the abnormal event points, but also takes into account the overall layout and pedestrian flow of the street sales activities; based on the consideration of the integrated prevention and control area and the street sales area, finally delimit a safety prevention and control area that not only covers the abnormal event points but also takes into account the overall safety of the street sales activities.

[0088] Specifically, assume that we have a shopping mall H located on a bustling street in the city center. There are multiple floors and various stores in the shopping mall, and there are also many retail stores on both sides of the street; in step S151, we have determined a floor (such as the first floor) inside the shopping mall H as the first prevention and control area because there have been multiple incidents of customer property loss on this floor recently.

[0089] Determine the second prevention and control area according to the abnormal events and store locations: analysis shows that the abnormal events in the first prevention and control area are mainly concentrated near clothing and jewelry stores, and the customer flow of these stores is relatively large; it is identified that there are also potential risks in similar stores (such as women's clothing stores, men's clothing stores, jewelry stores, etc.) on another floor (such as the second floor) adjacent to the first prevention and control area because the customer flow of these stores is also relatively large and the value of the goods is relatively high; therefore, delimit the area where these potentially risky stores are located on the second floor as the second prevention and control area.

[0090] Integrate the first floor (the first prevention and control area) and the second floor (the second prevention and control area) to form a prevention and control scope covering two floors inside the shopping mall; considering the street sales area outside the shopping mall, especially the stores on both sides of the street adjacent to the shopping mall entrance, these stores have a close pedestrian flow connection with the inside of the shopping mall and are also affected by the abnormal events inside the shopping mall; therefore, expand the prevention and control scope inside the shopping mall to the stores on both sides of the street adjacent to the shopping mall to form a larger safety prevention and control area. This area not only covers the abnormal event points inside the shopping mall but also takes into account the overall layout and pedestrian flow of the street sales activities to ensure overall safety.

[0091] Therefore, multiple abnormal events and the flow of people in each store are marked in the security control area, and the first safety factor is determined according to the form of the security control area and the multiple abnormal events. The second safety factor is determined according to the form of the security control area and the flow of people in each store. The corresponding security control measures are determined based on the first safety factor, the second safety factor, and the mapping relationship between the control measures. The overall consideration of the first safety factor, the second safety factor, and the mapping relationship between the control measures is compatible to ensure the accuracy of the corresponding security control measures. At the same time, the overall consideration of the security control area, multiple abnormal events, and the flow of people in each store is realized, which improves the accuracy of the security control measures and ensures the remote inspection effect of the store.

[0092] At this time, within the security control area, multiple abnormal events that have been identified previously are marked by marking the location, type, occurrence time and other information of the abnormal events on the map; at the same time, the flow of people in each store within the security control area is monitored in real time, which is accomplished by installing crowd counters, using video surveillance analysis software or using mobile data analysis; the flow of people includes key indicators such as crowd density, flow speed and direction.

[0093] Determine the first safety factor: Analyze the morphology of the security control area, including the area, shape, boundary characteristics, etc.; combine the type, number, distribution and historical data of abnormal events to evaluate the overall impact of abnormal events on the security control area; calculate the first safety factor based on the results of morphological analysis and abnormal event impact assessment. This first safety factor reflects the quantitative assessment of the potential risks of abnormal events to the security control area.

[0094] Determine the second safety factor: conduct an in-depth analysis of the flow of people in each store within the safety control area to identify peak hours, crowded areas and bottlenecks; based on the results of the flow analysis, assess the risks that the flow of people brings to the safety control area, such as stampedes and the risk of lost items; combine the results of the flow analysis and the flow risk assessment to calculate the second safety factor, which reflects the quantitative assessment of the potential risks of the flow of people to the safety control area.

[0095] Establish in advance a mapping relationship between prevention and control measures and safety factors, which includes the types of prevention and control measures, implementation methods, priorities, etc. corresponding to different safety factor ranges; select corresponding safety prevention and control measures based on the calculation results of the first safety factor and the second safety factor, as well as the mapping relationship between prevention and control measures; formulate a detailed implementation plan, including the execution time, execution personnel, required resources, etc. of the measures, and ensure the effective implementation of the measures.

[0096] Specifically, assume we have a large shopping mall I, and its security prevention and control area has been determined according to step S152; within the security prevention and control area of shopping mall I, the locations and types of recent customer property loss incidents, employee-customer dispute incidents, and product quality complaint incidents have been marked; through the installed video monitoring analysis software of the pedestrian flow counter, the pedestrian flow status of each store within the security prevention and control area has been monitored in real time, including pedestrian flow density, flow speed, and pedestrian flow direction.

