CN117893383A - Urban functional area identification method, system, terminal equipment and medium - Google Patents
Urban functional area identification method, system, terminal equipment and medium Download PDFInfo
- Publication number
- CN117893383A CN117893383A CN202410289290.XA CN202410289290A CN117893383A CN 117893383 A CN117893383 A CN 117893383A CN 202410289290 A CN202410289290 A CN 202410289290A CN 117893383 A CN117893383 A CN 117893383A
- Authority
- CN
- China
- Prior art keywords
- space
- characterization
- poi
- functional
- paragraph
- 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 83
- 239000013598 vector Substances 0.000 claims abstract description 120
- 238000012512 characterization method Methods 0.000 claims abstract description 103
- 230000006870 function Effects 0.000 claims abstract description 95
- 230000002708 enhancing effect Effects 0.000 claims abstract description 7
- 238000003064 k means clustering Methods 0.000 claims abstract description 7
- 238000011160 research Methods 0.000 claims description 38
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 13
- 238000013135 deep learning Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 238000011161 development Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 2
- 238000007418 data mining Methods 0.000 abstract description 2
- 238000009826 distribution Methods 0.000 description 23
- 230000008569 process Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005070 sampling Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 4
- 230000019771 cognition Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000003915 cell function Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000013361 beverage Nutrition 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application is applicable to the technical field of space-time big data mining, and provides a city functional area identification method, a system, terminal equipment and a medium, wherein the method comprises the following steps: acquiring POI data, and dividing the POI data into space units; constructing a Delaunay triangle network, and constructing a space unit paragraph after long-side constraint; extracting effective generalized words in each space unit vector characterization paragraph aiming at each space unit vector characterization paragraph, and learning the characterization vector of each space unit vector paragraph based on a designed learning function for enhancing the similarity between the characterization vector of the effective generalized words and the space unit paragraph characterization vector to obtain the function characterization of each space unit; k-means clustering is carried out on the space units based on the similarity of the functional characterization, the average density ratio and the type ratio of each cluster are calculated respectively, and the functional areas of the researched city are identified. The method and the device can improve accuracy of urban functional area identification.
Description
Technical Field
The application belongs to the technical field of space-time big data mining, and particularly relates to a city functional area identification method, a system, terminal equipment and a medium.
Background
The current urban area function cognition method can be mainly divided into cognition based on human activity statistical characteristics and cognition based on deep learning characterization.
The method based on the human activity statistical characteristics is based on the characteristics of the density and the like of the point of interest data (POI, point of Interest, etc.) and the like, and combines the urban human activity statistical characteristics of the traffic flow, the track density and the like to estimate the urban area functions. However, because the POI text labels are complex in semantics and difficult to directly calculate through a mathematical method, the calculation of human activity features is too subjective, so that the region features recognized by the method based on the human activity statistical features are shallow, and the method is difficult to directly apply to fine recognition of urban functional areas. With the rapid development of machine learning and deep learning technologies, the deep learning characterization-based method improves Word vector deep learning methods such as Word2Vec (a natural language processing technology based on a neural network, which is used for representing words into vector form, so that mathematical operations and analysis can be performed on the words on a computer), gloVe (Global Vectors for Word Representation, an algorithm for learning Word vector representation) and the like to characterize regional functions. According to the method, the POI tag is regarded as a paragraph word, the POI data in a certain area are orderly sampled by designing a specific sampling mode to generate an area paragraph, the POI context co-occurrence feature is further learned based on the area paragraph, a POI tag characterization vector is obtained, and finally the area feature is expressed by adopting methods such as average or weighted average of the POI tag characterization vector. However, the distribution mode of POI points also has a certain influence on the regional function, for example, densely distributed commercial consumption POI may represent a commercial square type, and uniformly distributed commercial consumption POI may represent a general resident living shopping supermarket region, so that the regional function vector obtained by adopting feature average or weighted average currently cannot reflect the spatial distribution difference of POI facilities.
In summary, to recognize the urban area function structure, it is necessary to first characterize the urban area function, and the actual urban area function is affected by the semantic and spatial distribution modes of the urban POI tags, so that the accuracy of identifying the urban area based on the POI data is low.
Disclosure of Invention
The application provides a method, a system, terminal equipment and a medium for identifying urban functional areas, which can solve the problem of improving the accuracy of urban functional area identification.
In a first aspect, the present application provides a method for identifying a city functional area, including:
acquiring POI data of a research city, and dividing the POI data into space units of the research city according to the position information of the POI data; the research city comprises a plurality of space units, and the POI data comprises position information of different types of places in the research city;
respectively aiming at each space unit, constructing a Delaunay triangle network according to POI data of the space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network; the Delaunay triangular net corresponds to the space units one by one, long-side constraint is used for deleting abnormal long sides in the Delaunay triangular net, and the space unit vector representation section is used as a subsequent deep learning training sample library;
Extracting effective generalized words in the space unit vector characterization paragraphs respectively for each space unit vector characterization paragraph, constructing graph paragraph vectors of the space unit vector characterization paragraph, and learning the characterization vector of each space unit vector paragraph based on a designed learning function for enhancing the similarity between the characterization vector of the effective generalized words and the space unit paragraph characterization vector to obtain the function characterization of each space unit; the effective generalized word is used for representing frequent POI rooted subgraphs in the research city, and the functional representation is used for describing the functions of the space units;
calculating the similarity of the functional representation among the space units, clustering the space units based on the similarity of the functional representation to obtain a plurality of cluster clusters, and respectively calculating the average density ratio and the type ratio of each cluster; functional characterization similarity is used to describe the similarity between two spatial units;
and identifying the functional areas of the researched city according to the average density ratio and the type ratio.
