CN114022772A - Method, system, device and storage medium for predicting spatial distribution of mobile vendor - Google Patents
Method, system, device and storage medium for predicting spatial distribution of mobile vendor Download PDFInfo
- Publication number
- CN114022772A CN114022772A CN202111353459.6A CN202111353459A CN114022772A CN 114022772 A CN114022772 A CN 114022772A CN 202111353459 A CN202111353459 A CN 202111353459A CN 114022772 A CN114022772 A CN 114022772A
- Authority
- CN
- China
- Prior art keywords
- mobile
- subset
- street view
- spatial distribution
- vendor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012549 training Methods 0.000 claims description 46
- 238000012795 verification Methods 0.000 claims description 20
- 238000011176 pooling Methods 0.000 claims description 9
- 239000000969 carrier Substances 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 9
- 238000001514 detection method Methods 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 239000004677 Nylon Substances 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 229920001778 nylon Polymers 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a system and a device for predicting the spatial distribution of mobile vendors and a storage medium, which can be applied to the technical field of image processing. The method comprises the following steps: acquiring road network data of a preset area and a plurality of street view images corresponding to preset coordinate points in the road network data; storing a plurality of street view images into a first subset and a second subset respectively, and carrying out type marking on the street view images in the first subset by a mobile vendor; inputting the first subset and the second subset into a mobile vendor identification model to obtain the spatial position of mobile vendors in a preset area; acquiring an association factor of a mobile vendor; and inputting the spatial position and the correlation factor of the mobile vendors in the preset area into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendors in the area to be predicted. The invention can reduce labor cost and workload, and can quickly and accurately obtain the distribution state of the mobile booths in different areas.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method, a system, a device and a storage medium for predicting the spatial distribution of mobile vendors.
Background
The stall economy is an indispensable component of the urban social ecosystem: on the one hand, it helps to alleviate the problem of large low income or new immigration loss; on the other hand, the method can meet the requirements of low consumers and improve the vitality of urban communities. Therefore, the distribution range of the mobile vendors in the stall economy is reasonably guided, and the negative effects of the mobile vendors are favorably reduced. In the related art, the knowledge of the distribution state of a mobile vendor is mainly performed by the following two ways: first, manual individual case investigation: the method comprises the steps of selecting a typical area where mobile vendors are active, collecting personal case data of the vendors, consumers, community residents and city law enforcement personnel through observation, interview, visit and other investigation modes, and analyzing the distribution state of the mobile vendors in the current area through the collected data. Such a survey method requires high labor cost, and the survey data can only reflect the state of the surveyed area. Second, urban census: namely analyzing the scale and distribution state of the vendor by means of census data. The nature of such a survey method also relies on a process of manually collecting data, and therefore, the collected data can reflect only the state of the area to be surveyed.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art. Therefore, the invention provides a method, a system, a device and a storage medium for predicting the spatial distribution of mobile vendors, which can quickly obtain the distribution states of mobile booths in different areas without the help of artificially collected data.
In one aspect, an embodiment of the present invention provides a method for predicting spatial distribution of a mobile vendor, including the following steps:
acquiring road network data of a preset area;
acquiring a plurality of street view images corresponding to preset coordinate points in the road network data;
storing the street view images into a first subset and a second subset respectively, and performing type marking on the street view images in the first subset by a mobile vendor;
inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area;
acquiring an association factor of a mobile vendor;
and inputting the spatial position of the mobile vendors in the preset area and the correlation factor into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendors in the area to be predicted.
In some embodiments, the inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the predetermined area includes:
extracting partial data of the first subset and partial data of the second subset and saving the partial data and the residual data of the first subset to a training set, and saving the residual data of the first subset to a verification set;
inputting the training set into a mobile vendor identification model, and sequentially extracting features, pooling the features and fusing the features of street view images in the training set to obtain a first prediction category and a first boundary frame of a mobile vendor on the street view images in the training set;
inputting the verification set into a mobile vendor identification model, and sequentially extracting features, pooling the features and fusing the features of street view images in the verification set to obtain a second prediction category and a second boundary frame of mobile vendors on the street view images in the verification set;
determining a confidence score and an intersection region corresponding to the second bounding box;
and when the confidence score is greater than or equal to a confidence threshold and the intersection area is greater than or equal to an intersection area threshold, inputting the street view image into a mobile vendor identification model, and obtaining an identification result of the mobile vendor in the preset area after sequentially extracting features, pooling features and fusing features of the street view image.
