Urban renewal monitoring method and related equipment based on multi-factor temporal interaction characteristics

Abstract

The present invention relates to the field of urban renewal monitoring technology, and specifically to an urban renewal monitoring method and related equipment based on multi-factor time series interaction characteristics, the method comprising obtaining a time series image of a city, calculating the area ratio of each key urban landscape element in each pixel in the time series image, and constructing a time series of the area ratio of each key urban landscape element for the corresponding pixel; based on the above time series, generating a time series dynamic curve of the area ratio of each key urban landscape element; performing trend, breakpoint and peak-valley analysis on the time series dynamic curve to extract feature information; based on the feature information of multi-factor interaction and preset judgment criteria, determining the spatiotemporal distribution range of urban renewal and the renewal stage to which it belongs. In this way, a technical method for large-scale, high-precision spatiotemporal distribution of urban renewal and quantitative identification of the entire process is provided, thereby improving the accuracy of identifying the urban renewal development process and realizing large-scale continuous monitoring of the entire urban renewal process.

Description

Technical Field

[0001] The present invention relates to the technical field of urban renewal monitoring, and in particular to an urban renewal monitoring method and related equipment based on multi-factor temporal interaction characteristics. Background Art

[0002] Urban renewal is generally considered to be the process of replacing low, disordered and vast buildings with dense high-rise buildings and green landscapes. Obtaining accurate information on their spatial and temporal distribution in a timely manner is crucial for sustainable urban development and management.

[0003] As a type of transformation process with high frequency and complex process of urban cover, there are still certain challenges in carrying out complete, detailed and accurate remote sensing monitoring of the spatiotemporal evolution of urban renewal. At present, although the change detection method based on image difference and threshold division has the advantages of low image acquisition cost, simplicity and speed, it is often difficult to depict the detailed evolution process of urban land cover change, and the adaptability and reliability support for urban renewal process monitoring are insufficient; in addition, the traditional post-classification remote sensing monitoring method based on urban land classification is more often used. By establishing a multi-temporal vegetation, bare land, impervious surface, vacant land and other classification systems, the spatiotemporal characteristics of urban land renewal can be tracked using specific transformation rules, but the monitoring accuracy is overly dependent on the availability and accuracy of high-frequency land cover classification.

[0004] Therefore, how to efficiently and accurately carry out continuous monitoring of the entire process of large-scale urban renewal, so as to obtain accurate information on its temporal and spatial distribution, and then carry out sustainable urban development and management, has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide an urban renewal monitoring method and related equipment based on the time series interaction characteristics of multiple factors, so as to overcome the problem that the current urban renewal monitoring cannot accurately realize continuous monitoring of the entire process of large-scale urban renewal.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides an urban renewal monitoring method based on multi-factor time series interaction characteristics, comprising:

[0008] Acquire time series images of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series images, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation results; wherein the key landscape elements of the city include impervious surfaces, green vegetation, and bare land / vacant land;

[0009] Based on the time series of the area ratios of the key landscape elements of each city, a time series dynamic curve of the area ratios of the key landscape elements of each city is generated;

[0010] Based on a preset time series detection algorithm, trend, breakpoint and peak-valley analysis is performed on the time series dynamic curve to extract feature information;

[0011] Based on the characteristic information of the interaction of multiple key elements and preset judgment criteria, the spatiotemporal distribution range and renewal stage of urban renewal are determined.

[0012] Furthermore, in some embodiments of the present application, before calculating the area ratio of each key landscape element of each city in each pixel in the time series image, the method further includes:

[0013] Get the city boundary dataset;

[0014] Based on the boundary dataset, the city boundary range is generated by multi-source image superposition and maximum voting method;

[0015] Removing irrelevant land types within the city boundary to determine the area of ​​interest, wherein the irrelevant land types include roads, hills and small water bodies within the boundary;

[0016] The calculating the area ratio of each key urban landscape element in each pixel in the time series image includes: calculating the area ratio of each key urban landscape element in each pixel in the area of ​​interest in the time series image.

[0017] Further, in some embodiments of the present application, the calculating the area ratio of each key urban landscape element in each pixel in the time series image, and constructing a time series of the area ratio of each key urban landscape element for the corresponding pixel based on the calculation result, includes:

[0018] Calculating the area ratio of each basic urban landscape element in each pixel in the time series image by a regression model or a mixed pixel decomposition model; wherein the basic urban landscape elements include high-reflectivity landscape, low-reflectivity landscape, vegetation and bare soil;

[0019] Based on the area ratio of each basic urban landscape element in each pixel in the time series image, the area ratio of each key urban landscape element in each pixel in the time series image is determined, wherein the high-reflectivity landscape and the low-reflectivity landscape correspond to the impermeable surface, the vegetation corresponds to the green vegetation, and the bare soil corresponds to the bare land / vacant land.