Optionally, performing long-side constraint on the Delaunay triangle network includes:
obtain the firstAverage side length of Delaunay triangularly>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>，/>Represents the total number of Delaunay triangulations, < > >Representation ofEdge->Euclidean distance of>，/>Indicate->Total number of edges in the individual Delaunay triangulation;
by calculation formula
Obtain the firstSide length standard deviation of Delaunay triangularly>；
By calculation formula
Obtain the firstAbnormal long side threshold of Delaunay triangulation>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>The control factor is represented by a value representing,；
respectively for the firstEach edge in the Delaunay triangulation network is deleted if the edge length is greater than the abnormal long edge threshold value, so as to obtain the +.>Delaunay triangulation after long side constraint.
Optionally, constructing a plurality of space unit vector characterization paragraphs includes:
for each spatial unit, the following steps are performed:
delaunay triangulation after constraint for each long sideBy calculation formulaObtain->Is->Order neighboring POI Point set->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representation->And->At->The nearest figure-order distance on +.>Indicates that satisfy for->Nodes of nearest graph order distance, +.>Delaunay triangle net after long side constraint>Middle POI Point->Corresponding nodes;
constructionCorresponding->Step rooted map->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>；
By calculation formulaObtain->Corresponding generalized word->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicating splice->Representation->Corresponding POI Point- >POI type of (c);
respectively aiming at any two generalized words in each space unitAnd->If->，And->Then determine +.>Obtaining a plurality of space unit vector characterization paragraphs +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein, when->When (I)>，/>Representing a preset maximum order of the rooted sub-graph.
Alternatively, the expression of the learning function is as follows:
wherein,space unit vector characterization paragraph representing the correspondence of all space units,/->，Representing a similarity function, +.>，/>Representing space element vector characterization paragraph->Token vector of valid generalized words of +.>Representation->Corresponding word vector, ">Representation->Corresponding paragraph vectors, which are used to characterize the space cell function,/for example>Representing generalized valid words, < >>Representing natural constant->An exponential function of the base.
Optionally, the expression of the functional characterization similarity is as follows:
wherein,representing space element->And space unit->Similarity of functional characterization between->Representing space element->Functional characterization of->Representing space element->Functional characterization of->When->Space unit->And space unit->The functions are the same when->Space unit->And space unit->The functions are reversed.
Alternatively, the average density ratio is expressed as The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster->Middle->Density and clustering of type POI points>Middle->Ratio of average density of type POI spots, +.>Representing cluster->Middle->Number of types of POI spots->Representing cluster->Total area of all space units in +.>Representing>Number of types of POI spots->Representing the total area of the study city;
the expression of the type proportion isThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster->Middle->Average Density ratio of type POI Point>Ratio of average density ratio of all types, +.>Representing the total number of POI types in the study city.
Optionally, the identification result of the functional area is a weak functional area or a mixed functional area, the weak functional area is an area with insufficient development in the city, and the mixed functional area is an area without special functions in the city;
identifying the functional area of the researched city according to the average density ratio and the type ratio, wherein the method comprises the following steps:
separately for each clusterThe following operations are performed:
if it isThe maximum value of the average density ratio of the POI types of all spatial units in the cluster is less than 0.7, the cluster is +.>All space units in the network are identified as weak functional areas;
if it isThe maximum value of the average density ratio of the POI types of all spatial units in the cluster is more than 0.7 and less than 1.3, the cluster is clustered +. >All space units in the network are identified as mixed functional areas;
if it isThe maximum value of the average density ratio of the POI types of all the space units in (a) is more than 1.3, the pairs are in the order of the type ratio from big to small>The POI types in the middle are ordered from big to small, and the POI types in the first 30% are judged according to the accumulated addition>The functional type of all spatial units in the system.