In some embodiments, prior to performing the step after inputting the training set into a mobile vendor recognition model, the method further comprises the steps of:
turning street view images in the training set to obtain turned images;
zooming the turned image to obtain a zoomed image;
and performing color gamut transformation on the scaled image to obtain a target image for training.
In some embodiments, the inputting the spatial positions of the mobile vendors in the preset area and the correlation factors into a spatial distribution prediction model to predict the spatial distribution state of the mobile vendors in the area to be predicted includes:
inputting the spatial position of the mobile vendor in the preset area into a spatial distribution prediction model to obtain an ROC curve of model performance;
and when the ROC curve meets the preset requirement, inputting the correlation factor into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendor in the region to be predicted.
In some embodiments, the acquiring a plurality of street view images corresponding to preset coordinate points in the road network data includes:
setting preset coordinate points at preset intervals in the road network data;
and acquiring a plurality of street view images corresponding to the preset coordinate points.
In some embodiments, the acquiring a plurality of street view images corresponding to the preset coordinate point includes:
and acquiring a plurality of street view images at different shooting angles on the preset coordinate point.
In some embodiments, the type tagging of the street view image within the first subset by the floating vendor comprises:
extracting cargo carriers within the street view image within a first subset;
labeling the type of the mobile vendor according to the cargo carrier.
In another aspect, an embodiment of the present invention provides a method for predicting spatial distribution of a mobile vendor, including:
the first acquisition module is used for acquiring road network data of a preset area;
the second acquisition module is used for acquiring a plurality of street view images corresponding to preset coordinate points in the road network data;
the decomposition module is used for respectively storing the street view images into a first subset and a second subset, and performing type marking on the street view images in the first subset by a mobile vendor;
the identification module is used for inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area;
a third obtaining module, configured to obtain an association factor of a mobile vendor;
and the prediction module is used for inputting the spatial positions of the mobile vendors in the preset area and the correlation factors into a spatial distribution prediction model to predict and obtain the spatial distribution state of the mobile vendors in the area to be predicted.
In another aspect, an embodiment of the present invention provides an apparatus for predicting spatial distribution of a mobile vendor, including:
at least one memory for storing a program;
at least one processor configured to load the program to perform the method for predicting the spatial distribution of a mobile vendor.
In another aspect, the present invention provides a storage medium, in which a computer-executable program is stored, and the computer-executable program is executed by a processor to implement the method for predicting the spatial distribution of a mobile vendor.
The prediction method for the space distribution of the mobile vendor provided by the embodiment of the invention has the following beneficial effects:
in the embodiment, a plurality of street view images corresponding to the preset coordinate point on the road network data of the preset area are obtained, and the plurality of street view images are respectively stored in the first subset and the second subset, and performing type labeling of the mobile vendor on the street view images in the first subset to improve the training precision of the recognition model, then inputting the first subset and the second subset into a mobile vendor identification model to obtain the spatial location of mobile vendors in a preset area, so as to provide location reference data for the mobile vendor distribution prediction in other areas, then inputting the spatial position of the mobile vendors in the preset area and the obtained correlation factors into a spatial distribution prediction model, predicting to obtain the spatial distribution state of the mobile vendors in the area to be predicted, therefore, data do not need to be artificially collected, the labor cost and the workload are reduced, and meanwhile, the distribution states of the mobile booths in different areas can be quickly and accurately obtained.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The invention is further described with reference to the following figures and examples, in which:
fig. 1 is a flowchart of a method for predicting spatial distribution of mobile vendors according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a mobile vendor identification model according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
In the description of the present invention, the meaning of a plurality is one or more, the meaning of a plurality is two or more, and the above, below, exceeding, etc. are understood as excluding the present numbers, and the above, below, within, etc. are understood as including the present numbers. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, unless otherwise explicitly defined, terms such as set, etc. should be broadly construed, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the detailed contents of the technical solutions.
In the description of the present invention, reference to the description of the terms "one embodiment," "some embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples," etc., means that a particular feature or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Referring to fig. 1, an embodiment of the present invention provides a method for predicting spatial distribution of a mobile vendor. The embodiment can be applied to the background processor corresponding to the server or the vendor management platform.
