Processing method, system and device for flowing vendor dredging area and medium

Abstract

The invention provides a processing method, system and device for a flowing vendor dredging area and a medium, and can be applied to the technical field of flowing vendor data processing. According to the method, after the streetscape image is obtained, the mobile vendor identification is carried out according to the streetscape image to obtain the spatial distribution data of the mobile vendor, so that data support of a large-range area can be provided for spatial design and site selection strategies without investigation, and the investigation cost is effectively saved; moreover, after the position preference factor and the negative external effect factor of the mobile vendor are analyzed according to the spatial distribution data, feature recognition is carried out according to the position preference factor and the negative external effect factor to obtain a feature recognition result, and the management cost of the mobile vendor is obtained at the same time. And the spatial layout of the mobile vendors is optimized according to the feature recognition result and the management cost to obtain the target dredging area of the mobile vendors, so that the accuracy of the optimized mobile vendor area demarcation result is higher, and the actual situation is better met.

Description

Technical Field

[0001] The present application relates to the technical field of mobile vendor data processing, and in particular to a method, system, device and medium for processing a mobile vendor diversion zone. Background Art

[0002] In the related technologies, the delineation of the regularized areas for mobile vendors is a method that combines literature research and actual research. It is to propose spatial design and site selection strategy suggestions after understanding and analyzing the current situation of mobile vendors, the characteristics of business activities, and the interests of multiple parties. However, this method has high research costs, cannot cover large areas of the city, and cannot give specific spatial range suggestions. Another method for delineating the regularized areas for mobile vendors is to build an indicator system based on policy and literature review, conduct spatial quantitative analysis through GIS, assign weights through expert consultation and entropy method, and determine appropriate areas and strategies. However, this method cannot capture the dynamic nature of mobile vendor activities and the weights of the division indicators are subject to human subjectivity.

[0003] In summary, the methods for demarcating regularized areas for mobile vendors in the prior art have low accuracy in demarcating areas, and the areas are quite different from the actual situation. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a method, system, device and medium for processing mobile vendor diversion areas, which can effectively improve the accuracy of mobile vendor area demarcation and make the division results more consistent with reality.

[0005] To achieve the above purpose, one aspect of an embodiment of the present application provides a method for processing a mobile vendor diversion zone, the method comprising the following steps:

[0006] Get Street View imagery;

[0007] Identify mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors;

[0008] Analyze the location preference factors and negative externality factors of the mobile vendors according to the spatial distribution data;

[0009] Performing feature identification according to the location preference factor and the negative externality factor to obtain a feature identification result, wherein the feature identification result includes the suitability of the mobile vendor and the negative externality of the mobile vendor;

[0010] Obtaining management costs of the mobile vendors;

[0011] The spatial layout of the mobile vendors is optimized according to the feature recognition results and the management costs to obtain the target diversion area for the mobile vendors.

[0012] In some embodiments, the step of identifying mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors includes:

[0013] Predicting preset area boundaries and target labels in the street view image by using a preset recognition network model;

[0014] Counting the number of mobile vendors in the preset area corresponding to the boundary of the preset area according to the target tag;

[0015] The spatial distribution data of the mobile vendors is generated according to the preset area boundaries and the number of the mobile vendors.

[0016] In some embodiments, performing feature recognition according to the location preference factor and the negative externality factor to obtain a feature recognition result includes:

[0017] Obtaining a first personalized feature attribution data set and a second personalized feature attribution data set;

[0018] Calculating the suitability of the mobile vendors' operation in each diversion zone according to the first personalized feature attribution data set and the location preference factor;

[0019] The negative externalities of the mobile vendors in each diversion area are calculated based on the second personalized feature attribution data set and the negative external effect factors.