In a second aspect, the present application provides a city function area identification system, comprising:
the data acquisition module is used for acquiring POI data of the research city and dividing the POI data into space units of the research city according to the position information of the POI data; the research city comprises a plurality of space units, and the POI data comprises position information of different types of places in the research city;
the long-side constraint module is used for respectively constructing a Delaunay triangle network according to POI data of each space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network; the Delaunay triangular net corresponds to the space units one by one, long-side constraint is used for deleting abnormal long sides in the Delaunay triangular net, the space unit vector representation paragraph is used for representing, and the space unit vector representation paragraph is used as a subsequent deep learning training sample library;
The function characterization module is used for respectively aiming at each space unit vector characterization paragraph, extracting effective generalized words in the space unit vector characterization paragraph, constructing a graph paragraph vector of the space unit vector characterization paragraph, and iterating the effective generalized words and the graph paragraph vector based on a designed learning function for enhancing the similarity between the effective generalized words and the graph paragraph vector until the preset iteration termination condition is met, so as to obtain the function characterization of each space unit; the effective generalized word is used for representing frequent POI rooted subgraphs in the research city, and the functional representation is used for describing the functions of the space units;
the recognition parameter calculation module is used for calculating the functional representation similarity among the space units, K-means clustering is carried out on the space units based on the functional representation similarity to obtain a plurality of cluster clusters, and the average density ratio and the type ratio of each cluster are calculated respectively; functional characterization similarity is used to describe the similarity between two spatial units;
and the identification module is used for identifying the functional area of the researched city according to the average density ratio and the type ratio.
In a third aspect, the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the above-mentioned urban function area identification method when executing the computer program.
In a fourth aspect, the present application provides a computer readable storage medium storing a computer program which when executed by a processor implements the above-described urban function area identification method.
The scheme of the application has the following beneficial effects:
according to the urban function area recognition method, after long-side constraint is carried out on the Delaunay triangulation network, the local scale POI space distribution structure can be considered, meanwhile, interference of POIs with far distances on a distribution mode is avoided, and accuracy of urban function area recognition is improved; by constructing a space unit vector characterization paragraph and extracting effective generalized words in the space unit vector characterization paragraph, the function characterization learning considering the multi-scale space distribution mode can be realized, so that the depth characterization learning model can distinguish the space unit function differences with different POI distribution modes, and the accuracy of urban function region identification is improved.
Other advantages of the present application will be described in detail in the detailed description section that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following description will briefly introduce the drawings that are needed in the embodiments or the description of the prior art, it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a city function area identification method according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a city function area identifying system according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As used in this specification and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
In addition, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
Aiming at the problem of low accuracy of urban function area identification based on POI data at present, the application provides an urban function area identification method, an urban function area identification system, terminal equipment and medium, and the method can consider the local scale POI space distribution structure after long-side constraint is carried out on a Delaunay triangulation network, meanwhile, the interference of POIs with far distances on a distribution mode is avoided, and the accuracy of urban function area identification is improved; by constructing a space unit vector characterization paragraph and extracting effective generalized words in the space unit vector characterization paragraph, the function characterization learning considering the multi-scale space distribution mode can be realized, so that the depth characterization learning model can distinguish the space unit function differences with different POI distribution modes, and the accuracy of urban function region identification is improved.
The city function area identifying method provided by the application is exemplified below.
Specifically, as shown in fig. 1, the urban functional area identification method includes the following steps:
and 11, acquiring POI data of the research city, and dividing the POI data into space units of the research city according to the position information of the POI data.
The research city includes a plurality of spatial units. The POI data includes location information for different types of sites in the research city.
Illustratively, in embodiments of the present application, POI data may be obtained from a public dataset (e.g., a hadamard map data open platform), the types of POI data including: traffic facilities, recreational entertainment, corporate enterprises, business houses, tourist attractions, life services, scientific and educational cultures, shopping consumption, financial institutions and food and beverage delicates.
In order to ensure the validity of POI data, after the POI data are collected, the collected POI data are cleaned, and the specific cleaning process is as follows:
first, POI data whose spatial position is outside the study city is deleted. POI data of unknown semantic type (e.g., type "other") is then deleted. Further, for a plurality of POI data having the same name and a spatial linear distance of not more than 50 meters (m), in the embodiment of the present application, it is considered to be essentially the same city facility, one POI data in which the closest to the road network is reserved. Then, the POI semantics are reclassified according to the POI tag type.
In addition, spatial unit division is required for the research city. Specifically, since the road network may have parallel road segments, a buffer of 30m is first generated from the road network data. And then fusing the road network buffer areas with the space superposition relation, and extracting the road center line from the fused road network buffer areas. Then, the study city is divided into spatial units according to the road center line, and the divided spatial units with the minimum POI number or minimum POI area are combined to other units with side adjacent relation. For example, the spatial units of the research city may be represented as ，/>Indicate->And a space unit.
And finally, performing space matching of the POI data and the space unit. The method comprises the following steps: searching for the space unit that each POI falls into based on its spatial position in turn, so that each space unit can be represented asWherein->For POI within a spatial cell, +.>Is the total number of POIs within a spatial unit.
Step 12, respectively constructing a Delaunay triangle network according to POI data of each space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network.
The Delaunay triangular networks are in one-to-one correspondence with the space units, long-side constraint is used for deleting abnormal long sides in the Delaunay triangular networks, and the space unit vector representation paragraph is used for representing the space unit vector representation paragraph as a subsequent deep learning training sample library.
The current POI paragraph generation method treats spatially discrete POI and basic spatial units as words and text paragraphs composed of words, respectively. Unlike real text paragraph sequences which follow a certain grammar rule and expand in one-dimensional length, POIs in urban space units are distributed in two-dimensional urban space according to constraints of road networks, buildings and the like, and the current POI text paragraph generation method breaks up real distribution characteristics of POIs in the real urban space.