In the implementation process, the embodiment includes the following steps:
and S11, acquiring road network data of the preset area.
In this embodiment, the road network data includes city road network data, and the road network data may be directly called from a preset website, for example, from an OpenStreetMap. The OpenStreetMap is abbreviated as OSM, and is a world map which can be freely edited. Maps on OSM are drawn by a user holding a GPS device, aerial photographs, satellite imagery, other free content, even with the user having spatial knowledge due to familiarity with the target area.
And S12, acquiring a plurality of street view images corresponding to the preset coordinate points in the road network data.
In this embodiment, after the road network data is acquired, preset coordinate points are set on the road network data at preset intervals, for example, 10 meters is set as one interval, 20 meters is set as one interval, and the like. The size of the preset interval can be set by a user according to the actual situation of the road network data. After coordinate points on the current road network data are set, street view images corresponding to the coordinate points are collected. Specifically, the image collected at each coordinate point includes at least three images. The at least three images may be street view images corresponding to the same angle, or street view images corresponding to three different shooting angles. When the street view image is acquired from street view open service of the online map, web crawler software can be opened through an API of the online map, and the captured images at different angles corresponding to the preset coordinate points on the road network data are crawled. For example, street view images with shooting angles of 90 °, 180 °, and 270 °, respectively, are crawled.
And S13, storing the street view images into a first subset and a second subset respectively, and performing type marking on the street view images in the first subset.
In this embodiment, in order to improve the training accuracy of the mobile vendor identification model, after a plurality of street view images are acquired, the street view images are respectively stored in two different subsets, and then different processing methods are adopted for data in the two subsets. For example, street view images within the first subset are subject to the type criteria of a floating vendor, but street view images within the second subset are not processed to provide different types of data for model training. It will be appreciated that because different types of mobile vendors use different placement tools, the mobile vendors can be identified by first identifying the cargo carriers contained on the street view image within the first subset and then determining the type of mobile vendor based on the identified cargo carriers. For example, if the identified cargo carrier is a nylon bag that is placed on the ground and the cargo is placed on the nylon bag, then it can be determined that the type of mobile vendor is a ground booth. For another example, if the cargo carrier is a table and the cargo is placed on the table, it can be determined that the type of the mobile vendor is a table booth. If the cargo carrier is a tricycle and the cargo is placed in the tricycle, it can be determined that the type of the mobile vendor is a tricycle booth. If it is also recognized that the cargo carrier is a wagon and the cargo is placed in the wagon, it can be determined that the type of the mobile vendor is a wagon booth, for example.
And S14, inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area.
In this embodiment, after all street view images in the first subset are processed, the mobile vendor identification model is trained through the first subset and the second subset, and the identification result of the mobile vendor in the preset area is obtained through prediction, so as to provide data support for the mobile vendor prediction process in other areas. It will be appreciated that during the training process, the partial data extracted from the first subset may be saved to the training set together with the partial data extracted from the second subset, while the remaining data extracted from the first subset is saved to the validation set. And the partial data of the first subset in the training set can be used as positive training samples, and the number of the positive training samples is equal to that of the samples of the second subset in the training set. For example, if a total of 10 samples are extracted from the first subset, wherein 9 samples are saved in the training set and the remaining 1 sample is used as the validation set, the number of samples extracted from the second subset is also 9. After the training set and the verification set are obtained, the training set is firstly input into a mobile vendor identification model, then features are sequentially extracted, pooled and fused from street view images in the mobile vendor identification model, and a first prediction category and a first boundary frame of the mobile vendor on the street view images in the training set are obtained through prediction and serve as identification results in the current training process. After one round of training is completed, inputting the verification set into a mobile vendor identification model, and sequentially extracting features, pooling the features and fusing the features of street view images in the verification set in the mobile vendor identification model to obtain a second prediction category and a second boundary frame of the mobile vendor in the street view images in the verification set as an identification result in the current verification process. And then determining a confidence score and an intersection region corresponding to the second bounding box, wherein the intersection region refers to an overlapping region of the predicted bounding box and the labeled bounding box. And when the confidence score is smaller than the confidence threshold or the intersection area is smaller than the intersection area threshold, extracting part of data from the first subset again to be respectively used as a training set and a verification set, extracting part of data from the second subset to be stored in the training set, then retraining the mobile vendor identification model through the acquired training set again, and simultaneously verifying the trained mobile vendor identification model by adopting the acquired verification set again. And circulating the steps until the confidence score is greater than or equal to the confidence threshold and the intersection area is greater than or equal to the intersection area threshold, inputting the street view image into the trained mobile vendor identification model, and obtaining the identification result of the mobile vendor in the preset area after sequentially extracting the features, pooling the features and fusing the features of the street view image.