[0020] In some embodiments, the optimizing the spatial layout of the mobile vendors according to the feature recognition results and the management cost includes:

[0021] Calculate the compactness function value of the spatial layout corresponding to the mobile vendors according to the management cost;

[0022] Calculating an optimization objective function according to the suitability of the mobile vendor's operation, the negative externalities of the mobile vendor and the compactness function value;

[0023] Calculating the occupancy probability of each grid cell in the spatial layout according to the optimization objective function;

[0024] The optimized boundary result of the spatial layout corresponding to the mobile vendors is determined according to the occupancy probability.

[0025] In some embodiments, the formula of the optimization objective function is as follows:

[0026] Utility=W s SW n N+W c C;

[0027] In the formula, W srepresents the weight of the suitability of the mobile vendor business, W n represents the weight of the negative externality of the mobile vendors, W c represents the weight of the compactness function value, S represents the function value of the suitability of mobile vendors, N represents the function value of the negative externality of the mobile vendors, and C represents the compactness function value;

[0028]

[0029] In some embodiments, calculating the occupancy probability of each grid unit in the spatial layout according to the optimization objective function includes:

[0030] Calculating pheromone parameters according to the optimization objective function;

[0031] Get the specific objective function value for each grid cell;

[0032] Calculate the heuristic parameter according to the specific objective function value;

[0033] The occupation probability of each grid cell in the spatial layout is calculated according to the pheromone parameters and the heuristic parameters.

[0034] In some embodiments, the calculating pheromone parameters according to the optimization objective function comprises:

[0035] Calculating the pheromone retention rate according to the optimization objective function;

[0036] The pheromone parameters are calculated according to the pheromone retention rate and the pheromone evaporation rate.

[0037] Another aspect of the present application provides a system for processing a mobile vendor diversion zone, the system comprising:

[0038] A first acquisition module, used to acquire a street view image;

[0039] A first recognition module is used to identify mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors;

[0040] An analysis module, used for analyzing the location preference factors and negative external effect factors of the mobile vendors according to the spatial distribution data;

[0041] A second identification module is used to perform feature identification according to the location preference factor and the negative external effect factor to obtain a feature identification result, wherein the feature identification result includes the suitability of the mobile vendor and the negative externality of the mobile vendor;

[0042] A second acquisition module is used to acquire the management cost of the mobile vendor;

[0043] The optimization module is used to optimize the spatial layout of the mobile vendors according to the feature recognition results and the management costs to obtain the target diversion area for the mobile vendors.

[0044] Another aspect of the present application provides a computer device, including:

[0045] at least one processor;

[0046] at least one memory for storing at least one program;

[0047] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0048] Another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0049] The embodiments of the present application include at least the following beneficial effects: after acquiring street view images, the present application identifies mobile vendors based on the street view images to obtain spatial distribution data of mobile vendors, thereby providing data support for spatial design and site selection strategies in a large area without the need for investigation, effectively saving investigation costs; and, after analyzing the location preference factors and negative external effect factors of mobile vendors based on the spatial distribution data, the present application performs feature identification based on the location preference factors and negative external effect factors to obtain feature identification results, and simultaneously obtains the management costs of the mobile vendors, and then optimizes the spatial layout of mobile vendors based on the feature identification results and the management costs to obtain target diversion areas for mobile vendors, thereby making the optimized mobile vendor area demarcation results more accurate and more in line with actual conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of a method for processing a mobile vendor diversion zone provided in an embodiment of the present application;

[0051] Figure 2 It is an application flow chart of the method for handling mobile vendor diversion areas provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application, they are only examples of devices and methods consistent with some aspects of the embodiments of the present application.

[0053] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0054] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0056] In the relevant technology, the street stall economy, as one of the most active parts of the informal economy, has always been regarded as illegal and difficult to manage by modern cities. The negative externalities of the street stall economy include affecting the city appearance, blocking traffic, weakening street safety and polluting the environment. These negative effects make urban managers unfriendly to mobile vendors. However, the street stall economy also has its positive effects, namely, eliminating the survival difficulties of low-income groups and immigrant groups, meeting the needs of low-income consumers, increasing the vitality of urban streets and even helping the recycling of urban garbage. Therefore, in the process of urbanization, the governance policies on mobile vendors in the street stall economy are also constantly being revised.