[0020] Further, in some embodiments of the present application, the characteristic information includes: the maximum interference amount of the impervious surface, the initial value of the impervious surface, the final value of the impervious surface, the dominant period of green vegetation and bare land / vacant land, the peak time point of bare land / vacant land and the valley time point of the impervious surface;

[0021] Among them, the maximum interference amount of the impervious surface is used to characterize the maximum change in the area ratio of the impervious surface during the urban renewal process; the initial value of the impervious surface is used to characterize the average area ratio of the impervious surface in the early stage of urban renewal; the final value of the impervious surface is used to characterize the average area ratio of the impervious surface in the later stage of urban renewal; the dominant period of green vegetation and bare land / vacant land is used to characterize the period when the area ratio of bare land / vacant land and the area ratio of green vegetation occupy the largest proportion of the entire pixel area; the bare land / vacant land peak time point is used to characterize the year corresponding to the peak value of the bare land / vacant land area ratio in the time series dynamic curve; the impervious surface valley time point is used to characterize the year corresponding to the valley value of the impervious surface area ratio in the time series dynamic curve.

[0022] Furthermore, in some embodiments of the present application, the determining of the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information of the interaction of multiple key elements and preset judgment criteria includes:

[0023] If, in the time series dynamic curve of the target time series, the maximum interference amount of the impervious surface is greater than or equal to the first threshold, then the area corresponding to the target time series is determined as a potential urban change area;

[0024] If, in the time series dynamic curve of the target time series corresponding to the potential urban change area, the initial value of the impervious surface is greater than or equal to the second threshold, the area is determined to be a potential urban renewal area;

[0025] If, in the time series dynamic curve of the target time series corresponding to the potential urban renewal area, there is a period when the area ratio of the green vegetation and bare land / vacant land is greater than the area ratio of the impervious surface, the area is determined to be an urban renewal area.

[0026] Furthermore, in some embodiments of the present application, the determining of the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information and the preset judgment criteria further includes:

[0027] Based on the time information of the rapid increase in the area ratio of bare land / vacant land and the sharp decrease in the area ratio of impervious surface in the time series dynamic curve of the target time series corresponding to the urban renewal area, the renewal year of the urban renewal area is determined.

[0028] Furthermore, in some embodiments of the present application, the determining of the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information and the preset judgment criteria further includes:

[0029] If the terminal value of the impervious surface in the time series dynamic curve of the target time series corresponding to the urban renewal area is less than the second threshold, it is determined that the urban renewal area is in the demolition stage at the end time node of this target time series; if it is greater than or equal to the second threshold, it is determined that the urban renewal area is in the reconstruction stage at the end time node of this target time series.

[0030] Furthermore, in some embodiments of the present application, the trend, breakpoint and peak-valley analysis of the time series dynamic curve based on a preset time series segmentation algorithm includes:

[0031] Calculate the rate of change of the target time series and identify whether the trend of the target time series is increasing or decreasing;

[0032] Identify the breakpoints of the target time series by using the LandTrendr time series breakpoint detection method;

[0033] Based on a preset discriminant function, the peaks and valleys of the target time series are identified.

[0034] In a second aspect, an embodiment of the present application provides an urban renewal monitoring device based on multi-factor time series interaction characteristics, including:

[0035] The first generation module is used to obtain a time series image of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series image, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation result; wherein the key landscape elements of the city include impervious surface, green vegetation and bare land / vacant land;

[0036] A second generating module is used to generate a time series dynamic curve of the area ratio of each key landscape element in each city based on the time series of the area ratio of each key landscape element in each city;

[0037] An analysis and extraction module, used to perform trend, breakpoint and peak-valley analysis on the time series dynamic curve based on a preset time series segmentation algorithm, and extract feature information;

[0038] The judgment and determination module is used to determine the spatiotemporal distribution range and renewal stage of urban renewal based on the characteristic information of the interaction of multiple key elements and preset judgment criteria.

[0039] In a third aspect, an embodiment of the present application provides an urban renewal monitoring device based on multi-factor time series interaction characteristics, characterized in that it includes a processor and a memory, wherein the processor is connected to the memory:

[0040] Wherein, the processor is used to call and execute the program stored in the memory;

[0041] The memory is used to store the program, and the program is at least used to execute the above-mentioned urban renewal monitoring method based on multi-factor time series interaction characteristics.