[0097] Determine the first safety factor: Analyze the shape of the security prevention and control area of shopping mall I and find that the area has a regular shape and clear boundaries, but there are areas with dense pedestrian flow (such as near the entrance, elevator entrance, etc.); combined with the impact assessment of abnormal events, calculate that the first safety factor is 0.75, indicating that the abnormal events have a relatively high overall impact on the security prevention and control area.

[0098] Determine the second safety factor: Conduct an in-depth analysis of the pedestrian flow status and find that weekends and holidays are peak pedestrian flow periods, and there are areas with dense pedestrian flow in some popular stores (such as fashion clothing stores, electronics stores, etc.), and the pedestrian flow speed is relatively fast; based on the pedestrian flow risk assessment, calculate that the second safety factor is 0.65, indicating that the risk brought by the pedestrian flow status to the security prevention and control area is moderate.

[0099] According to the pre-established mapping relationship of prevention and control measures, a first safety factor of 0.75 corresponds to measures such as strengthening monitoring and patrol efforts and increasing warning signs; a second safety factor of 0.65 corresponds to measures such as optimizing the pedestrian flow guidance plan and setting up temporary evacuation channels; therefore, safety prevention and control measures such as strengthening monitoring and patrol efforts, increasing warning signs, optimizing the pedestrian flow guidance plan, and setting up temporary evacuation channels are selected; a detailed implementation plan is formulated, including specific measures such as increasing the patrol frequency of security personnel on weekends and holidays, adding monitoring cameras in areas with dense pedestrian flow, setting warning signs in prominent positions, and dynamically adjusting the elevator operation direction and floor stay time according to the pedestrian flow status, and ensuring the effective implementation of these measures.

[0100] Please refer to Figure 7 , Figure 7 which is a schematic structural composition diagram of the remote inspection system for stores in the embodiment of the present invention; the remote inspection system for stores includes: The street sales area module 21 is used to collect the locations of each store and determine the corresponding street sales area according to the locations and types of each store. The pedestrian flow status module 22 is used to determine the personnel distribution in the street sales area based on the personnel detection in the street sales area, and determine the pedestrian flow status of each store according to the personnel distribution in the street sales area and the locations of each store. The remote inspection system module 23 is used to determine the remote inspection system for the street sales area according to the pedestrian flow status of each store, the location of each store, and the area of the street sales area. The remote inspection system includes the inspection routes between each store; The abnormal event module 24 is used to determine the abnormal area of the corresponding store based on the remote inspection of each store in the remote inspection system, and determine the abnormal event of the store according to the abnormal area and the store type of the corresponding store; The safety prevention and control measure module 25 is used to determine the safety prevention and control area based on multiple abnormal events and the locations of the corresponding stores, and determine the corresponding safety prevention and control measures based on the safety prevention and control area, multiple abnormal events, and the pedestrian flow status of each store.

[0101] For any combination of the technical features of the above embodiments, for the sake of brevity of description, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

Claims

1. A remote inspection method for a store, characterized in that: include: Collect the location of each store and determine the corresponding street sales area based on the location and type of each store; Determine the personnel distribution of the street sales area based on personnel detection in the street sales area, and determine the flow of people in each store according to the personnel distribution in the street sales area and the location of each store; Determine the remote inspection system of the street sales area according to the flow of people in each store, the location of each store and the area of ​​the street sales area, and the remote inspection system includes the inspection routes between each store; In the remote inspection system, the abnormal area of ​​the corresponding store is determined based on the remote inspection of each store, and the abnormal event of the store is determined according to the abnormal area and the store type of the corresponding store; The security control area is determined based on multiple abnormal events and the locations of the corresponding stores, and the corresponding security control measures are determined based on the security control area, multiple abnormal events and the flow of people in each store.

2. The remote inspection method of a store according to claim 1, characterized in that: The collecting of the locations of each store and determining the corresponding street sales area according to the location and type of each store include: Collecting the name of the building, determining a distribution map of the building according to the name of the building and a building database, marking a plurality of stores in the distribution map of the building, and collecting the positions of the plurality of stores relative to the building; In each store, trigger corresponding area detection based on the location of each store to collect the business area of ​​each store, and determine the first sub-area according to the location of each store and the corresponding business area of ​​the store; The second sub-area is determined according to the location and type of each store, and the corresponding street sales area is determined based on the first sub-area, the second sub-area and the street of the building, and the street sales area presents the corresponding street characteristics.