In view of this, the present application proposes a graph constraint sampling method based on spatial distribution of POI, which fully considers the spatial distribution mode of POI points, and transforms the POI in a spatial unit into a graph paragraph model, and the method will be described in detail later.
And 13, respectively aiming at each space unit vector characterization paragraph, extracting effective generalized words in the space unit vector characterization paragraph, constructing a graph paragraph vector of the space unit vector characterization paragraph, and learning the characterization vector of each space unit vector paragraph based on a designed learning function for enhancing the similarity between the characterization vector of the effective generalized words and the space unit paragraph characterization vector to obtain the functional characterization of each space unit.
The valid generalized word is used for representing frequent POI rooted subgraphs in the research city, and the functional representation is used for describing the functions of the space units.
Specifically, the expression of the learning function is as follows:
wherein,space unit vector characterization paragraph representing the correspondence of all space units,/->，Representing a similarity function, +.>，/>Representing space element vector characterization paragraph->Token vector of valid generalized words of +.>Representation->Corresponding word vector, ">Representation- >Corresponding paragraph vectors, which are used to characterize the space cell function,/for example>Representing generalized valid words, < >>Representing natural constant->An exponential function of the base.
The formula measures the overall association closeness between all generalized word vectors in the graph paragraph and the graph paragraph vectors thereof, and the larger the formula shows that the graph paragraph vectors are more consistent with the generalized word vectors, the smaller the formula shows that the current vector characterization is difficult to embody the consistency between the graph paragraph and the generalized word, and the iterative parameter learning needs to be continued.
Because the calculated amount and the absolute value are too large, in order to accelerate the convergence speed of the model, a negative sample contrast learning strategy is introduced. An adjustable superparameter is set to represent the number of negative samples, which are obtained from a specified noise distribution for the internal positive samples, where the parameter is a free parameter and is a word frequency function. The essence of negative sampling is that the probability of positive samples is maximized while the probability of negative samples is minimized, and the calculation formula of the negative sample strategy is as follows:
；
；
wherein,representing a negative sample value,/->The larger the parameter representing the probability of smooth noise sampling, the more easily the generalized valid word with higher word frequency is selected as noise, typically taking a value of 0.75.
The negative sampling technology enables each sample to update only a small part of weights in training, so that the calculated amount in the gradient descent process is reduced, and finally the aim of improving the training speed is achieved. Finally, training generalized word vectors and paragraph vectors in an iterative mode through random gradient descent and back propagation, and finally outputting the paragraph vectors as space unit function vectors.
When the training batch is greater than or equal toThe average cosine similarity between corresponding space unit characterizations obtained by (preset maximum training batch) or two times of training is greater than +.>And when (the preset average cosine similarity), the preset iteration termination condition is met, the training process is stopped, and the functional representation of each space unit is obtained.
And 14, calculating the similarity of the functional representation among the space units, and carrying out K-means clustering on the space units based on the similarity of the functional representation to obtain a plurality of clusters, and respectively calculating the average density ratio and the type ratio of each cluster.
Functional characterization similarity is used to describe the similarity between two spatial units.
The expression of the above functional characterization similarity is as follows:
wherein,indicating emptyInterlude(s)>And space unit->Similarity of functional characterization between->Representing space element->Functional characterization of->Representing space element->Functional characterization of->When->Space unit->And space unit->The functions are the same when->Space unit->And space unit->The functions are reversed.
The expression of the average density ratio isThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster- >Middle->Density and clustering of type POI points>Middle->Ratio of average density of type POI spots, +.>Representing cluster->Middle->Number of types of POI spots->Representing cluster->Total area of all space units in +.>Representing>Number of types of POI spots->Representing the total area of the city under investigation.
The expression of the type proportion isThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster->Middle->Average Density ratio of type POI Point>Ratio of average density ratio of all types, +.>Representing the total number of POI types in the study city.
And 15, identifying the functional area of the researched city according to the average density ratio and the type ratio.
The recognition result of the functional area is a weak functional area or a mixed functional area, the weak functional area is an area with insufficient development in the city, and the mixed functional area is an area without special functions in the city.
Specifically, for each clusterThe following operations are performed:
if it isThe maximum value of the average density ratio of the POI types of all spatial units in the cluster is less than 0.7, the cluster is +.>All space units in the network are identified as weak functional areas;
if it isThe maximum value of the average density ratio of the POI types of all spatial units in the cluster is more than 0.7 and less than 1.3, the cluster is clustered +. >All space units in the network are identified as mixed functional areas;
if it isThe maximum value of the average density ratio of the POI types of all the space units in (a) is more than 1.3, the pairs are in the order of the type ratio from big to small>The POI types in the middle are ordered from big to small, and the POI types in the first 30% are judged according to the accumulated addition>The functional type of all spatial units in the system.