[0042] The present invention relates to the field of urban renewal monitoring technology, and specifically to an urban renewal monitoring method and related equipment based on multi-element time series interaction characteristics, the method comprising obtaining a time series image of a city, calculating the area ratio of each key urban landscape element in each pixel in the time series image, and constructing a time series of the area ratio of each key urban landscape element for the corresponding pixel based on the calculation result; wherein the key urban landscape elements include impervious surfaces, green vegetation and bare land / vacant land; based on the time series of the area ratio of each key urban landscape element, a time series dynamic curve of the area ratio of each key urban landscape element is generated; based on a preset time series detection algorithm, the time series dynamic curve is analyzed for trend, breakpoint and peak and valley, and feature information is extracted; based on the feature information of multi-element interaction and preset judgment criteria, the spatiotemporal distribution range of urban renewal and the renewal stage to which it belongs are determined. In this way, a technical method for quantitative identification of the spatiotemporal distribution and the whole process of urban renewal with large range and high precision is provided, thereby improving the accuracy of identification of the urban renewal development process and realizing continuous monitoring of the whole process of large-scale urban renewal. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a flow chart of a method for monitoring urban renewal based on multi-factor temporal interaction characteristics provided by an embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of a process of generating a time series dynamic curve of the area ratio of each key landscape element of each city in the urban renewal monitoring method based on multi-element time series interaction characteristics provided by an embodiment of the present invention;

[0046] Figure 3It is an example schematic diagram of each characteristic parameter in the urban renewal monitoring method based on multi-factor time series interaction characteristics provided by an embodiment of the present invention;

[0047] Figure 4 It is a schematic diagram of a process for judging and determining monitoring results in an urban renewal monitoring method based on multi-factor time series interaction characteristics provided by an embodiment of the present invention;

[0048] Figure 5 It is a schematic diagram of an urban renewal process represented by characteristic parameters in an urban renewal monitoring method based on multi-factor time series interaction characteristics provided by an embodiment of the present invention;

[0049] Figure 6 It is a specific flow chart of the urban renewal monitoring method based on multi-factor time series interaction characteristics provided by an embodiment of the present invention;

[0050] Figure 7 It is a structural schematic diagram of an urban renewal monitoring device based on multi-factor time series interaction characteristics provided by an embodiment of the present invention;

[0051] Figure 8 It is a structural schematic diagram of an urban renewal monitoring device based on multi-factor time series interaction characteristics provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0053] Method Example:

[0054] Figure 1 This is a flow chart of the urban renewal monitoring method based on multi-factor time series interaction characteristics provided by an embodiment of the present invention. Figure 1 , this embodiment may include the following steps:

[0055] S101. Acquire time series images of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series images, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation results; and generate a time series dynamic curve of the area ratio of each key landscape element of the city based on the time series of the area ratio of each key landscape element of the city.

[0056] Among them, key urban landscape elements include impervious surfaces (ISA), green vegetation (GV) and bare land / vacant land (BS).

[0057] This step is actually the reconstruction of the time series of urban landscape components (i.e. key urban landscape elements). Specifically, the time series images can be long-time series, year-by-year wide-band remote sensing images, such as long-time series, year-by-year wide-band remote sensing images from 1985 to 2020 (in the images of all years, the pixels corresponding to the same actual urban area are corresponding, such as the first pixel in the image of each year corresponds to the same actual urban area. In this step of the present application, the pixels in the images of multiple years corresponding to the same actual urban area are referred to as pixels corresponding to a certain pixel).

[0058] After obtaining the above time series images, the area ratio of each key urban landscape element in each pixel in the time series images is calculated. At this time, for each pixel in the wide-band remote sensing image of each year, the area ratio of its impervious surface, green vegetation and bare land / vacant land, that is, the abundance value, can be obtained. Then, the area ratio of the same key urban landscape elements of pixels distributed in different images corresponding to the same actual urban area is used to generate a time series of the area ratio of the key urban landscape elements in chronological order. For example, the proportion area of ​​the impervious surface in the first pixel in each image is used to generate a time series of the area ratio of the impervious surface for the pixel, and a time series dynamic curve corresponding to the time series of the time area ratio is generated.

[0059] S102: Based on a preset time series detection algorithm, trend, breakpoint and peak-valley analysis is performed on the time series dynamic curve to extract feature information.