For example, a mobile vendor identification model is constructed by using the YOLOv4 algorithm, so that the mobile vendor identification model can effectively detect the target object in image data of different lighting conditions, observation visual angles, blocked targets, complex backgrounds or scenes. The method for constructing the mobile vendor identification model by adopting the YOLOv4 algorithm comprises the following four steps: deep network construction, model training, model inspection and model application.
Specifically, as shown in fig. 2, the mobile vendor identification model constructed based on YOLOv4 algorithm includes three parts: the Backbones are used for extracting features from the training images, and CSPDarknet53 is adopted as the Backbones; the Neck consists of SPP and PANET, wherein the SPP is used for feature pooling, and the PANET is used for feature fusion; heads, used to predict and output the vendor's prediction categories and bounding boxes. After the model construction is completed, the street view images with the types marked are combined with some street view images without any vendor to form a model training set, and the training set is input into the recognition model. In this embodiment, in order to improve the recognition efficiency, a set of standard bounding boxes with group widths and heights of (17, 24), (29, 34), (38, 51), (50, 76), (65, 54), (69, 108), (91, 81), (107, 34), (165, 181) can be determined by using a K-means algorithm according to the distribution mean of the labeled bounding boxes for detecting vendors in the street view image. In addition, in order to eliminate the interference of complex colors and backgrounds in different images, the full-color images in the training set can be converted by various combinations such as turning, scaling, color gamut change and the like, and then the full-color images are imported into background bones for recognition model training. Specifically, the K-means algorithm is a clustering analysis algorithm for iterative solution, and the processing procedure is to decompose preset data into K groups, randomly calculate K objects as initial clustering centers, then calculate the distance between each object and each clustering center, and assign each object to the clustering center closest to the distance. The cluster center and the assigned objects represent a cluster. During training, the confidence threshold and the set IoU threshold formed by dividing the overlapped part of the two regions by the two regions are key parameters for the vendor to identify the model, and the embodiment controls the output result and influences the performance of the model through the confidence threshold and the IoU threshold. The higher the threshold, the lower the error tolerance of the detection result, and generally, neither the confidence threshold nor the IoU threshold is set too low, which may result in too many false detection results. To test the sensitivity of these two thresholds and calibrate the parameters, this example performed 8 sets of control experiments with different parameters. The confidence threshold values are set between 0.4 and 0.8 at intervals of 0.1, and the IoU threshold values are set between 0.5 and 0.8 at intervals of 0.1. And finally, finding out the parameter value with the optimal performance by comparing the model performance of the result obtained by combining different values of the two thresholds.
In the process of training the recognition model, when the detection result obtained by the recognition model meets two conditions, the recognition result of the recognition model is correct. Wherein the two conditions include a confidence score greater than or equal to the confidence threshold and an intersection region greater than or equal to the IoU threshold. In the decision process, the model performance uses F1Value (F)1Score, harmonic mean of Precision and recall) and mean Precision, mAP (mean Average Precision). F1The value is calculated by the formula (1), where P is the precision, and is calculated by the ratio of the number of correct vendor Tags (TP) to the total number of frames detected (TP + FP); r is the recall ratio, calculated from the ratio of the number of detected correct vendor Tags (TP) to the number of real frames (TP + FN). The mAP is the area of the region enclosed by the precision-recall curve and the x-axis, and can be calculated by the cumulative integral formula shown in formula (2):
wherein, P is TP/(TP + FP), R is TP/(TP + FN), and K is the number of tag classes.