[0057] The combination of formalization management and eviction management has the potential to guide mobile vendors to engage in business activities, maximize their positive impact and mitigate their negative effects. However, the premise is that the spatial governance boundaries of both managements are precisely delineated. Spatial mismatch of governance boundaries will lead to the failure of formalization management, that is, the designated diversion zones are not popular with mobile vendors and thus become ineffective. Since the demarcation of diversion zones led by urban managers is mainly for the convenience of management and the reduction of harmful externalities, rather than trying to meet the needs of mobile vendors, consumers and citizens. Limited diversion zone sites are usually located in areas with few pedestrians and traffic. In addition, mobile vendors must pay management fees or license fees and operate regularly as formal merchants, which will seriously reduce their income and competitiveness. Therefore, how to determine the boundaries of the diversion zones for formalization management and fully respect the business logic of mobile vendors is a key issue in achieving successful governance.

[0058] At present, the demarcation of the regularized area for mobile vendors is a method that combines literature research and actual research. It is to propose spatial design and site selection strategy suggestions after understanding and analyzing the current situation of mobile vendors, the characteristics of business activities, and the interests of multiple parties. However, this method has high research costs, cannot cover large areas of the city, and cannot give specific spatial range suggestions. Another method for demarcating the regularized area for mobile vendors is to build an indicator system based on policy and literature review, conduct spatial quantitative analysis through GIS (Geographic Information System), assign weights through expert consultation and entropy method, and determine appropriate areas and strategies. However, this method cannot capture the dynamic nature of mobile vendor activities and the weight of the division index is subject to human subjectivity. In summary, the demarcation method for the regularized area of ​​mobile vendors in the prior art has low accuracy in demarcating areas and is quite different from the actual situation.

[0059] Based on this, the embodiments of the present application provide a method, system, device and medium for processing mobile vendor diversion areas. After acquiring street view images, the present application identifies mobile vendors based on the street view images to obtain spatial distribution data of mobile vendors, thereby providing data support for large-scale areas for spatial design and site selection strategies without surveys, effectively saving survey costs; and, after analyzing the location preference factors and negative external effect factors of mobile vendors based on the spatial distribution data, the present application performs feature identification based on the location preference factors and negative external effect factors to obtain feature identification results, and simultaneously obtains the management costs of the mobile vendors, and then optimizes the spatial layout of mobile vendors based on the feature identification results and the management costs to obtain the target diversion areas for mobile vendors, thereby making the optimized mobile vendor area demarcation results more accurate and more in line with actual conditions.

[0060] Reference Figure 1The present application embodiment provides a method for processing a mobile vendor diversion zone, which includes but is not limited to the following steps:

[0061] Step S110, obtaining a street view image;

[0062] Step S120, identifying mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors;

[0063] Step S130, analyzing the location preference factors and negative externality factors of mobile vendors according to the spatial distribution data;

[0064] Step S140, performing feature recognition according to the location preference factor and the negative externality factor to obtain a feature recognition result, wherein the feature recognition result includes the suitability of the mobile vendors and the negative externalities of the mobile vendors;

[0065] Step S150, obtaining the management cost of the mobile vendors;

[0066] Step S160: Optimize the spatial layout of mobile vendors according to the feature recognition results and the management cost to obtain the target diversion area for mobile vendors.

[0067] In the embodiment of the present application, the process of identifying mobile vendors according to street view images to obtain spatial distribution data of mobile vendors may include but is not limited to the following steps:

[0068] Predicting preset area boundaries and target labels in street view images through a preset recognition network model;

[0069] Count the number of mobile vendors in a preset area corresponding to the boundary of the preset area according to the target label;

[0070] Generate spatial distribution data of mobile vendors based on preset area boundaries and the number of mobile vendors.