[0060] This step is actually the feature extraction of the multi-factor evolution process based on time series segmentation. Specifically, the feature information extracted through trend, breakpoint and peak-valley analysis may include the maximum disturbance of the impervious surface, the initial value of the impervious surface, the final value of the impervious surface, the dominant period of green vegetation and bare land / vacant land, the peak time point of bare land / vacant land and the valley time point of the impervious surface, which will be explained in detail later.

[0061] S103. Determine the spatiotemporal distribution range and renewal stage of urban renewal based on the characteristic information of the interaction of multiple key elements and preset judgment criteria.

[0062] This step is actually the construction and identification of the urban renewal area and process identification model based on multivariate comprehensive representation. Specifically, based on the interactive characteristic information of multiple key elements (i.e., impervious surface, green vegetation, and bare land / vacant land), the preset judgment criteria are used to make step-by-step judgments to determine the spatiotemporal distribution range of urban renewal and the renewal stage to which it belongs, where the renewal stage includes the demolition and reconstruction stages.

[0063] The urban renewal monitoring method based on the time series interaction characteristics of multiple elements provided in the present application first obtains the time series images of the city, calculates the area ratio of each key landscape element of the city in each pixel in the time series images, and constructs the time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation results; wherein the key landscape elements of the city include impervious surfaces, green vegetation and bare land / vacant land; then based on the time series of the area ratio of each key landscape element of the city, a time series dynamic curve of the area ratio of each key landscape element of the city is generated; then based on the preset time series detection algorithm, the trend, breakpoint and peak and valley analysis of the time series dynamic curve is performed to extract feature information; finally, based on the feature information of the interaction of multiple key elements and the preset judgment criteria, the spatiotemporal distribution range of urban renewal and the renewal stage to which it belongs are determined. In this way, a technical method for large-scale, high-precision spatiotemporal distribution and quantitative identification of the entire process of urban renewal is provided, thereby improving the accuracy of identification of the urban renewal development process and realizing large-scale continuous monitoring of the entire process of urban renewal.

[0064] Figure 2 : is a schematic diagram of a process of generating a time series dynamic curve of the area ratio of each key landscape element of each city in the urban renewal monitoring method based on the multi-element time series interaction characteristics provided by an embodiment of the present invention, such as Figure 2 As shown:

[0065] In an embodiment of the present application, based on the time series images of the city, the process of generating a time series dynamic curve of the area ratio of each city's key landscape elements pixel by pixel is the city's multi-element time series dynamic reconstruction process. In practical applications, the time series images of the city can be intensive time series remote sensing images.

[0066] Urban renewal is the process of replacing low, disordered and vast buildings with dense high-rise buildings and green landscapes. The urban renewal monitoring method provided by this application based on the time series interaction characteristics of multiple elements first establishes the dynamic evolution reconstruction and characterization of the area proportion of various basic landscape elements in urban renewal with time series information (time series information of time series images of the city), and establishes a time series dynamic curve of the landscape components of impervious surface-green vegetation-bare land / vacant land pixel by pixel, in order to fully track the entire process of urban demolition-reconstruction during the historical period of urbanization. The specific steps are as follows:

[0067] First, the city scope is defined based on multi-source images, that is, the city scope is defined by integrating multi-source images, specifically including: using the published city scope related data products (for example: GUB, Globeland30, GAIA, etc.) to determine the urbanization spatial scope, first, respectively obtain the first and last period (corresponding to the beginning and end of the time series image or the time series image to be studied) city boundary data sets, and use multi-source image overlay, maximum voting method, etc. to generate the first and last period high-confidence city boundary scope, and at the same time remove non-relevant land types such as roads, hills and small water bodies within the boundary, so as to define the region of interest (ROI) in the urban renewal monitoring method based on multi-factor time series interaction characteristics provided in this application, that is, the subsequent extraction and other processes are all performed on this region of interest.

[0068] On this basis, the area ratio information of the basic urban landscape elements is extracted: long-term series and wide-band remote sensing images are obtained year by year, and the area ratio of key urban landscape elements of each pixel is calculated through regression models, mixed pixel decomposition models, etc. (that is, the abundance value of urban landscape is estimated). Among them, in practical applications, the urban mixed space in this process is composed of four basic urban landscape elements, namely high-reflectivity landscape, low-reflectivity landscape, vegetation and bare soil. High-reflectivity features refer to buildings and roads made mainly of concrete, cement, metal and glass. Low-reflectivity features are mainly concentrated in old urban areas dominated by old buildings and heavy industrial areas with serious dust pollution, corresponding to the impervious surfaces in the key urban landscape elements (that is, the area ratio of impervious surfaces is determined based on the area ratio of high-reflectivity landscapes and low-reflectivity landscapes). Vegetation includes green vegetation in urban parks and other green areas, corresponding to the green vegetation in the key urban landscape elements. Bare land is scattered in urban vacant land with low vegetation coverage, or in areas where soil is exposed due to building demolition, corresponding to bare land / vacant land in the key urban landscape elements.