In the process of verifying the training effect of the recognition model, the embodiment adopts a ten-rule cross mode for verification. For example, sample data is randomly divided into 10 shares, 9 of which are used for training in turn, and the remaining 1 is used for model verification. The model test uses the average of 10 experiments, and the model finally used for testing is the one that performs best in 10 experiments.
In this embodiment, the model performance of the results obtained by comparing the confidence threshold and the IoU threshold with different values is shown in table 1.
TABLE 1
As can be seen from Table 1, when the confidence threshold monotonically decreases, the recall rate monotonically increases; the general trend of accuracy is descending, and the local trend is ascending or descending. While the recall rate and accuracy monotonically decrease as the threshold IoU monotonically increases. The first three of each index are labeled and a suitable value of two thresholds equal to 0.5 can be found, and therefore, the set of threshold parameters is used to calibrate the detection model. The detection model obtained by using the group of parameters for training is averaged to obtain F1The value is 0.77, the mAP is 66.75%, and the model performance meets the preset requirement, so that the method can be applied to detection of the whole street view image data set.
After the training and verification of the model are completed, the street view image is input into the trained recognition model, the recognition model can output a prediction frame for the street view image containing the mobile vendor, the point location corresponding to the street view image records the number of the appeared prediction frames, and the number reflects the number of the mobile vendor detected at the point location. For example, when street view image recognition is performed on a city a, the number of street view images obtained is 76108, and 3907 floating booths are recognized from the 76108 street view images; for example, when identifying street view images of city B, the number of acquired street view images is 3339062, and 26119 mobile booths are identified from 3339062 street view images. The mobile booths are marked on the street view image through a prediction frame, so that the spatial positions of the mobile booths in the current area can be obtained through analysis.
And S15, acquiring the association factor of the mobile vendor.
In this embodiment, the mobile vendors tend to prefer to gather and traffic-intensive areas, such as intersections, markets, and the vicinity of public transportation sites, in order to obtain more operational benefits when participating in the business. But is constrained by law enforcement and the proximity effect so that vendors may actively move away from areas where their aggregation is undesirable. Based on this, as shown in table 2, an index system covering 12 factors was established for characterizing the spatial distribution of vendors:
TABLE 2
Wherein "+" in table 2 indicates a correlation factor having a positive impact on a mobile vendor's stall position and "-" indicates a correlation factor having a negative impact on a mobile vendor's stall position.
Specifically, the population distribution in the index 1 is mainly used for representing refined population space gathering information inside a city, and the grid data of the population space distribution of the city can be formed by inversion of information such as mobile phone signaling and smart phone users, and the existing population space distribution data products can also be used, the pixel size is 100 × 100 meters, the number of population in the area is reflected by each piece of previous pixel information, and the projection coordinate is WGS 84.
The intersection of index 2 can be extracted from the acquired urban road network data. Specifically, firstly, a network data set constructing tool in GIS software is used for establishing a road network for the vector data of the city, and then the intersection in the network data set is extracted, namely intersection data. And finally, calculating the straight line distance between each position in the city range and the intersections by using the Euclidean distance in the GIS software, and outputting to form corresponding raster data.
The metro opening, pedestrian overpass, market, school, cultural leisure facility, large transportation hub, management and law enforcement, and CBDs (Central Business district) of indexes 3 to 10 can be calculated based on POIs (Points of Interest). Specifically, city infrastructure points, including subway stations, pedestrian bridges, markets, schools, cultural leisure facilities, large transportation hubs (e.g., airports, train stations, public transportation hubs), city management law enforcement, CBDs, are first obtained from API (application program interface) services of a high-end map or a Baidu map. And then, calculating Euclidean distances from each position to the POIs in the city range by using the linear distances in the GIS software, and outputting and forming corresponding raster data.
The road congestion degree in the index 11 comprises dynamic road traffic data, and is mainly used for measuring the traffic speed conditions of urban roads in different time periods, so as to evaluate the congestion degree, and further analyze the quantitative relation between vendor operation and road congestion. Dynamic traffic speed data can be acquired by using a high-grade map or Baidu map API (application program interface), and by designing crawler software, the dynamic traffic speed data are respectively input into road network sections of a research area, so that the traffic time of each section of road can be measured, the traffic speed is further obtained, and the congestion degree is further graded by combining the speed limit grade of the road.