[0071] It is understandable that the preset recognition network model can be a SIPSI (Street Informal Participant Spatial Investigation) model, which is based on the YOLO v4 deep learning neural network algorithm and uses 7545 manually labeled mobile vendor samples to train a detector. All street view images are input into the pre-trained detector for processing to obtain the predicted target label and preset area boundary. Subsequently, the number of mobile vendors in the preset area corresponding to the preset area boundary in each image is counted according to the target label, and recorded as the current count of the corresponding coordinate point, and then the spatial distribution data of mobile vendors is generated according to the preset area boundary and the number of mobile vendors. In addition, the preset recognition network model can also be other similar image target detection models, such as SSD, RCNN and other models.

[0072] Specifically, when choosing a business location, street vendors consider factors such as customer availability and maximizing commercial benefits. However, they also need to consider avoiding eviction and NIMBY effects, which may keep them away from core urban areas. In order to evaluate the micro-distribution characteristics of street vendors, this application uses spatial distribution data to analyze factors such as urban population aggregation, infrastructure distribution, housing rental levels, and traffic conditions in different spatiotemporal dimensions. An interpretable machine learning algorithm is then used to establish an accurate feature extraction model that characterizes the micro-operation preferences of street vendors and the spatial impact of negative externalities.

[0073] Ideally, the micro-location selection of mobile vendors only needs to consider the influencing factors' attraction to these vendors, such as their natural tendency to areas with high population density and high pedestrian traffic. However, in reality, influencing factors not only attract mobile vendors, but also have a repulsive effect. For example, mobile vendors may avoid areas under the jurisdiction of urban management departments. Therefore, the micro-location distribution of mobile vendors is mainly a state of equilibrium affected by various influencing factors. It can be divided into two aspects: the combined influence of various influencing factors that are conducive to street vending, which produces an attraction effect; and the repulsive effect generated by law enforcement agencies and residents based on the negative externalities associated with street vending. In addition, when demarcating the formalized area, the urban management authorities also need to consider their law enforcement capabilities and management costs. Therefore, the spatial scope of the formalized area should be neither too large nor too small to maintain appropriate compactness.

[0074] It can be understood that after obtaining the spatial distribution data of mobile vendors, this embodiment analyzes the location preference factors and negative external effect factors of mobile vendors according to the spatial distribution data, and performs feature identification based on the location preference factors and negative external effect factors to obtain feature identification results including the suitability of mobile vendors and the negative externalities of mobile vendors. Specifically, the location preference factor mainly affects the suitability of mobile vendors' operations. This embodiment adopts a multi-standard evaluation to evaluate the suitability target to reflect the comprehensive impact of various influencing factors on the attractiveness of mobile vendors. In order to attract more potential customers, mobile vendors usually choose locations with large traffic flow to operate. However, the urban population is mobile, and different groups gather in different locations at different time periods. For example, around the market, the high passenger flow during the day attracts mobile vendors to set up stalls and compete with regular merchants. In contrast, at night when the market is closed and there is no traffic, it loses its appeal to mobile vendors. Based on this, this embodiment calculates the suitability of mobile vendor operations in each diversion area by obtaining the location preference factor. As shown in Tables 1 and Figure 2 As shown, location preference factors include but are not limited to population, housing rent, intersections, subway stations, schools, road grades, pedestrian bridges, markets, parks and squares, etc.

[0075] Table 1

[0076]

[0077] As can be seen from Table 1, each location preference factor corresponds to a specific time period and selection criteria. This embodiment uses the XGBoost method to mathematically simulate the business behavior rules of mobile vendors, that is, to simulate the attractiveness of each location preference factor to mobile vendors, specifically as follows:

[0078] V=f(x 1 ,x 2 ,…,x i );

[0079] In the formula, V represents mobile vendors, x i Indicates the factors that influence the attractiveness of mobile vendors.