[0069] On this basis, the construction of the time series dynamic curve of urban multi-factors based on intensive time series remote sensing is completed. Based on long time series and wide-band intensive remote sensing images year by year, the dynamic change curve of the area ratio of urban multi-factors (i.e., key urban landscape elements) with the inter-annual time series is constructed at the pixel scale, and the corresponding quality inspection is carried out (the inspection can be carried out by Figure 2 The RMSE and MSE indices shown are used for quality verification, which can be understood by referring to the prior art and will not be described in detail here).

[0070] On this basis, feature information is extracted based on the dynamic change curve.

[0071] In an embodiment of the present application, before extracting feature information, it is necessary to first determine the feature variables to be extracted, including performing trend, breakpoint and peak-valley analysis on the time series dynamic curve based on a preset time series segmentation algorithm, such as calculating the rate of change of the target time series based on a preset trend test algorithm, and identifying the trend corresponding to the target time series as increasing or decreasing; identifying the breakpoints of the target time series through the LandTrendr time series breakpoint detection method; identifying the peaks, valleys, sudden declines and sudden rises of the target time series based on a preset discriminant function, and completing the multivariate extraction of the urban renewal evolution process based on time series segmentation.

[0072] Specifically, the first step is to analyze the trend. The specific principles are as follows:

[0073] First, the time series of the area ratio of each key landscape element of each city generated by each pixel (no special statement is made below, the time series mentioned are all this time series, in which in a certain analysis, it can be the time series of the area ratio of a key landscape element of a city of a certain pixel, or it can be the time series within a part of its time range, or it can be more time series, which is determined according to the actual research and analysis needs, so as to realize the trend analysis of the time series of the entire city through multiple analyses) is SG smoothed, and then based on the SG smoothed time series Y, its time series dynamic curve is analyzed. In the embodiment of the present application, a linear regression model is selected to analyze the trend of the time series, and the specific formula is as follows:

[0074]

[0075] Among them, X i is the sequence number of the SG smoothed time series Y (X i =1,2,3....,n), N is the length of the time series signal (N=30), Y i is the SG smoothed time series Y value at position i, Slope is the rate of change of the SG smoothed time series Y. For the above processing, if Slope>0, it means that the trend of the time series signal is increasing, otherwise it is a decreasing trend.

[0076] It should be noted that since the non-parametric MK test has no restrictions on the distribution of samples and does not significantly interfere with outliers, this method is often not used for trend testing of ecological environment time series. Therefore, the embodiment of the present application can also use the MK test to verify the significance of the time series trend based on time series trend detection.

[0077] Specifically, the null hypothesis (H0) of the nonparametric MK test is that the time series samples are all from a population with independent realizations and these samples are equally distributed, while the alternative hypothesis (H1) is that the time series presents a monotonic trend. For a time series Y with N independent samples, for any two points k in the time series, j≤N., and if k≠j, the test statistic S is defined as:

[0078]

[0079] Among them, the sgn function can be defined as,

[0080]

[0081] Since the statistic S of the MK test follows a normal distribution and E[S] = 0, the variance of S Var(S) can be defined as:

[0082]

[0083] Where p is a parallel array in the time series Y, t j is the frequency of the p-value in the parallel array. Since the statistic S of the MK test follows a normal distribution, it is Z-transformed to obtain:

[0084]

[0085] Through the MK test, when Z is greater than 0, it can be considered that the trend of the time series is rising, and when Z is less than 0, the trend of the time series is falling. In addition, when the absolute value of Z is greater than 1.96, it can be recognized that the trend change of the time series is significant at a confidence level of 95%.

[0086] Furthermore, for breakpoint analysis and detection, the specific principles are as follows:

[0087] The LandTrendr time series breakpoint detection method is used to fit the pixel-scale spectral trajectory through a series of straight lines to identify the sudden and gradual characteristics of the time series, so as to exclude noise to the greatest extent while retaining the necessary details. Among them, the input information of the algorithm mainly includes the inter-annual time series of the abundance values ​​(i.e., area ratio) of each urban element per pixel (A j-1 , A j , A j+1 ) and the corresponding year date (j-1, j, j+1). The main steps include: eliminating peak outliers caused by noise; identifying potential breakpoints; fitting trajectory lines; simplifying the model (i.e., the formula model used to analyze and detect breakpoints); determining the optimal model, etc.

