CN106021499B - Construction land classification method and device based on volunteer geographic information - Google Patents

Abstract

The invention relates to a construction land classification method and device based on volunteer geographic information. The method includes: determining the area where the construction land to be classified is located; acquiring volunteer geographic information in the area; divide the area according to the main road to obtain multiple first-level plots; divide the area according to all roads to obtain multiple second-level plots; determine each plot in the area The level of a point of interest; according to the level of each point of interest, determine the level of the plot to which the point of interest belongs, and assign the attributes of the point of interest to the plot at the level of the plot; The construction land type of the property assigned to the plot is obtained, and the classification result is obtained. The invention can reduce cost, improve automation, shorten cycle and improve real-time performance, and can also improve the accuracy of construction land classification.

Description

Construction land classification method and device based on volunteer geographic information

technical field

The invention relates to the technical field of land classification, in particular to a construction land classification method and device based on volunteer geographic information.

Background technique

In recent years, my country is in the process of rapid urbanization. Although the growth of construction land has kept pace with economic growth, the rapid growth of urban land, rural settlements and industrial and mining land across the country, and the "two-way expansion" of urban and rural construction land shows that China's construction land has not embarked on the road of saving land. The low utilization efficiency of construction land will cause more other land types to be converted into construction land under the same demand, which will affect the amount of cultivated land and reduce ecological land. The existing spatial layout of various types of construction land is one of the foundations for improving utilization efficiency and strengthening the monitoring and control of planning implementation. In order to realize the spatial layout of various types of construction land, it is necessary to classify construction land.

According to the classification standard of land use status (GB/T 21010-2007), construction land can be divided into commercial land, industrial and mining storage land, residential land, public management and public service land, transportation land and other construction land. At present, the combination of traditional remote sensing technology and geographic information system can be used. This method is difficult to subdivide the construction land. Therefore, the method of on-site investigation is generally used. This method has the following disadvantages: (1) The degree of automation is low. , consumes a lot of human resources, capital and time, and the cost is too high; (2) the cycle is long, after a cycle ends, the use of some construction land has been changed, that is, the real-time performance is poor.

Professor Goodchild first proposed Volunteered Geographic Information (VGI) in 2007. He believes that in the era of Web 2.0, spatial data is not only produced by experts and consumed by the public, but everyone is a sensor for acquiring data, not only Data users are also data producers and disseminators. In 2010, Li Deren et al. defined the generalized volunteer geographic information as the concept, mode, method and technology related to the spontaneous geographic information in the narrow sense. As a result, crowdsourcing data and crowdsourcing platforms are increasingly driving scientific and technological development into the field of vision of scientific researchers. In 2012, Sarah Elwood et al. explored the application potential of volunteer geographic information in various fields. In the same year, Rodrigues F. et al. used points of interest to consider and analyze land use from the perspective of urban planning. In 2013, J.J.Arsanjani et al. only used the Open Street Map (OSM) to map the current state of land use in Vienna, and used official data for testing, with an accuracy of more than 65%. In the same year, Long Ying et al. used the type and quantity of Sina Weibo check-in point data to divide their research area by function. Although this kind of research has many improvements compared to the traditional construction land classification method, there are still some problems: (1) The amount of volunteer geographic information data used is small, and in cities with high construction land complexity, it will lead to accuracy (2) It does not take into account that the actual ground objects represented by different types of point data have different occupied areas, and their importance in the process of type determination is also different, which will lead to low accuracy.

SUMMARY OF THE INVENTION

In view of the above defects, the present invention provides a construction land classification method and device based on volunteer geographic information, which can reduce costs, improve automation, shorten cycle and improve real-time performance, and can also improve the accuracy of construction land classification.