The house rent in index 12 is used to analyze the spatial correlation between vendor location preferences and high, medium and low-level residential areas and people of different consumption levels in the city. Rent data may be obtained from related departments, or from larger property portals or industry research institutions. In the embodiment, the web crawler tool is adopted to obtain house rental listing data from large house rental information portal websites, each piece of obtained data records house area, rental price and geographic coordinates, repeated release or error data is cleaned, the data is converted into vector points containing area unit rental information, and a house rental spatial distribution map of a research area is obtained by using a spatial interpolation tool.
And S16, inputting the spatial position and the correlation factor of the mobile vendors in the preset area into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendors in the area to be predicted.
In this embodiment, when predicting the spatial distribution of mobile vendors in the region to be predicted, the spatial position of the mobile vendors in the preset region may be input into the spatial distribution prediction model to obtain an ROC curve of the model performance, and then when the ROC curve meets the preset requirement, the correlation factor is input into the spatial distribution prediction model to predict the spatial distribution state of the mobile vendors in the region to be predicted. The ROC curve is a coordinate-based analysis tool, and is used for the following two aspects: firstly, selecting an optimal signal detection model, and discarding a second best model; second, an optimal threshold is set in the same model. It is understood that the embodiment can adopt the MaxEnt ecological niche model as a spatial distribution prediction model to predict the probability of the mobile vendor activities in the urban area to be predicted. Specifically, the identified mobile vendor is taken as a presence sample, 75% of the samples are randomly selected as training samples of the MaxEnt ecological niche model, and the rest 25% of the samples are taken as verification samples. The grid images of the correlation factors listed in table 2 are converted into ASCII format and input as environment layers into the MaxEnt ecological model compiled from the R language. The MaxEnt ecological niche model is based on a training sample and a verification sample calibration model, then a spatial distribution map of the occurrence probability of the mobile vendor is predicted, and after the spatial distribution map is superposed and analyzed with the existing streets and open spaces of the area to be predicted, the spatial distribution state of the mobile vendor in the area to be predicted can be obtained.
In summary, the identification result of the mobile vendors in the area where the street image is located is obtained through the identification model according to the known road network data and the street image, and then the spatial distribution state of the mobile vendors in the area to be predicted is obtained through the spatial distribution prediction model by combining the correlation factor and the recognition result of the vendors in the known area, so that the spatial distribution state of the mobile vendors in the designated area can be quickly obtained without manually collecting a large amount of data, and meanwhile, the influence of errors caused by manually collecting data can be avoided.
The embodiment of the invention provides a method for predicting the spatial distribution of mobile vendors, which comprises the following steps:
the first acquisition module is used for acquiring road network data of a preset area;
the second acquisition module is used for acquiring a plurality of street view images corresponding to preset coordinate points in the road network data;
the decomposition module is used for respectively storing the street view images into a first subset and a second subset, and performing type marking on the street view images in the first subset by a mobile vendor;
the identification module is used for inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area;
a third obtaining module, configured to obtain an association factor of a mobile vendor;
and the prediction module is used for inputting the spatial positions of the mobile vendors in the preset area and the correlation factors into a spatial distribution prediction model to predict and obtain the spatial distribution state of the mobile vendors in the area to be predicted.
The content of the embodiment of the method of the invention is all applicable to the embodiment of the system, the function of the embodiment of the system is the same as the embodiment of the method, and the beneficial effect achieved by the embodiment of the system is the same as the beneficial effect achieved by the method.
The embodiment of the invention provides a device for predicting the spatial distribution of mobile vendors, which comprises:
at least one memory for storing a program;
at least one processor configured to load the program to perform the method for predicting the spatial distribution of a mobile vendor as shown in fig. 1.
The content of the method embodiment of the present invention is applicable to the apparatus embodiment, the functions specifically implemented by the apparatus embodiment are the same as those of the method embodiment, and the beneficial effects achieved by the apparatus embodiment are also the same as those achieved by the method.
An embodiment of the present invention provides a storage medium, in which a computer-executable program is stored, and the computer-executable program is executed by a processor to implement the method for predicting the spatial distribution of a mobile vendor as shown in fig. 1.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention. Furthermore, the embodiments of the present invention and the features of the embodiments may be combined with each other without conflict.