It should be noted that, according to the recognition result of the functional area of the research city, reasonable guidance can be provided for planning the functional area of the research city.
An exemplary procedure for long-side constraining the Delaunay triangle network in step 12 is described below. The method comprises the following steps:
step I, through a calculation formula
Obtain the firstAverage side length of Delaunay triangularly>。/>，/>Representing the total number of Delaunay triangulation networks.
Wherein,representing edge->Euclidean distance of>，/>Indicate->Total number of edges in the Delaunay triangle network.
Step II, through a calculation formula
Obtain the firstSide length standard deviation of Delaunay triangularly>。
Step III, through a calculation formula
Obtain the firstAbnormal long side threshold of Delaunay triangulation>。
Wherein,representing control factors->。
Step IV, respectively for the firstEach side in the Delaunay triangle network is deleted if the side length of the side is larger than the threshold value of the abnormal long side, and the +. >Delaunay triangulation after long side constraint.
The process of constructing a plurality of space unit vector characterization paragraphs in step 12 is exemplarily described below. The method comprises the following steps:
for each spatial unit, the following steps are performed:
step a, respectively aiming at Delaunay triangulation network after each long side constraintBy calculation formulaObtain->Is->Order neighboring POI Point set->。
Wherein,representation->And->At->The nearest figure-order distance on +.>Representation ofDelaunay triangle net after long side constraint +.>Middle POI Point->Corresponding node->Delaunay triangle net after long side constraint>Middle POI Point->A corresponding node.
Step b, constructingCorresponding->Step rooted map->。
Wherein,。
step c, through a calculation formulaObtain->Corresponding generalized words。
Wherein,indicating splice->Representation->Corresponding POI Point->POI type of (c).
Step d, aiming at any two generalized words in each space unit respectivelyAnd->If->，/>And->Then determine +.>Obtaining a plurality of space unit vector characterization paragraphs +.>。
Wherein whenWhen (I)>，/>Representing a preset maximum order of the rooted sub-graph.
In another embodiment of the present application, an island is selected as the study city, and the data used are POI data and road data within the island. In this example, a total of 70039 POI data points.
The following describes an exemplary urban functional area recognition system provided in the present application, specifically as follows:
as shown in fig. 2, the urban functional area recognition system 200 includes:
the data acquisition module 201 is configured to acquire POI data of a research city, and divide the POI data into space units of the research city according to position information of the POI data; the research city includes a plurality of space units;
the long-side constraint module 202 is configured to construct a Delaunay triangle network according to POI data of each space unit, and construct a plurality of space unit vector characterization paragraphs after long-side constraint is performed on the Delaunay triangle network; the Delaunay triangular net corresponds to the space units one by one, long-side constraint is used for deleting abnormal long sides in the Delaunay triangular net, the space unit vector representation paragraph is used for representing, and the space unit vector representation paragraph is used as a subsequent deep learning training sample library;
the function characterization module 203 is configured to extract valid generalized words in the space unit vector characterization paragraphs for each space unit vector characterization paragraph, construct a graph paragraph vector of the space unit vector characterization paragraph, and iterate the valid generalized words and the graph paragraph vector based on a learning function designed to enhance similarity between the characterization vector of the valid generalized words and the graph paragraph vector until a preset iteration termination condition is satisfied, so as to obtain a function characterization of each space unit; the effective generalized word is used for representing frequent POI rooted subgraphs in the research city, and the functional representation is used for describing the functions of the space units;
The recognition parameter calculation module 204 is configured to calculate a functional representation similarity between each spatial unit, perform K-means clustering on the plurality of spatial units based on the functional representation similarity to obtain a plurality of cluster clusters, and calculate an average density ratio and a type ratio of each cluster respectively; functional characterization similarity is used to describe the similarity between two spatial units;
and the identification module 205 is used for identifying the functional area of the researched city according to the average density ratio and the type ratio.
It should be noted that, because the content of information interaction and execution process between the modules/units is based on the same concept as the method embodiment of the present application, specific functions and technical effects thereof may be referred to in the method embodiment section, and details thereof are not repeated herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
As shown in fig. 3, an embodiment of the present application provides a terminal device, as shown in fig. 3, a terminal device D10 of the embodiment includes: at least one processor D100 (only one processor is shown in fig. 3), a memory D101 and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, the processor D100 implementing the steps in any of the various method embodiments described above when executing the computer program D102.
Specifically, when the processor D100 executes the computer program D102, POI data of the research city is collected, and the POI data is divided into space units of the research city according to the position information of the POI data; respectively aiming at each space unit, constructing a Delaunay triangle network according to POI data of the space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network; extracting effective generalized words from each space unit vector representation paragraph, constructing a graph paragraph vector of the space unit vector representation paragraph, and iterating the effective generalized words and the graph paragraph vector based on a designed learning function for enhancing the similarity between the effective generalized words and the graph paragraph vector until a preset iteration termination condition is met, so as to obtain the functional representation of each space unit; calculating the similarity of the functional representation among the space units, and carrying out K-means clustering on the space units based on the similarity of the functional representation to obtain a plurality of clustering clusters, and respectively calculating the average density ratio and the type ratio of each clustering cluster; and identifying the functional areas of the researched city according to the average density ratio and the type ratio. After long-side constraint is carried out on the Delaunay triangulation network, the spatial distribution structure of the local scale POI can be considered, meanwhile, the interference of the POI with a longer distance on a distribution mode is avoided, and the accuracy of urban functional area identification is improved; by constructing a space unit vector characterization paragraph and extracting effective generalized words in the space unit vector characterization paragraph, the function characterization learning considering the multi-scale space distribution mode can be realized, so that the depth characterization learning model can distinguish the space unit function differences with different POI distribution modes, and the accuracy of urban function region identification is improved.
