A method and device for identifying urban informal green spaces

Abstract

Embodiments of the present application provide a kind of urban informal green land identification method and device.The method comprises: determining the green land quadrat in target city, and obtaining the measured data of each green land quadrat, so that each green land quadrat is classified for the first time, determine the first category of each green land quadrat, according to the remote sensing image data of target city, calculate NDVI, and determine the NDVI corresponding to each green land quadrat, according to the NDVI corresponding to each informal green land quadrat, calculate the quartile of NDVI, and use it as threshold, according to the NDVI corresponding to each green land quadrat, each green land quadrat is classified for the second time, determine the second category of each green land quadrat, with the remote sensing image data corresponding to each green land quadrat as sample, with the second category corresponding to each green land quadrat as label, build data set, so that the initial city informal green land identification model is trained, obtain city informal green land identification model, and then based on the model, city informal green land is efficiently identified.

Description

Technical Field

[0001] This invention relates to the field of urban planning technology, and in particular to a method and apparatus for identifying informal green spaces in cities. Background Technology

[0002] Over the past few decades, with the rapid development of globalization and urbanization, urban ecosystems have faced unprecedented challenges. Urban green spaces, as a crucial component for improving urban living quality, maintaining biodiversity, and mitigating the urban heat island effect, have become increasingly important for effective planning and management. However, traditional urban green space planning often overlooks the value of urban spontaneous vegetation (USV). These vegetation plants, which grow spontaneously in urban gaps and fringe areas, not only enrich urban biodiversity but also play a vital role in regulating urban climate, improving air quality, and increasing the connectivity of urban green spaces. Currently, the identification of informal urban green spaces often relies on manual on-site identification, which is inefficient. Therefore, improving the efficiency of informal urban green space identification has become an urgent technical problem to be solved. Summary of the Invention

[0003] In a first aspect, embodiments of the present invention provide a method for training a model for identifying informal green spaces in cities, the method comprising:

[0004] Based on the distribution map of green space and ecological space in the target city, determine the green space quadrats in the target city and obtain the measured data of each green space quadrat;

[0005] Based on the measured data of each green space quadrat, the green space quadrat was classified for the first time to determine the first category of each green space quadrat. The first category includes: cultivated plant land and informal green space.

[0006] Based on remote sensing image data of the target city, the Normalized Difference Vegetation Index (NDVI) is calculated, and the NDVI corresponding to each green space quadrat is determined based on the location information of each green space quadrat.

[0007] Based on the NDVI corresponding to each informal green space quadrat, the quartiles of the NDVI are calculated. Using these quartiles as thresholds, each green space quadrat is further classified according to its NDVI to determine its second category. The second category includes: cultivated plant areas, low-degree free-growing plant areas, moderate-degree free-growing plant areas, and high-degree free-growing plant areas. Low-degree free-growing plant areas, moderate-degree free-growing plant areas, and high-degree free-growing plant areas are considered informal green spaces.

[0008] Using remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels, a dataset is constructed. The initial urban informal green space identification model is trained based on the dataset to obtain the urban informal green space identification model.

[0009] Among the feasible methods of the first aspect, determining green space quadrats within the target city based on a green space and ecological space distribution map of the target city includes:

[0010] Based on the distribution map of green space and ecological space in the target city, the boundary range of each green space patch in the target city is determined, and green space quadrats are randomly deployed using ArcGIS. Green space quadrats falling on buildings, roads and water surfaces are deleted, and finally the green space quadrats in the target city are determined.

[0011] In some feasible implementations of the first aspect, based on the measured data of each green space quadrat, each green space quadrat is initially classified to determine its first category, including:

[0012] For any green space plot, the proportion of cultivated plant area and the proportion of free-growing plant area are calculated based on the measured data. If the proportion of cultivated plant area is greater than a preset threshold, it is determined to be a cultivated plant plot; if the proportion of free-growing plant area is greater than a preset threshold, it is determined to be an informal green space.

[0013] In some feasible approaches to the first aspect, the Normalized Difference Vegetation Index (NDVI) is calculated based on remote sensing image data of the target city, and the NDVI corresponding to each green space quadrat is determined based on the location information of each green space quadrat, including:

[0014] Based on multiple remote sensing images of the target city taken during the vegetation growing season, the average NDVI is calculated. Then, based on the location information of each green space quadrat, the NDVI of the corresponding pixels of each green space quadrat is determined. For any green space quadrat, the average NDVI of its corresponding pixels is calculated and used as the NDVI of the green space quadrat.

[0015] In some feasible implementations of the first aspect, the quartiles of the NDVI are calculated based on the NDVI corresponding to each informal green space quadrat, and these quartiles are used as thresholds to perform a second classification of each green space quadrat based on the NDVI corresponding to each green space quadrat, determining the second category of each green space quadrat, including:

[0016] The NDVI values ​​of each informal green space quadrat were sorted in ascending order. The quartiles of the NDVI were calculated and used as thresholds. If the NDVI of a green space quadrat was lower than the first quartile, it was identified as a cultivated plant site. If the NDVI of a green space quadrat was between the first and second quartiles, it was identified as a low-degree free-growing plant site. If the NDVI of a green space quadrat was between the second and third quartiles, it was identified as a moderately free-growing plant site. If the NDVI of a green space quadrat was higher than the third quartile, it was identified as a highly free-growing plant site.

[0017] In some possible implementations of the first aspect, an initial urban informal green space identification model is trained based on a dataset to obtain an urban informal green space identification model, including:

[0018] The dataset is divided into a training set and a validation set;

[0019] The initial urban informal green space identification model is trained using the training set, and the performance of the trained urban informal green space identification model is validated using the validation set to obtain the final urban informal green space identification model.

[0020] Secondly, embodiments of the present invention provide a method for identifying informal green spaces in cities, the method comprising:

[0021] Acquire remote sensing image data of the target city;

[0022] The remote sensing image data is input into the urban informal green space identification model, which then identifies the remote sensing image data to obtain the urban informal green space identification results.

[0023] The urban informal green space identification model is obtained based on the training method for the urban informal green space identification model described above.

[0024] In some possible implementations of the second aspect, the method further includes:

[0025] Based on the identification results of informal green spaces in the city, a distribution map of informal green spaces in the city is generated.

[0026] Thirdly, embodiments of the present invention provide a training device for an urban informal green space identification model, the device comprising:

[0027] The determination module is used to determine the green space quadrats in the target city based on the green space and ecological space distribution map of the target city, and to obtain the measured data of each green space quadrat.

[0028] The classification module is used to classify each green space quadrat based on the measured data of each green space quadrat, and determine the first category of each green space quadrat. The first category includes: cultivated plant land and informal green space.

[0029] The calculation module is used to calculate the Normalized Difference Vegetation Index (NDVI) based on the remote sensing image data of the target city, and to determine the NDVI corresponding to each green space quadrat based on the location information of each green space quadrat.

[0030] The classification module is also used to calculate the quartiles of the NDVI corresponding to each informal green space quadrat, and use these quartiles as thresholds to perform a second classification of each green space quadrat based on the NDVI corresponding to each green space quadrat, determining the second category of each green space quadrat. The second category includes: cultivated plant land, low-degree freestanding plant land, moderate-degree freestanding plant land, and high-degree freestanding plant land; low-degree freestanding plant land, moderate-degree freestanding plant land, and high-degree freestanding plant land belong to informal green spaces.

[0031] The training module is used to construct a dataset using remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels. The initial urban informal green space identification model is trained based on the dataset to obtain the urban informal green space identification model.

[0032] Fourthly, embodiments of the present invention provide an urban informal green space identification device, the device comprising:

[0033] The acquisition module is used to acquire remote sensing image data of the target city;

[0034] The identification module is used to input remote sensing image data into the urban informal green space identification model, and the urban informal green space identification model identifies the remote sensing image data to obtain the urban informal green space identification results.

[0035] The urban informal green space identification model is obtained based on the training method for the urban informal green space identification model described above.

[0036] In this embodiment of the invention, green space quadrats within the target city can be determined based on a map of the green space and ecological space distribution of the target city. Measured data for each green space quadrat can be obtained, and these data are used to perform an initial classification of each green space quadrat, determining its first category, which includes: cultivated plant areas and informal green spaces. Based on remote sensing image data of the target city, the NDVI is calculated, and the NDVI corresponding to each green space quadrat is determined based on its location information. Based on the NDVI corresponding to each informal green space quadrat, the quartiles of the NDVI are calculated, and these quartiles are used as... A threshold is used to classify each green space quadrat a second time based on the NDVI corresponding to each green space quadrat, determining the second category of each green space quadrat, including: cultivated plant land, low-level free-growing plant land, medium-level free-growing plant land, and high-level free-growing plant land. The latter three categories belong to informal green spaces. Using the remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels, a dataset is constructed. This dataset is used to train the initial urban informal green space identification model, resulting in an urban informal green space identification model. This model is then used to efficiently identify urban informal green spaces.

[0037] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0038] The above and other features, advantages, and aspects of the various embodiments of the present invention will become more apparent from the accompanying drawings and the following detailed description. The drawings are provided for a better understanding of the invention and are not intended to limit the invention. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0039] Figure 1 The flowchart illustrates a training method for an urban informal green space identification model provided by an embodiment of the present invention.

[0040] Figure 2 This diagram illustrates a random forest workflow provided by an embodiment of the present invention.

[0041] Figure 3 A flowchart of a method for identifying informal urban green spaces provided by an embodiment of the present invention is shown;

[0042] Figure 4 This invention provides a map showing the distribution of informal urban green spaces according to an embodiment of the invention.

[0043] Figure 5 This diagram illustrates the structure of a training device for an urban informal green space identification model provided in an embodiment of the present invention.

[0044] Figure 6 This diagram illustrates the structure of an urban informal green space identification device according to an embodiment of the present invention.

[0045] Figure 7 A structural diagram of an exemplary electronic device capable of implementing embodiments of the present invention is shown. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Furthermore, the term "and / or" in this invention is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this invention generally indicates that the preceding and following related objects have an "or" relationship.

[0048] To address the technical problems mentioned in the background art, embodiments of the present invention provide a method and apparatus for identifying informal green spaces in cities. Specifically, based on a map of the distribution of green spaces and ecological spaces in a target city, green space quadrats within the target city are determined, and measured data for each green space quadrat are acquired. This data is used to perform an initial classification of each green space quadrat, determining its first category, which includes: cultivated plant areas and informal green spaces. Based on remote sensing image data of the target city, the NDVI is calculated, and based on the location information of each green space quadrat, the corresponding NDVI is determined. Based on the NDVI corresponding to each informal green space quadrat, the quartiles of the NDVI are calculated and used as a threshold. Based on the NDVI corresponding to each green space quadrat, a second classification is performed on each green space quadrat to determine the second category of each green space quadrat, including: cultivated plant land, low-degree native plant land, medium-degree native plant land, and high-degree native plant land. The latter three categories belong to informal green spaces. Using the remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels, a dataset is constructed. This dataset is used to train the initial urban informal green space identification model to obtain the urban informal green space identification model, and then the model is used to efficiently identify urban informal green spaces.

[0049] In this way, machine learning can be used to efficiently identify informal green spaces in cities, providing accurate data support for urban planners, environmental scientists, and policymakers, promoting the optimal allocation and management decisions of urban green space resources, and thus improving the quality of the urban environment and the quality of life of residents.

[0050] The following detailed description, with reference to the accompanying drawings and specific embodiments, illustrates a method and apparatus for identifying informal urban green spaces provided by the present invention.

[0051] Figure 1 A flowchart of a method for identifying informal urban green spaces provided by an embodiment of the present invention is shown, as follows: Figure 1 As shown, the urban informal green space identification method 100 may include the following steps:

[0052] S110. Based on the green space and ecological space distribution map of the target city, determine the green space quadrats within the target city and obtain the measured data of each green space quadrat.

[0053] In some embodiments, the boundary range of each green space patch in the target city can be determined based on the green space and ecological space distribution map of the target city and other supporting data (such as urban land use / cover remote sensing interpretation map, urban topographic DEM map), and green space quadrats can be randomly deployed using ArcGIS, green space quadrats falling on buildings, roads and water surfaces can be deleted, and some spare green space quadrats can be reserved, so as to finally determine the green space quadrats in the target city.

[0054] For example, the number of green space quadrats in the target city can be 3,500, covering various land types such as park green spaces, residential green spaces, street trees, riverbanks, urban greenways, abandoned land, and landfills. In addition, considering that urban green spaces are scattered and small in area, the surveyed native plants are mostly herbaceous, with occasional seedlings of trees and shrubs; therefore, the size of the green space quadrats here can be set to 1m*1m.

[0055] In some embodiments, researchers may conduct field measurements (i.e., plant surveys) on each green space quadrat during a specific time period (e.g., August-September 2023) and upload the measured data. The measured indicators (i.e., plant survey indicators) involved here include:

[0056] (1) Basic information on species: Record the plant species names, average height, coverage and growth status in each green space quadrat. Plant species identification is based on the Flora of China and local floras.

[0057] (2) Species composition information: Based on the plant composition, growth status, whether it is a seedling, whether it has been pruned or mechanically damaged, etc., we distinguish between "cultivated plants" and "wild plants". On-site, we judge that tall trees, shrubs and regularly distributed herbs that have been artificially planted and pruned are cultivated plants, while most herbs and scattered tree or shrub seedlings that are scattered and irregularly distributed are wild plants.

[0058] S120. Based on the measured data of each green space quadrat, the first classification of each green space quadrat is carried out to determine the first category of each green space quadrat.

[0059] The first category includes cultivated plant areas and informal green spaces. This first classification can be further detailed as follows:

[0060] For any green space plot, the proportion of cultivated plant area and the proportion of free-growing plant area are calculated based on the measured data. If the proportion of cultivated plant area is greater than a preset threshold (e.g., 50%), it is determined to be a cultivated plant plot; if the proportion of free-growing plant area is greater than a preset threshold (e.g., 50%), it is determined to be an informal green space (i.e., a free-growing plant plot).

[0061] S130: Based on the remote sensing image data of the target city, calculate the Normalized Difference Vegetation Index (NDVI), and determine the NDVI corresponding to each green space quadrat based on the location information of each green space quadrat.

[0062] The location information of each green space quadrat mentioned here is the geographical location information recorded by a handheld high-precision GPS device (accuracy <1 meter) during the actual measurement of the green space quadrat.

[0063] In some embodiments, the average NDVI can be calculated based on multiple remote sensing images of the target city taken during the vegetation growing season (e.g., Sentinel-2 satellite data). Based on this, the NDVI of the pixels corresponding to each green space quadrat can be determined according to the location information of each green space quadrat. For any green space quadrat, the average value of the NDVI of its corresponding pixels can be calculated and used as the NDVI of the green space quadrat.

[0064] As an example, the Normalized Difference Vegetation Index (NDVI) can be calculated using Sentinel-2 satellite data (10-meter resolution) from remote sensing imagery. The NDVI calculation formula used here is: NDVI = (NIR - RED) / (NIR + RED), where NIR is the reflectance in the near-infrared wavelength range, and RED is the reflectance in the red wavelength range. It is evident that NDVI is an important tool for assessing vegetation density, cover, growth status, and health, and is widely used worldwide for quality assessment of green spaces, farmland, etc., unlike other applications. This analysis, comparison, reasoning, and judgment are based on the characteristics of native plants and cultivated plants (including landscaping plants and grain crops) to differentiate them. Specifically, because cultivated plants require biomass removal for their respective purposes (e.g., landscaping plants require regular pruning, while cultivated plants require harvesting), the NDVI of native plants is higher than that of cultivated plants during the vegetation growing season (May to October each year). Therefore, the annual average NDVI is used in the algorithm and remote sensing image interpretation. Specifically, from May to October each year, four remote sensing images covering the target city are downloaded monthly, with cloud cover below 15%. These images are then merged, geometrically corrected, and cloud-covered pixels are removed. In the final analysis, the annual average NDVI is calculated at each pixel level. Based on this, the NDVI of the pixels corresponding to each green space quadrat is determined according to the location information of each green space quadrat. The average NDVI of these pixels is calculated and used as the NDVI of the corresponding green space quadrat.

[0065] S140. Calculate the quartiles of NDVI based on the NDVI corresponding to each informal green space quadrat, and use these quartiles as thresholds to perform a second classification of each green space quadrat based on the NDVI corresponding to each green space quadrat, thereby determining the second category of each green space quadrat.

[0066] The second category includes: cultivated plant areas, low-grade freestanding plant areas, moderate-grade freestanding plant areas, and high-grade freestanding plant areas; low-grade freestanding plant areas, moderate-grade freestanding plant areas, and high-grade freestanding plant areas belong to informal green spaces. This second classification can be further detailed as follows:

[0067] The NDVI values ​​of each informal green space quadrat were sorted in ascending order. The quartiles of the NDVI were calculated and used as thresholds. If the NDVI of a green space quadrat was lower than the first quartile, it was identified as a cultivated plant site. If the NDVI of a green space quadrat was between the first and second quartiles, it was identified as a low-degree free-growing plant site. If the NDVI of a green space quadrat was between the second and third quartiles, it was identified as a moderately free-growing plant site. If the NDVI of a green space quadrat was higher than the third quartile, it was identified as a highly free-growing plant site.

[0068] S150. Using remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels, a dataset is constructed. The initial urban informal green space identification model is trained based on the dataset to obtain the urban informal green space identification model.

[0069] In some embodiments, the dataset can be divided into a training set and a validation set in a 4:1 ratio. The initial urban informal green space identification model is then trained using the training set, and the performance of the trained urban informal green space identification model is validated using the validation set to obtain the final urban informal green space identification model.

[0070] For example, the above-mentioned model for identifying informal green spaces in cities can be a Random Forest (RF), the principle of which is as follows:

[0071] Random forest is a supervised classification method that extracts multiple samples from the original dataset. Each sample is used to construct a decision tree based on the principle of fastest entropy reduction. The predictions from these decision trees are then combined. Because decision trees are highly robust to interference, they can avoid overfitting in complex datasets. The optimal classification result is determined by the mode of the voting results from the decision trees. It has been widely applied in land use classification, land change detection, and biological and ecological sciences, demonstrating excellent accuracy. Figure 2 As shown, random forest classification mainly utilizes random forest to randomize the training samples, generate single decision trees, construct a decision tree forest, and count the votes with the most votes as the classification result.

[0072] Based on the urban informal green space identification model training method 100 provided in the embodiments of the present invention, the embodiments of the present invention also provide an urban informal green space identification method, such as... Figure 3 As shown, the urban informal green space identification method 300 may include the following steps:

[0073] S310: Acquire remote sensing image data of the target city.

[0074] The remote sensing image data of the target city is obtained by synthesizing multiple remote sensing image data of the target city taken during the vegetation growing season.

[0075] S320: Input remote sensing image data into the urban informal green space identification model, and the urban informal green space identification model will identify the remote sensing image data to obtain the urban informal green space identification result.

[0076] The urban informal green space identification model is obtained based on the training method for the urban informal green space identification model described above.

[0077] It is worth noting that, in order to facilitate a more intuitive display of the results of urban informal green space identification, the urban informal green space identification method 300 may also include:

[0078] Based on the identification results of informal green spaces in the city, a distribution map of informal green spaces in the city is generated.

[0079] To facilitate further understanding, the following detailed description of the training method 100 and the identification method 300 for the urban informal green space identification model are provided with reference to a specific embodiment, as shown below:

[0080] Taking the Chengdu Ring Road ecological zone and the area formed by a 500-meter buffer zone as an example, Chengdu (30°05'-31°26'N, 102°5'-104°53'E) is located in western Sichuan Province, in the heart of the Chengdu Plain, and is one of the central cities and transportation hubs in western my country. It is bordered by the Longquan Mountains to the east and the Longmen Mountains to the west. The terrain is high in the west and low in the east, mainly consisting of hills and mountains, with elevations mostly between 1000 and 3000 meters. The entire area has a subtropical humid monsoon climate with four distinct seasons, an average annual temperature of 16℃, and an average annual precipitation of 895.6 mm.

[0081] Driven by the construction of a park city, Chengdu, adhering to the concepts of "landscaped, scenic, accessible, and participatory," has created functional complexes through multi-faceted integration, constructing an ecological "green vein" that blends mountains and waters, connects urban and rural areas, and covers the entire region, achieving comprehensive green well-being. The Jinjiang Greenway forms a ring of seven wedge-shaped plots, with a total length of 100 kilometers for the entire primary greenway, spanning 12 districts, connecting 121 distinctive ecological parks, and featuring 78 scenic bridges, encompassing 133 square kilometers of ecological land. As the outer edge of Chengdu's central urban area, the ring area boasts diverse land use types, including farmland, parks, residential areas, and waterways, providing a typical area for identifying informal green spaces in the city.

[0082] The dataset here was collected using Google Earth Engine. The samples in this dataset form the basis for supervised classification of remote sensing images. The accuracy of supervised classification is related to the quality of the sample points; high-quality samples can significantly increase classification accuracy. Here, 3500 samples from the Chengdu Ring Expressway area were collected. The specific process was as follows: First, the Sentinel-level-2 A-grade remote sensing image data was used as the base map in GEE. Then, a new layer was created (eight layers were created in this study). Sample point names and corresponding attributes were assigned to each layer, such as highly naturalized vegetation, moderately naturalized vegetation, lightly naturalized vegetation, cultivated vegetation (green space, cultivated vegetation (farmland), developed construction land, roads, and water bodies). Because Sentinel-2 has a spatial resolution of 10m and the sample data has obvious texture features, most samples were selected through visual interpretation. For the three categories of highly spontaneously vegetated land (also known as highly spontaneously vegetated land), moderately spontaneously vegetated land (also known as moderately spontaneously vegetated land), and lightly spontaneously vegetated land (also known as lightly spontaneously vegetated land), reference data obtained from field sampling were matched, and then classified using the method described above. This improved operational efficiency and facilitated large-scale remote sensing. During model training, 80% of the samples were used for training and 20% for validation. The final model achieved an overall classification accuracy greater than 0.87, and the confusion matrix classification error was less than 0.15. After model training, the urban informal green space identification model can be used to identify Chengdu's remote sensing image data, yielding the results. Based on these results, a distribution map of Chengdu's urban informal green spaces can then be generated. Specifically, this can be done as follows... Figure 4 As shown.

[0083] In summary, according to the embodiments of the present invention, at least the following technical effects are achieved:

[0084] This invention not only enables efficient and accurate identification and assessment of the distribution of native plants in cities, but also provides crucial support for urban ecological planning, green space management, and biodiversity conservation. Implementing this invention can significantly improve the management efficiency of informal urban green spaces and promote the optimized allocation of urban green resources and the layout of green infrastructure.

[0085] Ultimately, the scientific and rational utilization and management of informal urban green spaces can not only promote the restoration and improvement of the urban ecological environment, but also provide strong support for addressing climate change and enhancing the resilience of cities to future challenges. This requires the joint efforts and participation of governments, research institutions, social organizations, and the public to achieve the long-term goal of sustainable urban development.

[0086] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0087] The above is an introduction to the method embodiments. The following describes the solution of the present invention further through device embodiments.

[0088] Figure 5 The diagram shows a structural diagram of a training device for an urban informal green space identification model provided in an embodiment of the present invention. Figure 5 As shown, the urban informal green space identification model training device 500 may include:

[0089] The determination module 510 is used to determine the green space quadrats in the target city based on the green space and ecological space distribution map of the target city, and to obtain the measured data of each green space quadrat.

[0090] The classification module 520 is used to classify each green space quadrat for the first time based on the measured data of each green space quadrat, and to determine the first category of each green space quadrat. The first category includes: cultivated plant land and informal green space.

[0091] The calculation module 530 is used to calculate the Normalized Difference Vegetation Index (NDVI) based on the remote sensing image data of the target city, and to determine the NDVI corresponding to each green space quadrat based on the location information of each green space quadrat.

[0092] The classification module 520 is also used to calculate the quartiles of NDVI based on the NDVI corresponding to each informal green space quadrat, and use these quartiles as thresholds to perform a second classification of each green space quadrat based on the NDVI corresponding to each green space quadrat, thereby determining the second category of each green space quadrat. The second category includes: cultivated plant land, low-degree free-growing plant land, moderate-degree free-growing plant land, and high-degree free-growing plant land; low-degree free-growing plant land, moderate-degree free-growing plant land, and high-degree free-growing plant land belong to informal green spaces.

[0093] Training module 540 is used to construct a dataset using remote sensing image data corresponding to each green space quadrat as samples and the second category corresponding to each green space quadrat as labels. The initial urban informal green space identification model is trained based on the dataset to obtain the urban informal green space identification model.

[0094] Understandable, Figure 5Each module / unit in the urban informal green space identification model training device 500 shown has the ability to implement Figure 1 The functions of each step in the training method 100 for identifying informal green spaces in cities, as shown, and the corresponding technical effects they achieve, will not be elaborated here for the sake of brevity.

[0095] Figure 6 A structural diagram of an urban informal green space identification device provided by an embodiment of this disclosure is shown, such as... Figure 6 As shown, the urban informal green space identification device 600 may include:

[0096] The acquisition module 610 is used to acquire remote sensing image data of the target city.

[0097] The identification module 620 is used to input remote sensing image data into the urban informal green space identification model, and the urban informal green space identification model identifies the remote sensing image data to obtain the urban informal green space identification result.

[0098] The urban informal green space identification model is obtained based on the training method for the urban informal green space identification model described above.

[0099] Understandable, Figure 6 Each module / unit in the urban informal green space identification device 600 shown has the ability to implement Figure 3 The functions of each step in the urban informal green space identification method 300 shown, and the corresponding technical effects they achieve, will not be elaborated here for the sake of brevity.

[0100] Figure 7 A structural diagram of an exemplary electronic device capable of implementing embodiments of the present invention is shown. Electronic device 700 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 700 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown in this invention, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0101] like Figure 7As shown, the electronic device 700 may include a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. The RAM 703 may also store various programs and data required for the operation of the electronic device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0102] Multiple components in electronic device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of displays, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows electronic device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0103] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as method 100 or method 300. For example, in some embodiments, method 100 or method 300 may be implemented as a computer program product, including a computer program tangibly contained in a computer-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of method 100 or method 300 described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform method 100 or method 300 by any other suitable means (e.g., by means of firmware).

[0104] The various embodiments described above in this invention can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), payload programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0105] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0106] In the context of this invention, a computer-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0107] It should be noted that the present invention also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute method 100 or method 300 and achieve the corresponding technical effects achieved by executing the methods in the embodiments of the present invention. For the sake of brevity, they will not be described in detail here.

[0108] In addition, the present invention also provides a computer program product, which includes a computer program that implements method 100 or method 300 when executed by a processor.

[0109] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this invention does not impose any limitations on them.

[0110] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for training an urban informal green space identification model, characterized in that, The method comprises: According to the green space and ecological space distribution map of the target city, determine the green plot in the target city, and obtain the measured data of each green plot, including species basic information and species composition information; According to the measured data of each green plot, the first classification of each green plot is performed to determine the first category of each green plot, wherein the first category includes cultivated plant land and informal green land; According to the remote sensing image data of the target city, the normalized difference vegetation index NDVI is calculated, and the NDVI corresponding to each green plot is determined according to the position information of each green plot; According to the NDVI corresponding to each informal green plot, the quartile of NDVI is calculated, and the second classification of each green plot is performed according to the NDVI corresponding to each green plot, to determine the second category of each green plot, wherein the second category includes cultivated plant land, low self-sustaining plant land, medium self-sustaining plant land and high self-sustaining plant land; Low self-sustaining plant land, medium self-sustaining plant land and high self-sustaining plant land belong to informal green land; Taking the remote sensing image data corresponding to each green plot as a sample and the second category corresponding to each green plot as a label, a data set is constructed, and an initial city informal green land recognition model is trained according to the data set to obtain a city informal green land recognition model; According to the remote sensing image data of the target city, the normalized difference vegetation index NDVI is calculated, and the NDVI corresponding to each green plot is determined according to the position information of each green plot, comprising: According to the multiple target city remote sensing image data taken in each month from May to October in the vegetation growing season, the average NDVI is calculated at each pixel level, and on this basis, the NDVI of the pixels corresponding to each green plot is determined according to the position information of each green plot, and for any green plot, the average value of the NDVI of the pixels corresponding thereto is calculated and taken as the NDVI corresponding to the green plot; The method further comprises: Obtain the remote sensing image data of the target city; Input the remote sensing image data into the city informal green land recognition model, and identify the remote sensing image data by the city informal green land recognition model to obtain the city informal green land recognition result; According to the NDVI corresponding to each informal green plot, the quartile of NDVI is calculated, and the second classification of each green plot is performed according to the NDVI corresponding to each green plot, to determine the second category of each green plot, comprising: The NDVI corresponding to each informal green space plot is sorted in ascending order, the quartiles of the NDVI are calculated, and the quartiles are used as thresholds; if the NDVI corresponding to the green space plot is lower than the first quartile, the green space plot is determined to be a cultivated plant land; if the NDVI corresponding to the green space plot is between the first quartile and the second quartile, the green space plot is determined to be a low-degree self-growing plant land; if the NDVI corresponding to the green space plot is between the second quartile and the third quartile, the green space plot is determined to be a medium-degree self-growing plant land; and if the NDVI corresponding to the green space plot is higher than the third quartile, the green space plot is determined to be a high-degree self-growing plant land.

2. The method of claim 1, wherein, The green space plot in the target city is determined according to the green space and ecological space distribution map of the target city, and includes: According to the green space and ecological space distribution map of the target city, the boundary range of each green space patch in the target city is determined, and green space plots are randomly arranged by using ArcGIS, and green space plots falling on buildings, roads and water surfaces are deleted, so as to finally determine the green space plots in the target city.

3. The method of claim 1, wherein, The first classification of each green space plot is performed according to the measured data of each green space plot, and the first category of each green space plot is determined, and includes: For any green space plot, the cultivated plant area ratio and the self-growing plant area ratio are calculated according to the measured data, and if the cultivated plant area ratio is greater than a preset threshold, the green space plot is determined to be a cultivated plant land; and if the self-growing plant area ratio is greater than a preset threshold, the green space plot is determined to be an informal green space.

4. The method of claim 1, wherein, The initial city informal green space identification model is trained according to the data set, and a city informal green space identification model is obtained, and includes: The data set is divided into a training set and a validation set; The initial city informal green space identification model is trained using the training set, and the performance of the trained city informal green space identification model is verified according to the validation set, so as to obtain a final city informal green space identification model.

5. The method of claim 1, wherein, The method further includes: According to the city informal green space identification result, a city informal green space distribution map is generated.

6. An urban informal green space identification model training apparatus, characterized in that, The device includes: A determination module is configured to determine green space plots in a target city according to a green space and ecological space distribution map of the target city, and to obtain measured data of each green space plot, wherein the measured data includes species basic information and species composition information; A classification module is configured to perform a first classification of each green space plot according to the measured data of each green space plot, and to determine a first category of each green space plot, wherein the first category includes a cultivated plant land and an informal green space; A calculation module is configured to calculate a normalized difference vegetation index (NDVI) according to remote sensing image data of the target city, and to determine NDVI corresponding to each green space plot according to position information of each green space plot. The classification module is further configured to calculate the quartile of NDVI according to the NDVI corresponding to each informal green space plot, and perform a second classification on each green space plot according to the NDVI corresponding to each green space plot, to determine a second category of each green space plot, wherein the second category includes cultivated plant land, low self-sustaining plant land, medium self-sustaining plant land, and high self-sustaining plant land; the low self-sustaining plant land, the medium self-sustaining plant land, and the high self-sustaining plant land belong to informal green space. The training module is configured to construct a data set by taking the remote sensing image data corresponding to each green space plot as a sample and taking the second category corresponding to each green space plot as a label, and train the initial urban informal green space identification model according to the data set to obtain the urban informal green space identification model. The calculation module is specifically configured to: calculate the average NDVI at each pixel level according to a plurality of target urban remote sensing image data taken in each month of the vegetation growing season from May to October, and then determine the NDVI of the pixels corresponding to each green space plot according to the position information of each green space plot, and calculate the average value of the NDVI of the pixels corresponding to any green space plot as the NDVI corresponding to the green space plot; The device further includes The acquisition module is configured to acquire remote sensing image data of a target city. The identification module is configured to input the remote sensing image data into the urban informal green space identification model, and identify the remote sensing image data by the urban informal green space identification model to obtain an urban informal green space identification result. The classification module is specifically configured to: sort the NDVI corresponding to each informal green space plot in ascending order, calculate the quartile of NDVI, and take the quartile as a threshold value; if the NDVI corresponding to a green space plot is lower than the first quartile, the green space plot is determined as cultivated plant land; if the NDVI corresponding to a green space plot is between the first quartile and the second quartile, the green space plot is determined as low self-sustaining plant land; if the NDVI corresponding to a green space plot is between the second quartile and the third quartile, the green space plot is determined as medium self-sustaining plant land; and if the NDVI corresponding to a green space plot is higher than the third quartile, the green space plot is determined as high self-sustaining plant land.

Images