[0088] Furthermore, for the analysis and identification of peaks, valleys, sudden drops and sudden rises, the specific principles are as follows:

[0089] The time series spectral trajectory segmentation method of wide-band remote sensing (such as Landsat satellite) is adopted, and the peak and valley characteristics of the time series are identified through the innovative time series breakpoint detection algorithm, so as to distinguish the breakpoint types and change points of multi-factor coupling. The model algorithm process is divided into the following steps:

[0090] Determine peaks and valleys through peak / valley detection algorithm analysis:

[0091] The embodiment of the present invention constructs a discriminant function G(j) to separate the time series (A j-1 , A j , A j+1 ) peak / valley and mutation, the specific formula is as follows:

[0092]

[0093] Among them, 0 is a sequence without mutation, p is a peak, v is a valley, F(j) is the result obtained by the breakpoint detection algorithm using the above-mentioned Landtrendr, F(j) = 1 means that there is a breakpoint at time series j, and F(j) = 0 means that there is no breakpoint at time series j.

[0094] Determine the trend through trend local change function analysis:

[0095] In addition, in order to judge the effectiveness and availability of the detected breakpoints, a trend local change function H(j) is constructed to evaluate the trend and degree of breakpoint change. The specific formula is as follows:

[0096]

[0097] For a time series breakpoint where G(j) is not 0, the embodiment of the present application evaluates the degree of the time series breakpoint based on the change of A between the breakpoint j and the next breakpoint z.

[0098] Model Evaluation:

[0099] In order to verify the accuracy of breakpoint detection, the piecewise linear modeling method can be used to simulate the time series based on the detected breakpoints, and the simulated residual mean error (RMSE) can be used to evaluate the breakpoints.

[0100] On the basis of the above analysis and detection, the characteristic parameters of the multi-factor evolution process are extracted, that is, through the analysis and detection of time series segmentation algorithms such as trend detection and valley / peak detection, the characteristic parameters that can describe the multi-factor and multi-stage evolution process of the city are derived and determined. Combined with the target time series in the actual monitoring process, the characteristic parameters of the multi-factor process such as urban impervious surface-vegetation-bare land, that is, the characteristic information, are determined, which can provide multi-dimensional input variables for the accurate identification and division of multiple construction process stages such as urban renewal, demolition, idleness, and reconstruction.

[0101] The specific characteristic parameters may include at least one of: the maximum disturbance of the impervious surface, the initial value of the impervious surface, the final value of the impervious surface, the dominant period of green vegetation and bare land / vacant land, the peak time point of bare land / vacant land, and the valley time point of the impervious surface;

[0102] Among them, the maximum interference amount of impervious surface ISA_magnitude is used to characterize the maximum change in the area ratio of impervious surface during the urban renewal process, such as Figure 3 a(in Figure 3 In the box of ac, the solid black line represents the time series dynamic curve generated based on the time series of the impervious surface); the initial value of the impervious surface, i.e., ISA-initial, is used to characterize the average value of the area ratio of the impervious surface in the early stage of urban renewal (the early stage, late stage, first stage and final stage in this application are all based on the time information of the time series), as shown in Figure 3 b; the final value of impervious surface, i.e., the final value of ISA, is used to characterize the average value of the area ratio of impervious surface in the later stage of urban renewal, as shown in Figure 3 c; the period dominated by green vegetation and bare land / vacant land is F _BS-GV The dominant period is used to characterize the period when the area ratio of bare land / vacant land and the area ratio of green vegetation occupy the largest proportion of the entire pixel area, such as Figure 3 d shows the peak time of bare land / idle land, i.e. P BS The time node is used to represent the year corresponding to the peak value of the bare land / vacant land area ratio in the time series dynamic curve, such as Figure 3 e; the valley point of the impervious surface is V ISA The time node is used to represent the year corresponding to the valley value of the area ratio of impervious surface in the time series dynamic curve, such as Figure 3 f.