3. The remote inspection method of a store according to claim 1, characterized in that: The determining of the personnel distribution of the street sales area based on the personnel detection in the street sales area, and the determining of the flow of people in each store according to the personnel distribution in the street sales area and the location of each store, includes: Collecting street sales areas, performing personnel detection on the street sales areas to present marks of people in the street sales areas, and determining the distribution of people in the street sales areas according to the marks of the street sales areas and people; Determine the personnel situation of each store based on the matching of the personnel distribution situation of the street sales area and the location of each store; The crowding coefficient of each store is determined according to the number of personnel in each store and the area of ​​each store, and the flow of people in each store is determined based on the crowding coefficient of each store, the location of each store and the state mapping relationship.

4. The remote inspection method of a store according to claim 1, characterized in that: The remote inspection system for determining the street sales area according to the flow of people in each store, the location of each store and the area of ​​the street sales area includes inspection routes between each store, including: In the street sales area, the first training data is determined based on the traffic status of each store and the location of each store, and the second training data is determined based on the traffic status of each store and the area of ​​the street sales area.

5. The remote inspection method of a store according to claim 4, characterized in that: The remote inspection system of the street sales area is determined based on the traffic status of each store, the location of each store and the area of ​​the street sales area. The remote inspection system includes the inspection routes between each store and also includes: Collecting the location of the street sales area, and determining a corresponding remote inspection training mode according to the location of the street sales area and the shape of the street sales area, and determining a remote inspection system for the street sales area based on the remote inspection training mode, the first training data, and the second training data; The remote inspection system of the street sales area is tested, and the inspection routes between each store are determined based on the detection of the remote inspection system of the street sales area, taking into account the impact of the flow of people in each store.

6. The remote inspection method of a store according to claim 1, characterized in that: In the remote inspection system, based on the remote inspection of each store, the abnormal area of ​​the corresponding store is determined, and the abnormal event of the store is determined according to the abnormal area and the store type of the corresponding store, including: Monitor the remote inspection system in real time, and conduct remote inspections of each store along the inspection routes in the remote inspection system to achieve remote inspections of each store.

7. The remote inspection method of a store according to claim 6, characterized in that: In the remote inspection system, based on the remote inspection of each store, the abnormal area of ​​the corresponding store is determined, and the abnormal event of the store is determined according to the abnormal area and the store type of the corresponding store, and further includes: In each store, multiple abnormal features are determined based on remote inspections of each store, the multiple abnormal features are in the same store, and the abnormal area of ​​the corresponding store is determined based on the multiple abnormal features and the sales area of ​​the store; A plurality of abnormal areas are collected, a plurality of abnormal contents are determined according to the locations of the plurality of abnormal areas and corresponding stores, and an abnormal event of the store is determined according to the plurality of abnormal contents and the store types of the corresponding stores.

8. The remote inspection method of a store according to claim 1, characterized in that: The method of determining a safety control area based on multiple abnormal events and the locations of corresponding stores, and determining corresponding safety control measures based on the safety control area, multiple abnormal events and the flow of people in each store, includes: Collect multiple abnormal events, and determine the corresponding first prevention and control area according to the multiple abnormal events; The second control area is determined based on multiple abnormal events and the locations of corresponding stores, and the safety control area is determined based on the first control area, the second control area and the street sales area.

9. The remote inspection method of a store according to claim 8, characterized in that: The method further includes: determining a safety control area based on multiple abnormal events and the locations of corresponding stores, and determining corresponding safety control measures based on the safety control area, multiple abnormal events and the flow of people in each store. Mark multiple abnormal events and the flow of people in each store in the security control area, determine a first safety factor based on the shape of the security control area and the multiple abnormal events, determine a second safety factor based on the shape of the security control area and the flow of people in each store, and determine corresponding security control measures based on the mapping relationship between the first safety factor, the second safety factor and the control measures.

10. A remote inspection system for stores, characterized in that: The remote inspection system of the store is applied to the remote inspection method of the store as claimed in any one of claims 1 to 9, and the remote inspection system of the store includes: The street sales area module is used to collect the location of each store and determine the corresponding street sales area according to the location and type of each store; A crowd flow status module is used to determine the distribution of people in the street sales area based on the detection of people in the street sales area, and to determine the crowd flow status of each store according to the distribution of people in the street sales area and the location of each store; A remote inspection system module is used to determine the remote inspection system of the street sales area according to the flow of people in each store, the location of each store and the area of ​​the street sales area. The remote inspection system includes inspection routes between each store; The abnormal event module is used to determine the abnormal area of ​​the corresponding store based on the remote inspection of each store in the remote inspection system, and determine the abnormal event of the store according to the abnormal area and the store type of the corresponding store; The safety control measures module is used to determine the safety control area based on multiple abnormal events and the locations of corresponding stores, and to determine the corresponding safety control measures based on the safety control area, multiple abnormal events and the flow of people in each store.

Images