Claims (10)
1. A method for predicting the spatial distribution of mobile vendors is characterized by comprising the following steps:
acquiring road network data of a preset area;
acquiring a plurality of street view images corresponding to preset coordinate points in the road network data;
storing the street view images into a first subset and a second subset respectively, and performing type marking on the street view images in the first subset by a mobile vendor;
inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area;
acquiring an association factor of a mobile vendor;
and inputting the spatial position of the mobile vendors in the preset area and the correlation factor into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendors in the area to be predicted.
2. The method of claim 1, wherein the inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the predetermined area comprises:
extracting partial data of the first subset and partial data of the second subset and saving the partial data and the residual data of the first subset to a training set, and saving the residual data of the first subset to a verification set;
inputting the training set into a mobile vendor identification model, and sequentially extracting features, pooling the features and fusing the features of street view images in the training set to obtain a first prediction category and a first boundary frame of a mobile vendor on the street view images in the training set;
inputting the verification set into a mobile vendor identification model, and sequentially extracting features, pooling the features and fusing the features of street view images in the verification set to obtain a second prediction category and a second boundary frame of mobile vendors on the street view images in the verification set;
determining a confidence score and an intersection region corresponding to the second bounding box;
and when the confidence score is greater than or equal to a confidence threshold and the intersection area is greater than or equal to an intersection area threshold, inputting the street view image into a mobile vendor identification model, and obtaining an identification result of the mobile vendor in the preset area after sequentially extracting features, pooling features and fusing features of the street view image.
3. The method of claim 2, wherein the method further comprises the following steps before the step of inputting the training set into a mobile vendor identification model is performed:
turning street view images in the training set to obtain turned images;
zooming the turned image to obtain a zoomed image;
and performing color gamut transformation on the scaled image to obtain a target image for training.
4. The method of claim 1, wherein the step of inputting the spatial locations of mobile vendors in the predetermined area and the correlation factors into a spatial distribution prediction model to predict the spatial distribution of mobile vendors in the area to be predicted comprises:
inputting the spatial position of the mobile vendor in the preset area into a spatial distribution prediction model to obtain an ROC curve of model performance;
and when the ROC curve meets the preset requirement, inputting the correlation factor into a spatial distribution prediction model, and predicting to obtain the spatial distribution state of the mobile vendor in the region to be predicted.
5. The method of claim 1, wherein the obtaining of the street view images corresponding to the preset coordinate points in the road network data comprises:
setting preset coordinate points at preset intervals in the road network data;
and acquiring a plurality of street view images corresponding to the preset coordinate points.
6. The method of claim 5, wherein the obtaining the street view images corresponding to the preset coordinate points comprises:
and acquiring a plurality of street view images at different shooting angles on the preset coordinate point.
7. The method of claim 1, wherein the type labeling of the street vendor in the first subset comprises:
extracting cargo carriers within the street view image within a first subset;
labeling the type of the mobile vendor according to the cargo carrier.
8. A method for predicting the spatial distribution of mobile vendors, comprising:
the first acquisition module is used for acquiring road network data of a preset area;
the second acquisition module is used for acquiring a plurality of street view images corresponding to preset coordinate points in the road network data;
the decomposition module is used for respectively storing the street view images into a first subset and a second subset, and performing type marking on the street view images in the first subset by a mobile vendor;
the identification module is used for inputting the first subset and the second subset into a mobile vendor identification model to obtain an identification result of mobile vendors in the preset area;
a third obtaining module, configured to obtain an association factor of a mobile vendor;
and the prediction module is used for inputting the spatial positions of the mobile vendors in the preset area and the correlation factors into a spatial distribution prediction model to predict and obtain the spatial distribution state of the mobile vendors in the area to be predicted.
9. An apparatus for predicting the spatial distribution of a mobile vendor, comprising:
at least one memory for storing a program;
at least one processor configured to load the program to perform the method for predicting the spatial distribution of a mobile vendor of any one of claims 1-7.