[0103] After the above analysis and extraction of characteristic parameters of the time series dynamic curve of the time series to be monitored, the spatiotemporal distribution range and the renewal stage of urban renewal are determined based on the characteristic information and preset judgment criteria. Figure 4It is a flow chart of determining the monitoring results in the urban renewal monitoring method based on the multi-factor time series interaction characteristics provided by the embodiment of the present invention. In the present application, the process is the construction and determination process of the multi-criteria determination model of the urban renewal area and process development stage. In the embodiment of the present application, an identification framework of the spatiotemporal distribution range and process stage of urban renewal based on multi-factor coupling-multi-variable representation-multi-criteria determination is established, and an urban renewal model and process monitoring algorithm model is constructed. Specifically, Figure 4 As shown, there are several key steps:

[0104] The first step is to determine the scope of the urban renewal area:

[0105] It should be noted that urban renewal activities will inevitably lead to changes in landscape patterns and a significant change in the proportion of impervious surface areas. Therefore, the embodiment of the present application uses ISA_magnitude≥σ (i.e., the first threshold) to identify potential urban change areas, while excluding pseudo-change noise caused by image quality, that is, if ISA_magnitude≥σ in the time series dynamic curve of the time series, then the area corresponding to the time series is judged to be a potential urban change area (if it does not meet the judgment criterion, it can be judged as a stable urban area); secondly, the initial condition represents the development and utilization activities in the existing urban built-up areas, so in the embodiment of the present application, ISA_initial≥β (i.e., the second threshold) is used to identify whether the potential urban change area is a potential urban renewal area, that is, if ISA_initial≥β, then the potential urban change area is determined to be a potential urban renewal area (if it does not meet the judgment criterion, it can be judged as other urban change areas); finally, during the urban renewal process, building demolition will occur, resulting in the addition of a large number of bare land areas, presenting a dominant stage in which the proportion of bare land and vegetation areas exceeds the proportion of impervious surfaces. Therefore, in the embodiment of the present application, through F _BS-GV >F _ISA To determine whether the potential urban renewal area is an urban renewal area, that is, if there is F _BS-GV >F _ISA The period of urban potential renewal is judged as the urban renewal area (if it does not meet the judgment criteria, it can be judged as others), thus realizing the determination of the spatial scope of the urban renewal area.

[0106] Furthermore, the time for urban renewal is determined:

[0107] It should be noted that urban renewal will cause a rapid rise in bare land areas and a sharp drop in impervious surface areas. Therefore, in the embodiment of the present application, the years in which P_BS and V_ISA appear can be used as a necessary condition for identifying the renewal year; in addition, in order to prevent noise problems caused by pseudo-changes, the turning points can be further filtered by comparing the images before and after the renewal year determined by P_BS and V_ISA, so as to more accurately locate the urban renewal year.

[0108] Furthermore, the urban renewal stage is determined, that is, the renewal stage to which the urban renewal belongs:

[0109] It should be noted that the urban renewal process includes different development stages. The first stage is the demolition stage, which is the conversion of impervious surfaces into bare land and green vegetation. After demolition, the abundance value of impervious surfaces is lower than that before the change. Figure 5 As shown, ISA_end<β can be used to determine whether urban renewal is in the demolition stage; and after the demolition stage, bare land and green vegetation will be converted back into high-density impervious surfaces, that is, the abundance value of the impervious surface after reconstruction is not lower than the average abundance value of the stage before the change, that is, ISA_end≥β can be used to determine whether urban renewal is in the reconstruction stage.

[0110] The urban renewal monitoring method based on multi-factor time series interaction characteristics provided in this application first generates time series data of impervious surface-vegetation-bare land abundance that characterizes the urban renewal evolution process based on time series images, and then constructs a dynamic curve of urban multi-factor coupling time series; then, based on the above curve, the multivariate representation of the urban renewal evolution process is determined, and the characteristic information is extracted; finally, the urban renewal area and the process development stage are judged based on the extracted characteristic information and multiple criteria. The overall process is as follows: Figure 6 As shown. The urban renewal monitoring method provided in this application based on multi-factor time series interaction characteristics constructs a multivariate determination criterion for urban renewal based on multi-factor end-member time series characteristics such as impervious surface, vegetation and bare soil / vacant land, which is helpful for large-scale and accurate distinction and identification of urban renewal scenes and other similar scenes; as well as multi-dimensional feature extraction and fusion based on urban renewal geological knowledge, which provides precise guidance for the full-cycle and multi-stage process identification of urban renewal, and is conducive to the development of remote sensing monitoring technology suitable for large-scale and full-process urban renewal spatiotemporal patterns.