The processor D100 may be a central processing unit (CPU, central Processing Unit), the processor D100 may also be other general purpose processors, digital signal processors (DSP, digital Signal Processor), application specific integrated circuits (ASIC, application Specific Integrated Circuit), off-the-shelf programmable gate arrays (FPGA, field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory D101 may in some embodiments be an internal storage unit of the terminal device D10, for example a hard disk or a memory of the terminal device D10. The memory D101 may also be an external storage device of the terminal device D10 in other embodiments, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the terminal device D10. Further, the memory D101 may also include both an internal storage unit and an external storage device of the terminal device D10. The memory D101 is used for storing an operating system, an application program, a boot loader (BootLoader), data, other programs, etc., such as program codes of the computer program. The memory D101 may also be used to temporarily store data that has been output or is to be output.
Embodiments of the present application also provide a computer readable storage medium storing a computer program which, when executed by a processor, implements steps that may implement the various method embodiments described above.
The present embodiments provide a computer program product which, when run on a terminal device, causes the terminal device to perform steps that enable the respective method embodiments described above to be implemented.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to the metropolitan area identification system/terminal equipment, a recording medium, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/network device and method may be implemented in other manners. For example, the apparatus/network device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
The urban function area identification method provided by the application has the following advantages:
(1) The prior art generally trains POI word vectors independently, then uses methods such as summation or weighted summation as space unit function vectors, and essentially still belongs to a linear feature statistical mode, thereby ignoring the nonlinear comprehensive relation of functional semantics. According to the method, the space unit function vector and the POI word vector are trained simultaneously, and the biased aggregation from the POI facility function to the space unit function is avoided through the vector consistency constraint of the space unit function vector and the POI word vector.
(2) The function of the spatial distribution mode on the semantic recognition of the functional area cannot be considered in the prior art, the space unit diagram paragraph generation method based on the spatial neighborhood relation diagram of the POI facilities is designed, a single POI facility and a local sub-diagram of a multi-order POI are simultaneously used as generalized words, function characterization learning considering the multi-scale spatial distribution mode is realized, and the depth characterization learning model can distinguish the function difference of the spatial units with different POI distribution modes.
While the foregoing is directed to the preferred embodiments of the present application, it should be noted that modifications and adaptations to those embodiments may occur to one skilled in the art and that such modifications and adaptations are intended to be comprehended within the scope of the present application without departing from the principles set forth herein.
Claims (10)
1. A method for identifying urban functional areas, comprising:
acquiring POI data of a research city, and dividing the POI data into space units of the research city according to the position information of the POI data; the research city comprises a plurality of space units, and the POI data comprises position information of different types of places in the research city;
respectively constructing a Delaunay triangle network according to POI data of each space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network; the Delaunay triangulation network is in one-to-one correspondence with the space units, the long-side constraint is used for deleting abnormal long sides in the Delaunay triangulation network, and the space unit vector characterization section is used as a subsequent deep learning training sample library;
extracting effective generalized words in each space unit vector characterization paragraph, constructing a graph paragraph vector of the space unit vector characterization paragraph, and learning the characterization vector of each space unit vector paragraph based on a designed learning function for enhancing the similarity between the characterization vector of the effective generalized words and the space unit paragraph characterization vector to obtain the functional characterization of each space unit; the valid generalized words are used for representing frequent POI root graphs in the research city, and the functional representation is used for describing the functions of the space units;
Calculating the similarity of functional characterization among the space units, and carrying out K-means clustering on the space units based on the similarity of functional characterization to obtain a plurality of clusters, and respectively calculating the average density ratio and the type ratio of each cluster; the functional token similarity is used to describe the similarity between two spatial units;
and identifying the functional area of the research city according to the average density ratio and the type ratio.
2. The urban function area recognition method according to claim 1, wherein the long-side constraint is performed on the Delaunay triangulation network, comprising:
by calculation formula
Obtain the firstAverage side length of Delaunay triangularly>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>，/>Represents the total number of Delaunay triangulations, < >>Representing edge->Euclidean distance of>，/>Indicate->Total number of edges in the individual Delaunay triangulation;
by calculation formula
Obtain the firstSide length standard deviation of Delaunay triangularly>；
By calculation formula
Obtain the firstAbnormal long side threshold of Delaunay triangulation>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing control factors->；
Respectively for the firstEach edge in the Delaunay triangulation network is deleted if the edge length is greater than the abnormal long edge threshold value, so as to obtain the +. >Delaunay triangulation after long side constraint.