[0080] Then, the SHAP (SHapley Additive exPlanations) method is used to evaluate and interpret the contribution level of individual influencing factors in the XGBoost fitting results. The SHAP value quantifies the attractiveness of each factor, and its normalized value is used as the weight of the variable in the multi-criteria evaluation. The evaluation formula for the suitability target of mobile vendors in each diversion area is as follows:

[0081]

[0082] Among them, S represents the suitability of mobile vendors, w i represents the normalized SHAP value in the first personalized feature attribution dataset, x i Represents various location preference factors.

[0083] In addition, this embodiment can also use other machine learning methods and indicator importance to evaluate the impact of location preference factors on the spatial distribution of mobile vendors. Specifically, LightGBM, CatBoost, random forest, etc. can be used to fit the behavior rules of mobile vendors. LIME is used to replace SHAP to evaluate the importance of indicators.

[0084] In the present embodiment, the choice of business location by mobile vendors is affected by the NIMBY effect. In order to quantify the NIMBY effect in space, the present embodiment establishes the relationship between the business location and the negative external effect factors. Specifically, the negative external effect factors are shown in Table 2 and Figure 2 As shown:

[0085] Table 2

[0086]

[0087] As can be seen from Table 2, each negative externality factor corresponds to a specific time period and selection criteria. The exclusion effect between negative externality factors and the spatial distribution of mobile vendors can be modeled using XGBoost within a multi-criteria decision-making framework, as shown in the following formula:

[0088]

[0089] N represents the negative externality of mobile vendors in each diversion area, w j represents the normalized SHAP value in the second personalized feature attribution dataset, x j Represents negative external effect factors.

[0090] In the embodiment of the present application, due to the increased management cost caused by the excessive fragmentation of the formalized diversion area, the over-dispersed formalized diversion area may lead to a potential decline in law enforcement efficiency, constraint capacity and management scope, and too many isolated formalized areas may have an adverse impact on street vending because it runs counter to the spatial agglomeration nature of mobile vendors. This embodiment maximizes the compactness of the spatial layout of the formalized diversion area by establishing a compactness function value. The compactness function is the following formula:

[0091]

[0092] In the formula, represents the compactness function value during the optimization process, L Sum represents the perimeter of the normalized diversion area, L MaxSum and L MinSum Represents the maximum and minimum perimeters corresponding to the area constraint of the regularized grooming zone.

[0093] In the embodiment of the present application, after obtaining the suitability of mobile vendors, the negative externalities of mobile vendors, and the compactness function values, the mobile vendors' diversion area is spatially optimized based on this information. Specifically, this embodiment calculates the optimization objective function according to the suitability of mobile vendors' operations, the negative externalities of mobile vendors, and the compactness function values, and then spatially optimizes the mobile vendors' diversion area by optimizing the objective function.

[0094] Among them, the formula of optimization objective function is as follows:

[0095] Utility=W s SW n N+W c C;

[0096] In the formula, W s represents the weight of the suitability of the mobile vendor business, W n represents the weight of the negative externality of the mobile vendors, W crepresents the weight of the compactness function value, S represents the function value of the suitability of mobile vendors, N represents the function value of the negative externality of mobile vendors, and C represents the compactness function value;

[0097]

[0098] It is understandable that after obtaining the optimization objective function value, this embodiment uses the ant colony optimization algorithm (AOC) to construct a multi-objective spatial optimization model, and then iterates multiple times to determine the optimal regularized area. It is understandable that the spatial governance of mobile vendors is dynamic and can adapt to local conditions and changing environments. Therefore, the model needs to analyze different governance scenarios, formulate the optimal regularized area, and formulate a sophisticated spatiotemporal governance strategy.