In the first aspect, the method for classifying construction land based on volunteer geographic information provided by the present invention includes:

According to the remote sensing image data, determine the area where the construction land to be classified is located;

Obtain the volunteer geographic information of the area where the volunteer is located, where the geographic information of the volunteer includes all road data and all point-of-interest data in the area;

According to the road data, determine all the roads in the area and the main roads in the all roads;

Divide the area according to the main road to obtain a plurality of first-level plots;

Divide the area according to all the roads to obtain a plurality of second-level plots;

According to the point of interest data, determine the level of each point of interest in the area;

According to the level of each point of interest, determine the level of the plot to which the point of interest belongs, and assign the attribute of the point of interest to the plot of the plot level;

According to the land classification standard, determine the construction land types of the plots with attributes assigned by each point of interest, and obtain the classification results.

Optionally, the level of the point of interest includes a first level, a second level and a third level;

The POI of the first level is a POI including multiple buildings and the number of internal roads is more than a preset number;

The POI of the second level is a POI including a plurality of buildings and the number of internal roads is less than the preset number;

The third level POI is a POI occupying an entire building.

Optionally, according to the level of each point of interest, determining the level of the plot to which the point of interest belongs, including:

The level of the point of interest is the first level, and it is determined that the level of the plot to which the point of interest belongs is the first level; or

The level of the point of interest is the second level or the third level, and it is determined that the level of the plot to which the point of interest belongs is the second level.

Optionally, the level of the POI further includes a fourth level, and the POI of the fourth level is a single-storey building that shares a building with other POIs or that occupies a smaller area than a preset area; the method further includes :

If it is determined that the POI of the fourth level exists in the indoor map data of the Atlas mall, the POI of the fourth level is eliminated.

Optionally, before determining the level of each point of interest in the area according to the point of interest data, the method further includes:

According to the data of all points of interest, other types of land except for construction land in the said area are removed.

Optionally, before determining the level of each point of interest in the area according to the point of interest data, the method further includes:

According to the character feature of each attribute in the interest point data, a corresponding regular expression is constructed, and matching attribute information is set for each interest point according to the regular expression.

Optionally, the volunteer geographic information further includes check-in data of the social platform and/or feature pictures with geographic locations; the method further includes:

The classification result is verified according to the check-in data of the social platform and/or the feature picture with the geographic location, and the classification accuracy of the construction land is determined.

Optionally, before the area is divided according to the main road, and a plurality of first-level plots are obtained, the method further includes: extending each road by a preset length.

In the second aspect, the construction land classification device based on volunteer geographic information provided by the present invention includes:

In the method and device for classifying construction land based on volunteer geographic information provided by the present invention, since two divisions are performed when the area is divided into blocks, a plurality of first-level areas with larger footprints are obtained by dividing the main road into blocks. , and use all road blocks to obtain multiple second-level plots with smaller footprints. Then, according to the level of the point of interest, determine the level of the appropriate assigned plot to avoid dividing one point of interest or not dividing multiple points of interest, thereby improving the accuracy of construction land classification. The classification method provided by the present invention does not require manual field investigation, so it has the advantages of low cost, high degree of automation, short period and high real-time performance. At the same time, the classification of plots and points of interest are used, and then different levels of interest points are used to classify plots of different levels, that is, for different points of interest, the levels of the plots to which they belong are determined to be different, taking into account the interests of different levels The area occupied by the actual ground objects represented by the points is different, which improves the accuracy of the classification of construction land.

Description of drawings

Characteristic information and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way, in which:

1 shows a schematic flowchart of a method for classifying construction land based on volunteer geographic information according to an embodiment of the present invention;

Fig. 2 shows a structural block diagram of an apparatus for classifying construction land based on volunteer geographic information according to an embodiment of the present invention.

Detailed ways

In order to understand the above objects, features and advantages of the present invention more clearly, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments may be combined with each other in the case of no conflict.

Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific details disclosed below. Example limitations.

As shown in Figure 1, the present invention provides a construction land classification method based on volunteer geographic information, the method comprising:

S101. Determine the area where the construction land to be classified is located according to the remote sensing image data;

Understandably, the remote sensing image data is parsed to determine the area to be classified.