3. The urban functional area recognition method according to claim 2, wherein said constructing a plurality of space unit vector characterization paragraphs comprises:
the following steps are performed for each of the spatial units separately:
delaunay triangulation after constraint for each long sideBy calculation formulaObtain->Is->Order neighboring POI Point set->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representation->And->At->The nearest figure-order distance on +.>Indicates that satisfy for->Nodes of nearest graph order distance, +.>Delaunay triangle net ++representing the long side constraint>Middle POI Point->Corresponding nodes;
constructionCorresponding->Step rooted map->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>；
By calculation formulaObtain->Corresponding generalized word->The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicating splice->Representation->Corresponding POI Point->POI type of (c);
respectively aiming at any two generalized words in each space unitAnd->If->，And->Then determine +.>Obtaining a plurality of space unit vector characterization paragraphs +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein, when->When (I)>，/>Representing a preset maximum order of the rooted sub-graph.
4. A city function area identifying method of claim 3, wherein the learning function is expressed as follows:
Wherein,space unit vector characterization paragraph representing the correspondence of all space units,/->，/>Representing a similarity function, +.>，/>Representing space element vector characterization paragraph->Token vector of valid generalized words of +.>Representation->Corresponding word vector, ">Representation->Corresponding paragraph vector for characterizing spatial unit functions, +.>Representing generalized valid words, < >>Representing natural constant->An exponential function of the base.
5. The urban functional area recognition method according to claim 4, wherein the expression of the functional characterization similarity is as follows:
wherein,representing space element->And space unit->Similarity of functional characterization between->Representing space element->Functional characterization of->Representing space element->Functional characterization of->When->Space unit->And space unit->The functions are the same when->Space unit->And space unit->The functions are reversed.
6. The urban function area recognition method according to claim 5, wherein the expression of the average density ratio isThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster->Middle->Density and clustering of type POI points>Middle->Ratio of average density of type POI spots, +. >Representing cluster->Middle->Number of types of POI spots->Representing cluster->Total area of all space units in +.>Representing>Number of types of POI spots->Representing the total area of the study city;
the expression of the type proportion isThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing cluster->Middle->Average Density ratio of type POI Point>Ratio of average density ratio of all types, +.>Representing the total number of POI types in the study city.
7. The urban function area recognition method according to claim 6, wherein the recognition result of the function area is a weak function area or a mixed function area, the weak function area is an area of insufficient development in a city, and the mixed function area is an area of no distinctive function in the city;
and identifying the functional area of the research city according to the average density ratio and the type ratio, wherein the method comprises the following steps:
separately for each clusterThe following operations are performed:
if it isMaximum average density ratio of POI types for all spatial units in a systemThe value is less than 0.7, the cluster is +.>All space units in the network are identified as weak functional areas;
if it isThe maximum value of the average density ratio of the POI types of all spatial units in the cluster is more than 0.7 and less than 1.3, the cluster is clustered +. >All space units in the network are identified as mixed functional areas;
if it isThe maximum value of the average density ratio of the POI types of all the space units in (a) is more than 1.3, the pairs are in the order of the type ratio from big to small>The POI types in the middle are ordered from big to small, and the POI types in the first 30% are judged according to the accumulated additionThe functional type of all spatial units in the system.
8. A city function area identification system, comprising:
the data acquisition module is used for acquiring POI data of a research city and dividing the POI data into space units of the research city according to the position information of the POI data; the research city comprises a plurality of space units, and the POI data comprises position information of different types of places in the research city;
the long-side constraint module is used for respectively constructing a Delaunay triangle network according to the POI data of each space unit, and constructing a plurality of space unit vector characterization paragraphs after long-side constraint is carried out on the Delaunay triangle network; the Delaunay triangulation network is in one-to-one correspondence with the space units, the long-side constraint is used for deleting abnormal long sides in the Delaunay triangulation network, and the space unit vector characterization section is used as a subsequent deep learning training sample library;
The function characterization module is used for respectively aiming at each space unit vector characterization paragraph, extracting effective generalized words in the space unit vector characterization paragraph, constructing a graph paragraph vector of the space unit vector characterization paragraph, and iterating the effective generalized words and the graph paragraph vector based on a designed learning function for enhancing the similarity between the characterization vector of the effective generalized words and the graph paragraph vector until the preset iteration termination condition is met, so as to obtain the function characterization of each space unit; the valid generalized words are used for representing frequent POI root graphs in the research city, and the functional representation is used for describing the functions of the space units;