[0099] Specifically, ACO starts with randomly distributed ants in the candidate area. Each ant occupies one grid cell in one iteration. As the optimization process ends, the final positions of the ants are merged into the optimal solution set of the regularized area. The task of the ants in each iteration is to identify grid cells with higher objective function values, which it can occupy. Among them, the process of calculating the occupation probability of each grid cell in the spatial layout according to the optimization objective function can be to calculate the pheromone parameters according to the optimization objective function, and after obtaining the specific objective function value of each grid cell, calculate the heuristic parameters according to the specific objective function value, and then calculate the occupation probability of each grid cell in the spatial layout according to the pheromone parameters and the heuristic parameters. Specifically, the process of calculating the pheromone parameters according to the optimization objective function can be to calculate the pheromone retention rate according to the optimization objective function, and then calculate the pheromone parameters according to the pheromone retention rate and the pheromone evaporation rate.

[0100] It can be understood that the calculation formula for the occupancy probability of each grid cell is as follows:

[0101]

[0102] In the formula, α and β represent the relative importance of pheromone and heuristic parameters, respectively, which are adjusted by the model user; allowed k is a taboo table, indicating cells that are allowed to be occupied by ants and are not occupied by other ants; the probability of being occupied Mainly determined by two parameters: the heuristic parameter η i and pheromone parameter τ i , the heuristic parameter η i and pheromone parameter τ i The value of can be calculated using the following formula:

[0103]

[0104] τ i (t+1)=τ i (t)(1-ρ)+△τ i (t);

[0105] △τ i (t) = r·Utilitty;

[0106] Among them, Suit i =max(0,W s SW n N) represents the objective function value of a specific grid cell, W s represents the weight of the suitability of the mobile vendor business, W n represents the weight of the negative externality of the mobile vendors, S represents the function value of the suitability of mobile vendors, N represents the function value of the negative externality of mobile vendors; parameter ρ represents the pheromone evaporation rate, and △τ i (t) represents the pheromone retention rate. The parameter r represents the ratio between the pheromone retention rate and the utility function value, which is usually set to 0.1.

[0107] After multiple iterations of the above model, this embodiment finally outputs the optimized boundary result of the spatial layout after the regularized diversion area is optimized. It can be seen that this embodiment can generate a governance decision plan for the mobile vendor diversion area in different spaces by inputting the spatial data layer representing the location suitability and negative externalities of different scenarios.

[0108] In addition, when optimizing the spatial distribution of the mobile vendors' diversion area, a random optimization algorithm can also be used for optimization. Among them, the random optimization algorithm uses a heuristic search algorithm that constantly imitates the behavior of social animals, including but not limited to particle swarm optimization algorithm, bacterial foraging algorithm, firefly algorithm, artificial fish school algorithm, etc.

[0109] In summary, the method provided in the embodiment of the present application achieves low-cost and accurate delineation of the spatial and temporal boundaries of spatial governance of regularization policies and eviction policies by combining machine learning algorithms and ant colony optimization algorithms based on multi-dimensional social perception data, ensures the objectivity and scientific nature of the optimization results, and balances the interests of mobile vendors, urban managers, and urban residents.

[0110] The present application also provides a processing system for mobile vendor diversion areas, the system comprising:

[0111] A first acquisition module, used to acquire a street view image;

[0112] The first recognition module is used to identify mobile vendors based on street view images to obtain spatial distribution data of mobile vendors;

[0113] An analysis module is used to analyze the location preference factors and negative externality factors of mobile vendors based on spatial distribution data;

[0114] The second identification module is used to perform feature identification according to the location preference factor and the negative externality factor to obtain a feature identification result, wherein the feature identification result includes the suitability of the mobile vendors and the negative externality of the mobile vendors;

[0115] The second acquisition module is used to obtain the management cost of mobile vendors;

[0116] The optimization module is used to optimize the spatial layout of the mobile vendors according to the feature recognition results and the management cost to obtain the target diversion area for the mobile vendors.

[0117] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0118] The present application also provides a computer device, which includes:

[0119] at least one processor;

[0120] at least one memory for storing at least one program;

[0121] When at least one program is executed by at least one processor, the at least one processor implements the above method.