S102, obtaining the volunteer geographic information of the region where the volunteer is located, where the geographic information of the volunteer includes all road data and all point-of-interest data in the region;

Among them, volunteer geographic information can be obtained on the Internet, which is open and free. The so-called point of interest data refers to data related to POI (Point of Interest). Road data can be obtained in OSM (Open StreetMap). The obtained road data has a large amount, high accuracy and fast update speed.

S103, according to the road data, determine all the roads in the area and the main roads in the all roads;

S104, dividing the area according to the main road to obtain a plurality of first-level plots;

It is understandable that by using the main road to divide the area, the land area of the first level obtained is generally larger.

S105. Divide the area according to all the roads to obtain a plurality of second-level plots;

It is understandable that by dividing the area by all roads, the obtained second-level plot area is generally smaller.

S106, according to the point of interest data, determine the level of each point of interest in the area;

S107, according to the level of each point of interest, determine the level of the plot to which the point of interest belongs, and assign the attribute of the point of interest to the plot of the plot level;

The plot level here is the above-mentioned first level and second level. Due to the different levels of points of interest, the assigned parcels are of different levels.

S108 , according to the land classification standard, determine the construction land types of the plots with attributes assigned by each point of interest, and obtain a classification result.

Among them, the land classification standard can be the current land use classification standard. According to the classification standard, construction land can be divided into commercial land, industrial and mining storage land, residential land, public management and public service land, transportation land or other construction land. Land type.

For example, a POI includes multiple buildings and the number of internal roads is more than the preset number of POIs. Such POIs generally cover a large area or even occupy an entire block, such as colleges and universities, airports, etc. It is defined as the first-level POI. Since such POIs include many roads, the attribute of the POI should be assigned to the first-level plot to which the POI belongs when assigning it, so as to avoid the POI. Points are segmented, resulting in misclassification of construction land.

For another example, a point of interest is a point of interest that includes multiple buildings but the number of internal roads is less than the preset number. Although such a point of interest includes multiple buildings, each building is concentrated due to fewer roads, such as a residential area. Industrial parks, etc., are defined as points of interest at the second level. Such points of interest are less roads, and a main road can have multiple such points of interest. Therefore, when assigning, the attributes of the point of interest should be assigned to the second-level plot to which the point of interest belongs to avoid unintended consequences. Segmenting multiple points of interest leads to misclassification of construction land.

For another example, a point of interest occupies the point of interest of a whole building, such as a large company, a star-rated hotel, a library, etc. The same point of interest has no roads, so the attribute of the point of interest should be assigned when assigning The second level parcel to which this POI belongs.

In the method for classifying construction land provided by the present invention, since two kinds of subdivisions are carried out when subdividing the area, a plurality of first-level land plots with large area are obtained by subdividing the main road, and all road subdivisions are used Blocks get multiple second-level plots with smaller footprints. Then, according to the level of the point of interest, determine the level of the appropriate assigned plot to avoid dividing one point of interest or not dividing multiple points of interest, thereby improving the accuracy of construction land classification. The classification method provided by the present invention does not require manual field investigation, so it has the advantages of low cost, high degree of automation, short period and high real-time performance. At the same time, the classification of plots and points of interest are used, and then different levels of interest points are used to classify plots of different levels, that is, for different points of interest, the levels of the plots to which they belong are determined to be different, taking into account the interests of different levels The area occupied by the actual ground objects represented by the points is different, which improves the accuracy of the classification of construction land.

Of course, there is also a level of POI, which shares a building with other POIs or a single-storey building with an area smaller than a preset area, such as restaurants, shopping stores, etc., is defined as the fourth level Points of Interest. Among them, single-storey buildings with an area less than the preset area, such as bungalows with an area of less than 30 square meters. The processing methods for such points of interest can include:

If it is determined that the POI of the fourth level exists in the indoor map data of the Atlas mall, the POI of the fourth level is eliminated.

Among them, the Atlas shopping mall indoor map provides developers with real-time updated shopping mall maps, including data of more than 400 shopping malls and stores in Shanghai, Beijing, Guangzhou, Chengdu (and other places), and matches the group purchase, movie tickets, and comment data of each store.

Here, for the points of interest existing in the indoor map data of the Atlas mall, if the plots to which they belong are assigned and classified, it will cause classification overflow, so they are deleted here to further improve the classification accuracy. Of course, if it does not exist in the indoor map data of the Atlas mall, the POI will be regarded as a single feature. At this time, the plot level can be determined according to its size, and then assigned to complete the plot to which the POI belongs. Classification.

In the specific implementation, since the points of interest come from the electronic map, they have the advantages of detailed classification and wide range of types. In order to further ensure the classification accuracy, before the classification, you can also: according to all the data of the points of interest Other types of land use other than land use are removed, so as to avoid other types of land use other than construction land from interfering with the classification.

During the specific implementation, it is found through data inspection that there is a situation where the attribute information of a small number of interest points does not correspond. At this time, the following method can be used to set matching attribute information for each interest point: according to each type of interest point data. Character features of attributes, construct a corresponding regular expression, and set matching attribute information for each interest point according to the regular expression.

Here, by constructing a regular expression, and then setting the matching attribute information for each interest point according to the regular expression, that is, the corresponding text and numerical information, the accuracy of subsequent assignments is improved, thereby improving the accuracy of classification.

During specific implementation, since there is a problem that roads are not connected due to operating errors of OSM volunteers, each road may be extended by a preset length before dividing the area according to the main road. Of course, it is also possible to trim the part that is still not connected to other roads after the extension.

During specific implementation, step S103 may further include setting buffers of different widths according to different levels of roads, that is, widening different types of roads with different widths to improve the accuracy of determining the POI level.

During specific implementation, the geographic information of volunteers also includes check-in data of social platforms and/or pictures of features with geographic locations; the method may further include:

The classification result is verified according to the check-in data of the social platform and/or the feature picture with the geographic location, and the classification accuracy of the construction land is determined.

Among them, the feature pictures with geographic location can be obtained from the Panoramio platform. Check-in data of social platforms, such as check-in data of Sina Weibo.

Here, the accuracy of the classification is determined by means of verification, so as to know whether the above-mentioned classification method is reliable. Through verification, it can be seen that the classification accuracy is above 90%.

Based on the same inventive concept, the present invention also provides a construction land classification device based on volunteer geographic information. As shown in FIG. 2 , the device 200 includes:

The first determination module 201 is used for determining the area where the construction land to be classified is located according to the remote sensing image data;

an acquisition module 202, configured to acquire volunteer geographic information of the region where the volunteer is located, where the geographic information of the volunteer includes all road data and all point-of-interest data in the region;

A second determination module 203, configured to determine all roads in the area and the main roads in all the roads according to the road data;

a first dividing module 204, configured to divide the area according to the main road to obtain a plurality of first-level plots;

A second dividing module 205, configured to divide the area according to all the roads to obtain a plurality of second-level plots;

A third determining module 206, configured to determine the level of each point of interest in the area according to the point of interest data;

The assignment module 207 is used for determining the level of the plot to which the point of interest belongs according to the level of each point of interest, and assigning the attribute of the point of interest to the plot of the plot level;

The fourth determining module 208 is configured to determine, according to the land classification standard, the construction land type of the plot of which attributes are assigned by each point of interest, and obtain a classification result.

Optionally, the level of the point of interest includes a first level, a second level and a third level;

The POI of the first level is a POI including multiple buildings and the number of internal roads is more than a preset number;

The POI of the second level is a POI including a plurality of buildings and the number of internal roads is less than the preset number;

The third level POI is a POI occupying an entire building.

In the present invention, the terms "first", "second", "third", and "fourth" are used for descriptive purposes only, and should not be construed as indicating or implying relative importance. The term "plurality" refers to two or more, unless expressly limited otherwise.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, various modifications and variations can be made by those skilled in the art without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope of the appended claims within the limits of the requirements.