the identification parameter calculation module is used for calculating the functional representation similarity among the space units, K-means clustering is carried out on the space units based on the functional representation similarity to obtain a plurality of clustering clusters, and the average density ratio and the type ratio of each clustering cluster are calculated respectively; the functional token similarity is used to describe the similarity between two spatial units;
and the identification module is used for identifying the functional area of the research city according to the average density ratio and the type ratio.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the urban function area identification method according to any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the urban function area identification method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410289290.XA CN117893383B (en) | 2024-03-14 | Urban functional area identification method, system, terminal equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410289290.XA CN117893383B (en) | 2024-03-14 | Urban functional area identification method, system, terminal equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117893383A true CN117893383A (en) | 2024-04-16 |
| CN117893383B CN117893383B (en) | 2024-06-04 |
Family
ID=
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120282888A1 (en) * | 2011-05-02 | 2012-11-08 | Telefonica S.A. | Method for an automatic identification of urban dense areas from cell phones records |
| CN103927432A (en) * | 2014-02-28 | 2014-07-16 | 中南大学 | Constrained spatial clustering method for facility location programming |
| CN112699284A (en) * | 2021-01-11 | 2021-04-23 | 四川大学 | Bus stop optimization visualization method based on multi-source data |
| CN115100395A (en) * | 2022-06-27 | 2022-09-23 | 南京大学 | Urban block function classification method integrating POI pre-classification and graph neural network |
| CN116662468A (en) * | 2023-05-24 | 2023-08-29 | 武汉大学 | Urban functional area identification method and system based on geographic object space mode characteristics |
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120282888A1 (en) * | 2011-05-02 | 2012-11-08 | Telefonica S.A. | Method for an automatic identification of urban dense areas from cell phones records |
| CN103927432A (en) * | 2014-02-28 | 2014-07-16 | 中南大学 | Constrained spatial clustering method for facility location programming |
| CN112699284A (en) * | 2021-01-11 | 2021-04-23 | 四川大学 | Bus stop optimization visualization method based on multi-source data |
| CN115100395A (en) * | 2022-06-27 | 2022-09-23 | 南京大学 | Urban block function classification method integrating POI pre-classification and graph neural network |
| CN116662468A (en) * | 2023-05-24 | 2023-08-29 | 武汉大学 | Urban functional area identification method and system based on geographic object space mode characteristics |
Non-Patent Citations (4)
| Title |
|---|
| WANG, TF: "NS-DBSCAN: A Density-Based Clustering Algorithm in Network Space", 《ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION》, 25 June 2019 (2019-06-25) * |
| 刘慧敏: "顾及功能语义特征的建筑物空间分布模式识别方法", 《测绘学报》, vol. 49, no. 5, 31 May 2020 (2020-05-31), pages 622 - 631 * |
| 石岩: "利用层次约束Delaunay三角网探测空间点事件离群模式", 《武汉大学学报(信息科学版)》, vol. 41, no. 12, 26 August 2016 (2016-08-26), pages 1584 - 1592 * |
| 贾冲: "基于POI和GPS轨迹的城市功能区域识别与商业选址推荐——以兰州市为例", 《优秀硕士学位论文》, 15 January 2021 (2021-01-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106570505B (en) | Method and system for analyzing histopathological images | |
| CN110458187B (en) | Malicious code family clustering method and system | |
| Wang et al. | Identifying dominant factors for the calibration of a land-use cellular automata model using Rough Set Theory | |
| CN109871875B (en) | Building change detection method based on deep learning | |
| Van Dijk | Identifying activity-travel points from GPS-data with multiple moving windows | |
| CN112232384A (en) | Model training method, image feature extraction method, target detection method and device | |
| CN101710422B (en) | Image segmentation method based on overall manifold prototype clustering algorithm and watershed algorithm | |
| Chen et al. | Vectorization of historical maps using deep edge filtering and closed shape extraction | |
| CN109033944A (en) | A kind of all-sky aurora image classification and crucial partial structurtes localization method and system | |
| CN103136757A (en) | SAR image segmentation method based on manifold distance two-stage clustering algorithm | |
| Li et al. | Efficient interpretation of deep learning models using graph structure and cooperative game theory: Application to asd biomarker discovery | |
| Gargoum et al. | A voxel-based method for automated detection and mapping of light poles on rural highways using lidar data | |
| CN113641906A (en) | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data | |
| Xu et al. | Separation of wood and foliage for trees from ground point clouds using a novel least-cost path model | |
| CN115424053A (en) | Small sample image identification method, device and equipment and storage medium | |
| Hu et al. | A framework to detect and understand thematic places of a city using geospatial data | |
| CN113379269B (en) | Urban business function partitioning method, device and medium for multi-factor spatial clustering | |
| Li et al. | Identifying urban form typologies in Seoul using a new Gaussian mixture model-based clustering framework | |
| CN103605671A (en) | Scientific research information evolution analyzing method and device | |
| CN117893383B (en) | Urban functional area identification method, system, terminal equipment and medium | |
| CN113158084B (en) | Method, device, computer equipment and storage medium for processing movement track data | |
| CN117893383A (en) | Urban functional area identification method, system, terminal equipment and medium | |
| CN116628531A (en) | Crowd-sourced map road object element clustering method, system and storage medium | |
| CN112257970A (en) | Automatic city functional area dividing method based on interest point big data | |
| Chen et al. | Intelligent approach to Mining cultural tourism potential areas Based on YOLOv4: insights from Macau |
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 |