10. A storage medium having stored thereon a computer-executable program which, when executed by a processor, is configured to implement a method for predicting the spatial distribution of a mobile vendor as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111353459.6A CN114022772B (en) | 2021-11-16 | 2021-11-16 | Method, system, device and storage medium for predicting space distribution of mobile vendor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111353459.6A CN114022772B (en) | 2021-11-16 | 2021-11-16 | Method, system, device and storage medium for predicting space distribution of mobile vendor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114022772A true CN114022772A (en) | 2022-02-08 |
| CN114022772B CN114022772B (en) | 2024-05-03 |
Family
ID=80064601
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111353459.6A Active CN114022772B (en) | 2021-11-16 | 2021-11-16 | Method, system, device and storage medium for predicting space distribution of mobile vendor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114022772B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458333A (en) * | 2019-07-18 | 2019-11-15 | 华南农业大学 | A kind of population spatial distribution prediction technique and system based on POIs data |
| CN110992645A (en) * | 2019-12-06 | 2020-04-10 | 江西洪都航空工业集团有限责任公司 | Mobile vendor detection and alarm system in dynamic scene |
| CN111444818A (en) * | 2020-03-24 | 2020-07-24 | 哈尔滨工程大学 | CNN-based market stall violation stall detection method |
| CN112651293A (en) * | 2020-10-30 | 2021-04-13 | 华设设计集团股份有限公司 | Video detection method for road illegal stall setting event |
| CN112990517A (en) * | 2019-12-12 | 2021-06-18 | 中移雄安信息通信科技有限公司 | Crowd distribution prediction method and system |
-
2021
- 2021-11-16 CN CN202111353459.6A patent/CN114022772B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458333A (en) * | 2019-07-18 | 2019-11-15 | 华南农业大学 | A kind of population spatial distribution prediction technique and system based on POIs data |
| CN110992645A (en) * | 2019-12-06 | 2020-04-10 | 江西洪都航空工业集团有限责任公司 | Mobile vendor detection and alarm system in dynamic scene |
| CN112990517A (en) * | 2019-12-12 | 2021-06-18 | 中移雄安信息通信科技有限公司 | Crowd distribution prediction method and system |
| CN111444818A (en) * | 2020-03-24 | 2020-07-24 | 哈尔滨工程大学 | CNN-based market stall violation stall detection method |
| CN112651293A (en) * | 2020-10-30 | 2021-04-13 | 华设设计集团股份有限公司 | Video detection method for road illegal stall setting event |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114022772B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9547866B2 (en) | Methods and apparatus to estimate demography based on aerial images | |
| US20130226667A1 (en) | Methods and apparatus to analyze markets based on aerial images | |
| JP6781711B2 (en) | Methods and systems for automatically recognizing parking zones | |
| CN109493119B (en) | POI data-based urban business center identification method and system | |
| Laumer et al. | Geocoding of trees from street addresses and street-level images | |
| US20160063516A1 (en) | Methods and apparatus to estimate commercial characteristics based on geospatial data | |
| US20220357176A1 (en) | Methods and data processing systems for predicting road attributes | |
| CN112016605A (en) | Target detection method based on corner alignment and boundary matching of bounding box | |
| CN112668375B (en) | Tourist distribution analysis system and method in scenic spot | |
| CN114708521A (en) | City functional area identification method and system based on street view image target relation perception network | |
| Gargoum et al. | A voxel-based method for automated detection and mapping of light poles on rural highways using lidar data | |
| CN111242223B (en) | Street space quality evaluation method based on streetscape image multi-feature fusion | |
| CN111414878A (en) | Method and device for social attribute analysis and image processing of land parcel | |
| JP2022003548A (en) | Recognition system and program for the same | |
| Dipankar et al. | Pavement crack detection based on a deep learning approach and visualisation by using GIS | |
| CN116413740A (en) | Laser radar point cloud ground detection method and device | |
| CN116628531A (en) | Crowd-sourced map road object element clustering method, system and storage medium | |
| CN114022772B (en) | Method, system, device and storage medium for predicting space distribution of mobile vendor | |
| Panizzi et al. | Private or Public Parking Type Classifier on the Driver’s Smartphone | |
| Meedeniya et al. | Land‐Use Classification with Integrated Data | |
| CN112015937B (en) | Picture geographic positioning method and system | |
| Hammoudi et al. | Towards a model of car parking assistance system using camera networks: Slot analysis and communication management | |
| CN109686077A (en) | Traveling state of vehicle monitoring method and device | |
| US10783644B2 (en) | Methods and systems for algorithmically analyzing urban form | |
| CN111488771B (en) | OCR hooking method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |