Natural resource element dynamic monitoring method and system

Abstract

The invention provides a natural resource element dynamic monitoring method and system, and solves the problem that the spatial-temporal dynamic change of a large-area impervious surface is difficult to conveniently and systematically describe in the prior art. The method comprises the following steps: preparing a multi-time-sequence impervious surface data set; dynamically monitoring the time sequence change of the impervious surface; analyzing the driving force of the sequential change of the impervious surface; dynamically monitoring the spatial form change of the impervious surface; and dynamically monitoring the spatial directivity change of the impervious surface. According to the method, dynamic changes of the space-time pattern of the impervious surface are systematically described, urban space planning can be scientifically compiled, and reference is provided for optimizing the urban human settlement environment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of natural resource monitoring, and particularly relates to a natural resource element dynamic monitoring method and system. BACKGROUND

[0002] Urban expansion causes environmental problems such as heat island effect, urban waterlogging, and reduced biodiversity. By dynamically and systematically monitoring the urban expansion process, people can more scientifically understand the urban development situation, thereby providing a key basis for urban planning and decision-making and promoting high-quality urban development. As an artificial surface composed of cement, brick, asphalt and other materials that hinder the natural infiltration of rainwater, impervious surface is a core index for measuring the intensity of urban expansion. Dynamic monitoring of its spatiotemporal distribution has important strategic significance for optimizing urban spatial layout and promoting the coordinated development of ecology and cities.

[0003] Current impervious surface monitoring research has the following problems: Most methods focus on single-dimensional analysis in the time or spatial dimension, making it difficult to achieve systematic dynamic monitoring in the time-space coupling dimension; and the extraction method based on machine learning has problems such as low computational efficiency and poor generalization ability, which restricts the practical application of large-scale multi-temporal monitoring. SUMMARY

[0004] The application provides a natural resource element dynamic monitoring method and system, which realizes systematic characterization of the spatiotemporal pattern evolution process of natural resource elements through multi-dimensional spatiotemporal data analysis, effectively solving the problem that the prior art cannot conveniently and systematically describe the spatiotemporal dynamic changes of large-area impervious surface.

[0005] In a first aspect, a natural resource element dynamic monitoring method comprises: calculating an impervious surface coverage index FISI, segmenting the impervious surface coverage index FISI, and constructing a multi-temporal impervious surface dataset based on the segmentation result. The process of calculating the impervious surface coverage index FISI comprises: based on the blue, green, near-infrared and short-wave infrared band data of the remote sensing image, extracting the surface spectral feature; for each pixel of the remote sensing image, the following operations are performed: determining whether the pixel is water body by improved normalized difference water index, calculating the preliminary value based on the linear combination of the blue band and the near-infrared band, and subtracting the improved normalized difference water index to suppress the water body signal, and finally obtaining the impervious surface coverage index FISI; if there is no water body interference, the linear combination of the blue band and the near-infrared band is directly used to calculate the impervious surface coverage index FISI; based on the multi-temporal impervious surface dataset, the impervious surface area and the change speed and the impervious surface expansion intensity and the area proportion are calculated to evaluate the temporal change of the impervious surface in the research time; the correlation coefficient between the social economic index and the impervious surface area is calculated, the contribution rate of different social economic indexes to the temporal change of the impervious surface is evaluated by regression learning, and the driving effect of social economic development on the change of the impervious surface is evaluated; based on the multi-temporal impervious surface dataset, the shape change of the impervious surface in the research time is evaluated by the landscape pattern index; based on the multi-temporal impervious surface dataset, the directional change of the impervious surface in the research time is evaluated by the directional analysis, which comprises: dividing the research area into different quadrants, and comparing the similarities and differences of the dynamic changes of the impervious surface in different directions; the directional features of the spatial distribution of the impervious surface are described by the standard deviation ellipse.

[0006] In a second aspect, a natural resource element dynamic monitoring system comprises: a dataset construction module configured to: calculate an impervious surface coverage index FISI, segment the impervious surface coverage index FISI, and construct a multi-temporal impervious surface dataset based on the segmentation result, wherein the process of calculating the impervious surface coverage index FISI comprises: extracting ground spectral features based on remote sensing image data of blue, green, near-infrared and short-wave infrared bands; performing the following operations on each pixel of the remote sensing image: determining whether the pixel is water by a modified normalized difference water index, segmenting and calculating: if there is water, calculating a preliminary value based on a linear combination of the blue band and the near-infrared band, and subtracting the modified normalized difference water index to suppress the water signal, and finally obtaining the impervious surface coverage index FISI; if there is no water interference, directly calculating the impervious surface coverage index FISI using a linear combination of the blue band and the near-infrared band; a first evaluation module configured to calculate the impervious surface area and the change speed and the impervious surface expansion intensity and the area proportion based on the multi-temporal impervious surface dataset to evaluate the temporal change of the impervious surface in the research time; a second evaluation module configured to calculate the correlation coefficient between the social economic indicators and the impervious surface area, and evaluate the contribution rate of different social economic indicators to the temporal change of the impervious surface by regression learning to evaluate the driving effect of social economic development on the change of the impervious surface; a third evaluation module configured to evaluate the shape change of the impervious surface in the research time based on the multi-temporal impervious surface dataset through landscape pattern index; evaluate the directional change of the impervious surface in the research time based on the multi-temporal impervious surface dataset through directional analysis, which comprises: dividing the research area into different quadrants, and comparing the similarities and differences of the dynamic changes of the impervious surface in different directions; and describing the directional features of the impervious surface in spatial distribution through standard deviation ellipse.

[0007] In a third aspect, a computer comprises a processor and a memory storing one or more computer program modules configured to be executed by the processor to implement the natural resource element dynamic monitoring method.

[0008] In a fourth aspect, a non-volatile computer readable storage medium is provided, which stores a computer program capable of implementing the natural resource element dynamic monitoring method when executed by a computer.

[0009] The present application uses a multi-temporal natural resource element dataset to describe the dynamic changes of natural resource elements in the research time from the aspects of time and space, uses multiple indicators to systematically represent the spatio-temporal changes of natural resource elements, and analyzes the influence of social and economic factors on the dynamic changes of natural resource elements. The present application systematically describes the dynamic changes of the spatio-temporal pattern of natural resource elements, which is helpful for scientifically formulating land space planning and providing reference for guaranteeing ecological safety. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a flow chart of a natural resource element dynamic monitoring method according to an embodiment of the present application. DETAILED DESCRIPTION

[0011] Figure 1 A natural resource element dynamic monitoring process is shown. With the aid of the impervious surface coverage index and time constraints, the present application can efficiently construct a multi-temporal large-area impervious surface dataset, describe the dynamic changes of the impervious surface in the study time from the aspects of time and space, use multiple indicators to systematically represent the spatio-temporal changes of the impervious surface, and analyze the influence of social and economic factors on the dynamic changes of the impervious surface. The present application systematically describes the dynamic changes of the spatio-temporal pattern of the impervious surface, which is helpful for scientifically preparing urban spatial planning and providing a reference for optimizing urban living environment. The following will be described in detail Figure 1 the method shown.

[0012] Step 1, calculate the impervious surface coverage index (Fractional Impervious Surface Index, FISI) and construct a multi-temporal impervious surface dataset.

[0013] Step 1.1, calculate the impervious surface coverage index (FISI) based on remote sensing images.

[0014] FISI is a continuous value index calculated by remote sensing bands, which represents the impervious surface coverage degree of each pixel (between 0 and 1), and reflects the relative proportion of the impervious surface on the ground. The FISI value of each pixel is a real number between 0 and 1 (such as 0.3 represents 30% impervious surface coverage). Specifically, the calculation of FISI needs to use the blue band (Blue), the green band (Green), the near-infrared band (NIR) and the short-wave infrared band (SWIR). The index is processed by combining the modified normalized difference water index (MNDWI) for segmentation, and the formula is as follows:

[0015]

[0016] FISI enhances the ability to distinguish between impervious surface and water body through a dynamic segmentation function. When , subtract MNDWI to suppress water body interference.

[0017] Step 1.2: construct a multi-temporal impervious surface dataset based on the FISI segmentation result (FISI_Bin).

[0018] The FISI segmentation result is the result of binarizing the FISI index. Generally, a threshold (such as the 95th percentile) is set to convert the continuous FISI value into a binary classification layer of 0 or 1. That is, when the continuous FISI value is converted into a binarization result (0 or 1) by setting a threshold (such as the 95th percentile), only when the FISI value is greater than or equal to the threshold, it is marked as 1 (impervious surface), otherwise it is marked as 0 (non-impervious surface).

[0019] In the study area, select impervious surface samples (such as buildings, roads) and count their FISI value distribution. Take the upper limit of FISI value covering 95% of the samples as the segmentation threshold (i.e. the 95th percentile), to ensure high confidence extraction of impervious surface. Impervious surface result of initial time phase Determined from the initial FISI binarization result ( ), the subsequent time phase (t≥1) uses a dynamic updating rule:

[0020] Wherein, is the impervious surface extraction result of the current time phase, is the impervious surface extraction result of the previous time phase, is the impervious surface coverage index segmentation result of the current time phase. It can be seen that is the combination of historical data and the of the current time phase The final impervious surface distribution result generated by the dynamic updating rule.

[0021] Note: (current time phase impervious surface extraction result) and (current time phase impervious surface coverage index segmentation result) are two different concepts. is the binarization result (0 or 1) of the FISI index of the current time phase, which only reflects the preliminary classification of the current time phase. is a static result, based only on the FISI value and threshold of the current time phase, and does not depend on historical data. Combination of historical data and the dynamic updating result of the current time phase . is a dynamic result: continuity is maintained through time series rules (such as retaining historical labels and suppressing short-term fluctuations).

[0022] Through the above steps, the construction of the multi-time series impervious surface dataset is finally completed. The dataset is composed of binarized impervious surface distribution layers The composition can reflect the dynamic change characteristics of the impervious surface in the research area, provide high-confidence spatio-temporal data support for urban expansion monitoring, ecological impact assessment, and provide data sources for subsequent natural resource element spatio-temporal dynamic monitoring.

[0023] Step 2, based on the multi-temporal impervious surface data set, the temporal change of the impervious surface in the research time is evaluated.

[0024] Step 2.1, calculate the impervious surface area and change speed.

[0025] The impervious surface area is calculated by multiplying the number of pixels by the area of a single pixel, and the growth rate is calculated by dividing the area change by the research time.

[0026]

[0027]

[0028] In the formula, is the area of the impervious surface in the research area, is the total number of pixels of the impervious surface in the research area, is the actual area of a pixel in the research area. is the area change speed of the impervious surface in the research time, is the research start time, is the area of the impervious surface at the beginning of the research, is the research end time, is the area of the impervious surface at the end of the research.

[0029] Step 2.2, calculate the impervious surface expansion intensity and area proportion.

[0030] Expansion intensity represents the relative expansion speed of the impervious surface in the research time, and area proportion represents the proportion of the impervious surface area in the research area, and the calculation formula is as follows:

[0031]

[0032] Where U is the expansion intensity of the impervious surface, is the research start time, is the area of the impervious surface at the beginning of the research, is the research end time, is the area of the impervious surface at the end of the research. is the proportion of the impervious surface in the research area, is the area of the impervious surface in the research area, is the total area of the research area.

[0033] Step 3: Combine the temporal changes of impervious surfaces with socio-economic indicators to assess the driving effects of socio-economic development on impervious surface changes.

[0034] Step 3.1: Calculate the correlation coefficient between socio-economic indicators and impervious surface area.

[0035] The correlation coefficient is represented by the Pearson coefficient, and the calculation formula is as follows:

[0036] where, is the Pearson coefficient, X is the area of a certain impervious surface, is the mean value of a certain impervious surface in the study period, Y is the socio-economic indicator, is the mean value of the socio-economic indicator in the study period.

[0037] Step 3.2: Assess the contribution rate of different socio-economic indicators to impervious surface changes through regression learning.

[0038] Specifically, the influence of socio-economic factors on the dynamic changes of impervious surfaces is represented by the contribution rate given during the training process of the regression learning machine. Methods such as random forest can assess the contribution rate of input variables when learning the relationship between socio-economic factors and impervious surfaces, which can be used to represent the influence of different socio-economic indicators on the dynamic changes of impervious surfaces.

[0039] Step 4: Based on the multi-temporal impervious surface dataset, assess the shape changes of impervious surfaces in the study period through landscape pattern indices.

[0040] Step 4.1: Calculate the area, perimeter, and number of impervious surface patches. Specifically, first segment the impervious surface, and consider the same kind of impervious surface pixel connected to each other as a patch. Then, calculate the area, perimeter, and number of patches.

[0041] Step 4.2: Calculate the landscape pattern indices based on the above parameters.

[0042] The selected landscape pattern indices include compactness, fractal dimension, patch density, and average patch area, and the calculation formulas are as follows:

[0043]

[0044]

[0045]

[0046] where C is compactness, D is fractal dimension, PD is patch density, MPS is mean patch size, A is the area of impervious surface, P is the perimeter of impervious surface, N is the number of impervious surface patches in the study area, is the total area of the study area.

[0047] Step 5, based on the multi-temporal impervious surface dataset, assess the directional changes of impervious surface within the study period through directional analysis.

[0048] Step 5.1, assess the directional differences of impervious surface changes through the quadrant method.

[0049] The origin of the quadrant method is the geometric center of the study area, then the study area is divided into different quadrants by four or eight, and the similarities and differences of impervious surface dynamic changes in different directions are compared. The calculation formula of the geometric center is as follows:

[0050]

[0051] where, , are the X and Y coordinates of the geometric center of the study area, respectively. n is the number of pixels in the study area. and are the X and Y coordinates of the i-th pixel, respectively.

[0052] Step 5.2, assess the overall directional changes of impervious surface through standard deviation ellipse.

[0053] Standard deviation ellipse can describe the directional characteristics of impervious surface in spatial distribution, including four elements of center, direction, major axis and minor axis. The calculation formula is as follows:

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] where, and respectively represent the X coordinate and Y coordinate of the center of the standard deviation ellipse. , are the spatial coordinates of each impervious surface. and represent the arithmetic mean center of the impervious surface. is the direction of the standard deviation ellipse, and are the differences between the mean center and the coordinates of a certain impervious surface pixel. and are the lengths of the X axis and Y axis of the standard deviation ellipse, respectively.

[0062] The present application provides a natural resource element dynamic monitoring system. The system comprises a dataset construction module, a first evaluation module, a second evaluation module, a third evaluation module and a fourth evaluation module.

[0063] The dataset construction module is configured to calculate an impervious surface coverage index FISI, segment the impervious surface coverage index FISI, and construct a multi-temporal impervious surface dataset based on the segmentation results. The process of calculating the impervious surface coverage index FISI includes: based on the blue, green, near-infrared and short-wave infrared band data of the remote sensing image, extracting the surface spectral feature; for each pixel of the remote sensing image, the following operations are performed: determining whether the pixel is water by improved normalized difference water index, segmenting and calculating: if there is water, calculating the preliminary value based on the linear combination of the blue band and the near-infrared band, and subtracting the improved normalized difference water index to suppress the water signal, and finally obtaining the impervious surface coverage index FISI; if there is no water interference, directly use the linear combination of the blue band and the near-infrared band to calculate the impervious surface coverage index FISI.

[0064] The first evaluation module is configured to calculate the impervious surface area and the change speed, and the impervious surface expansion intensity and the area ratio based on the multi-temporal impervious surface dataset, to evaluate the temporal change of the impervious surface within the research time.

[0065] The second evaluation module is configured to calculate the correlation coefficient between the social economic indicators and the impervious surface area, and evaluate the contribution rate of different social economic indicators to the temporal change of the impervious surface through regression learning, to evaluate the driving effect of social economic development on the change of the impervious surface.

[0066] The third evaluation module is configured to evaluate the shape change of the impervious surface within the research time through landscape pattern index based on the multi-temporal impervious surface dataset.

[0067] The fourth evaluation module is configured to evaluate the directional change of the impervious surface in the research time based on the multi-temporal impervious surface data set through directional analysis, which comprises: dividing the research area into different quadrants, and comparing the similarities and differences of the dynamic changes of the impervious surface in different directions; and describing the directional characteristics of the impervious surface in the spatial distribution through the standard deviation ellipse.

[0068] Here, only the components of the system are outlined, and the specific implementation manner can be seen from the embodiment description of the natural resource element dynamic monitoring method associated therewith.

[0069] The application further provides a computer. The computer comprises a processor and a memory. The memory and the processor can be interconnected through a bus system and / or other forms of connection mechanism. The processor can comprise any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP). The memory can comprise a volatile memory such as a random access memory (RAM). The memory can further comprise a non-volatile memory such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid state drive (SSD). The memory stores executable program codes, and the processor executes the executable program codes to implement the natural resource element dynamic monitoring method. That is, the memory stores instructions for executing the natural resource element dynamic monitoring method.

[0070] The application further provides a computer readable storage medium. For example, the computer readable storage medium is a non-transitory computer readable storage medium such as a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device. The computer readable storage medium is used to store non-transitory computer readable instructions, and when the non-transitory computer readable instructions are executed by a computer, one or more steps of the natural resource element dynamic monitoring method can be implemented.

Claims

1. A method for dynamic monitoring of natural resource elements, characterized in that: include: Calculate the impervious surface coverage index (FISI), segment the impervious surface coverage index (FISI), and construct a multi-time series impervious surface dataset based on the segmentation results. The process of calculating the impervious surface coverage index (FISI) includes: extracting surface spectral characteristics based on the blue, green, near-infrared, and short-wave infrared band data of the remote sensing image; performing the following operations on each pixel of the remote sensing image: judging whether the pixel is a water body by using the modified normalized difference water index, and calculating in segments: if there is a water body, calculate the preliminary value based on the linear combination of the blue band and the near-infrared band, and subtract the modified normalized difference water index to suppress the water body signal, and finally obtain the impervious surface coverage index (FISI); if there is no water body interference, directly use the linear combination of the blue band and the near-infrared band to calculate the impervious surface coverage index (FISI); Based on multiple time-series impervious surface datasets, the area and change rate of impervious surfaces, as well as the intensity and area proportion of impervious surface expansion, were calculated to evaluate the temporal changes of impervious surfaces during the study period. The correlation coefficient between socioeconomic indicators and impervious surface area was calculated, and the contribution rate of different socioeconomic indicators to the temporal changes of impervious surface area was evaluated through regression learning to assess the driving effect of socioeconomic development on the changes of impervious surface area. Based on a multi-temporal impervious surface dataset, the shape changes of impervious surfaces during the study period were evaluated using landscape pattern indices. Based on a multi-temporal impervious surface dataset, directional changes in the impervious surface over the study period were evaluated through directional analysis. This included dividing the study area into quadrants and comparing the similarities and differences in the dynamic changes of the impervious surface in different directions; and describing the directional characteristics of the spatial distribution of the impervious surface using standard deviational ellipses.

2. The method according to claim 1, characterized in that The calculation formula of impervious surface coverage index FISI is: Where Blue represents the blue band, Green represents the green band, NIR represents the near-infrared band, SWIR represents the short-wave infrared band, and MNDWI represents the modified normalized difference water index.

3. The method according to claim 2, characterized in that The correlation coefficient between socioeconomic indicators and impervious surface area is the Pearson coefficient.

4. The method according to claim 3, characterized in that Landscape pattern indices include compactness, fractal dimension, patch density and average patch area.

5. A natural resource element dynamic monitoring system, characterized in that: include: The dataset construction module is configured to: calculate the impervious surface coverage index FISI, segment the impervious surface coverage index FISI, and construct a multi-time series impervious surface dataset based on the segmentation results. The process of calculating the impervious surface coverage index FISI includes: extracting surface spectral characteristics based on the blue, green, near-infrared and short-wave infrared band data of the remote sensing image; performing the following operations on each pixel of the remote sensing image: judging whether the pixel is a water body by using the modified normalized difference water index, and calculating in segments: if there is a water body, calculating a preliminary value based on the linear combination of the blue band and the near-infrared band, and subtracting the modified normalized difference water index to suppress the water body signal, and finally obtaining the impervious surface coverage index FISI; if there is no water body interference, directly using the linear combination of the blue band and the near-infrared band to calculate the impervious surface coverage index FISI; A first evaluation module is configured to calculate the area and change rate of the impervious surface and the expansion intensity and area ratio of the impervious surface based on a multi-time series impervious surface dataset, so as to evaluate the temporal change of the impervious surface during the study period; The second evaluation module is configured to calculate the correlation coefficient between the socioeconomic indicators and the impervious surface area, and evaluate the contribution rate of different socioeconomic indicators to the temporal change of the impervious surface through regression learning, so as to evaluate the driving effect of socioeconomic development on the change of the impervious surface; The third evaluation module is configured to evaluate the shape change of impervious surfaces during the study time through a landscape pattern index based on a multi-temporal impervious surface dataset; The fourth evaluation module is configured to evaluate the directional changes of impervious surfaces during the study period through directional analysis based on a multi-time series impervious surface dataset. The module includes: dividing the study area into different quadrants and comparing the similarities and differences in the dynamic changes of impervious surfaces in different directions; describing the directional characteristics of the spatial distribution of impervious surfaces through standard deviation ellipses.

6. The system according to claim 5, characterized in that The calculation formula of impervious surface coverage index FISI is: Where Blue represents the blue band, Green represents the green band, NIR represents the near-infrared band, SWIR represents the short-wave infrared band, and MNDWI represents the modified normalized difference water index.

7. The system according to claim 5, characterized in that The correlation coefficient between socioeconomic indicators and impervious surface area is the Pearson coefficient.

8. The system according to claim 5, wherein: Landscape pattern indices include compactness, fractal dimension, patch density and average patch area.

9. A computer, characterized in that: It includes a processor and a memory, the memory stores one or more computer program modules, and the computer program modules are configured to be executed by the processor to implement the dynamic monitoring method for natural resource elements according to any one of claims 1 to 4.

10. A non-volatile computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, the method for dynamic monitoring of natural resource elements described in any one of claims 1 to 4 can be implemented.

Images