[0122] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0123] In addition, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and the computer program implements the above method when executed by a processor.

[0124] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0125] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for handling a mobile vendor diversion area, characterized in that: The method comprises the following steps: Get Street View imagery; Identify mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors; Analyze the location preference factors and negative externality factors of the mobile vendors according to the spatial distribution data; Performing feature identification according to the location preference factor and the negative externality factor to obtain a feature identification result, wherein the feature identification result includes the suitability of the mobile vendor and the negative externality of the mobile vendor; Obtaining management costs of the mobile vendors; The spatial layout of the mobile vendors is optimized according to the feature recognition results and the management costs to obtain the target diversion area for the mobile vendors.

2. The method according to claim 1, characterized in that The step of identifying mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors includes: Predicting preset area boundaries and target labels in the street view image by using a preset recognition network model; Counting the number of mobile vendors in the preset area corresponding to the boundary of the preset area according to the target tag; The spatial distribution data of the mobile vendors is generated according to the preset area boundaries and the number of the mobile vendors.

3. The method according to claim 1, characterized in that The performing feature recognition according to the location preference factor and the negative external effect factor to obtain a feature recognition result includes: Obtaining a first personalized feature attribution data set and a second personalized feature attribution data set; Calculating the suitability of the mobile vendors' operation in each diversion zone according to the first personalized feature attribution data set and the location preference factor; The negative externalities of the mobile vendors in each diversion area are calculated based on the second personalized feature attribution data set and the negative external effect factors.

4. The method according to claim 3, characterized in that The optimizing the spatial layout of the mobile vendors according to the feature recognition results and the management cost includes: Calculate the compactness function value of the spatial layout corresponding to the mobile vendors according to the management cost; Calculating an optimization objective function according to the suitability of the mobile vendor's operation, the negative externalities of the mobile vendor and the compactness function value; Calculating the occupancy probability of each grid cell in the spatial layout according to the optimization objective function; The optimized boundary result of the spatial layout corresponding to the mobile vendors is determined according to the occupancy probability.

5. The method according to claim 4, characterized in that The formula of the optimization objective function is as follows: Utility=W s S-W n N+W c C; In the formula, W s represents the weight of the suitability of the mobile vendor business, W n represents the weight of the negative externality of the mobile vendors, W c represents the weight of the compactness function value, S represents the function value of the suitability of mobile vendors, N represents the function value of the negative externality of the mobile vendors, and C represents the compactness function value; 6. The method according to claim 4, characterized in that The calculating the occupancy probability of each grid unit in the spatial layout according to the optimization objective function includes: Calculating pheromone parameters according to the optimization objective function; Get the specific objective function value for each grid cell; Calculate the heuristic parameter according to the specific objective function value; The occupation probability of each grid cell in the spatial layout is calculated according to the pheromone parameters and the heuristic parameters.

7. The method according to claim 6, characterized in that The calculating of pheromone parameters according to the optimization objective function comprises: Calculating the pheromone retention rate according to the optimization objective function; The pheromone parameters are calculated according to the pheromone retention rate and the pheromone evaporation rate.

8. A processing system for mobile vendors' diversion area, characterized in that: The system comprises: A first acquisition module, used to acquire a street view image; A first recognition module is used to identify mobile vendors according to the street view image to obtain spatial distribution data of the mobile vendors; An analysis module, used for analyzing the location preference factors and negative external effect factors of the mobile vendors according to the spatial distribution data; A second identification module is used to perform feature identification according to the location preference factor and the negative external effect factor to obtain a feature identification result, wherein the feature identification result includes the suitability of the mobile vendor and the negative externality of the mobile vendor; A second acquisition module is used to acquire the management cost of the mobile vendor; The optimization module is used to optimize the spatial layout of the mobile vendors according to the feature recognition results and the management costs to obtain the target diversion area of ​​the mobile vendors.

9. A computer device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images