Claims (10)

1. a construction land classification method based on volunteer geographic information, is characterized in that, comprises: According to the remote sensing image data, determine the area where the construction land to be classified is located; Obtain the volunteer geographic information of the area where the volunteer is located, where the geographic information of the volunteer includes all road data and all point-of-interest data in the area; According to the road data, determine all the roads in the area and the main roads in the all roads; Divide the area according to the main road to obtain a plurality of first-level plots; Divide the area according to all the roads to obtain a plurality of second-level plots; According to the point of interest data, determine the level of each point of interest in the area; According to the level of each point of interest, determine the level of the plot to which the point of interest belongs, and assign the attribute of the point of interest to the plot of the plot level; According to the land classification standard, determine the construction land types of the plots with attributes assigned by each point of interest, and obtain the classification results; The land parcel level to which it belongs is the first level or the second level. 2. The method according to claim 1, wherein the level of the point of interest comprises a first level, a second level and a third level; The POI of the first level is a POI including multiple buildings and the number of internal roads is more than a preset number; The POI of the second level is a POI including a plurality of buildings and the number of internal roads is less than the preset number; The third level POI is a POI occupying an entire building. 3. The method according to claim 2, wherein the determining the level of the plot to which the point of interest belongs according to the level of each point of interest, comprising: The level of the point of interest is the first level, and it is determined that the level of the plot to which the point of interest belongs is the first level; or The level of the point of interest is the second level or the third level, and it is determined that the level of the plot to which the point of interest belongs is the second level. 4 . The method according to claim 2 , wherein the level of the POI further comprises a fourth level, and the POI of the fourth level is sharing a building with other POIs or having an area smaller than a preset level. 5 . area of a single-story building; the method further comprising: If it is determined that the POI of the fourth level exists in the indoor map data of the Atlas mall, the POI of the fourth level is eliminated. 5 . The method according to claim 1 , wherein, before determining the level of each point of interest in the region according to the point of interest data, the method further comprises: 6 . According to the data of all points of interest, other types of land except for construction land in the said area are removed. 6 . The method according to claim 1 , wherein, before determining the level of each interest point in the area according to the interest point data, the method further comprises: 6 . According to the character feature of each attribute in the interest point data, a corresponding regular expression is constructed, and matching attribute information is set for each interest point according to the regular expression. 7. The method according to claim 1, wherein the volunteer geographic information further comprises check-in data of a social platform and/or a feature picture with geographic location; the method further comprises: The classification result is verified according to the check-in data of the social platform and/or the feature picture with the geographic location, and the classification accuracy of the construction land is determined. 8 . The method according to claim 1 , wherein before dividing the area according to the main road to obtain a plurality of first-level plots, the method further comprises: dividing each Roads are extended by a preset length. 9. A construction land classification device based on volunteer geographic information, characterized in that, comprising: The first determination module is used for determining the area where the construction land to be classified is located according to the remote sensing image data; an acquisition module, configured to acquire volunteer geographic information of the region where the volunteer is located, where the geographic information of the volunteer includes all road data and all point-of-interest data in the region; a second determining module, configured to determine, according to the road data, all roads in the area and the main roads in all the roads; a first dividing module, configured to divide the area according to the main road to obtain a plurality of first-level plots; A second division module, configured to divide the area according to all the roads to obtain a plurality of second-level plots; a third determining module, configured to determine the level of each point of interest in the area according to the point of interest data; an assignment module, configured to determine the level of the plot to which the point of interest belongs according to the level of each point of interest, and assign the attribute of the point of interest to the plot of the plot level; The fourth determination module is used to determine the construction land type of the plots with attributes assigned by each point of interest according to the land classification standard, and obtain the classification result; The land parcel level to which it belongs is the first level or the second level. 10. The apparatus according to claim 9, wherein the levels of the points of interest include a first level, a second level and a third level; The first-level POIs include multiple buildings and the number of internal roads is more than a preset number of POIs; The POI of the second level is a POI including a plurality of buildings and the number of internal roads is less than the preset number; The third level POIs are POIs occupying an entire building.

Images