[0111] Device Example:

[0112] Based on the same inventive concept, the present invention also provides an urban renewal monitoring device based on multi-factor temporal interaction characteristics, which is used to implement the above method embodiment. Figure 7 Schematic diagram of the structure of the urban renewal monitoring device based on multi-factor time series interaction characteristics provided by an embodiment of the present invention. Figure 7 As shown, the device comprises:

[0113] The first generation module 11 is used to obtain a time series image of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series image, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation result; wherein the key landscape elements of the city include impervious surface, green vegetation and bare land / vacant land;

[0114] The second generating module 12 is used to generate a time series dynamic curve of the area ratio of each key landscape element in each city based on the time series of the area ratio of each key landscape element in the city;

[0115] The analysis and extraction module 13 is used to perform trend, breakpoint and peak-valley analysis on the time series dynamic curve based on a preset time series detection algorithm to extract feature information;

[0116] The judgment and determination module 14 is used to determine the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information of the interaction of multiple key elements and preset judgment criteria.

[0117] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0118] Device Example:

[0119] Based on the same inventive concept, the present invention also provides an urban renewal monitoring device based on multi-factor time series interaction characteristics, which is used to implement the above method embodiment. Figure 8 Schematic diagram of the structure of the urban renewal monitoring device based on multi-factor time series interaction characteristics provided by an embodiment of the present invention. Figure 8 As shown, the urban renewal monitoring device based on multi-factor time series interaction characteristics of this embodiment includes a processor 21 and a memory 22, and the processor 21 is connected to the memory 22. The processor 21 is used to call and execute the program stored in the memory 22; the memory 22 is used to store the program, and the program is at least used to execute the urban renewal monitoring method based on multi-factor time series interaction characteristics in the above embodiment.

[0120] The specific implementation scheme of the urban renewal monitoring device based on multi-factor time series interaction characteristics provided in the embodiment of the present application can refer to the implementation scheme of the urban renewal monitoring method based on multi-factor time series interaction characteristics in any of the above embodiments, which will not be repeated here.

[0121] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0122] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0123] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0124] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0125] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0127] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0128] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0129] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for monitoring urban renewal based on multi-factor temporal interaction characteristics, characterized in that: include: Acquire time series images of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series images, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation results; wherein the key landscape elements of the city include impervious surfaces, green vegetation, and bare land / vacant land; Based on the time series of the area ratios of the key landscape elements of each city, a time series dynamic curve of the area ratios of the key landscape elements of each city is generated; Based on a preset time series detection algorithm, trend, breakpoint and peak-valley analysis is performed on the time series dynamic curve to extract feature information; Based on the interactive characteristic information of multiple key elements and preset judgment criteria, determine the spatiotemporal distribution range and renewal stage of urban renewal; The characteristic information includes: the maximum interference of impervious surface, the initial value of impervious surface, the final value of impervious surface, the dominant period of green vegetation and bare land / vacant land, the peak time point of bare land / vacant land and the valley time point of impervious surface; wherein the maximum interference of impervious surface is used to characterize the maximum change of the area ratio of impervious surface in the urban renewal process; the initial value of impervious surface is used to characterize the average value of the area ratio of impervious surface in the early stage of urban renewal; the final value of impervious surface is used to characterize the average value of the area ratio of impervious surface in the late stage of urban renewal; the dominant period of green vegetation and bare land / vacant land is used to characterize the period when the area ratio of bare land / vacant land and the area ratio of green vegetation occupy the largest proportion of the entire pixel area; the peak time point of bare land / vacant land is used to characterize the year corresponding to the peak value of the area ratio of bare land / vacant land in the time series dynamic curve; the valley time point of impervious surface is used to characterize the year corresponding to the valley value of the area ratio of impervious surface in the time series dynamic curve; The feature information based on the interaction of multiple key elements and preset judgment criteria are used to determine the spatiotemporal distribution range and renewal stage of urban renewal, including: if the maximum interference amount of the impervious surface in the time series dynamic curve of the target time series is greater than or equal to a first threshold, then the area corresponding to the target time series is determined to be a potential urban change area; if the initial value of the impervious surface in the time series dynamic curve of the target time series corresponding to the potential urban change area is greater than or equal to a second threshold, then the area is determined to be a potential urban renewal area; if there is a period in the time series dynamic curve of the target time series corresponding to the potential urban renewal area when the area ratio of the green vegetation and bare land / vacant land is greater than the area ratio of the impervious surface, then the area is determined to be an urban renewal area.

2. The urban renewal monitoring method based on multi-factor temporal interaction characteristics according to claim 1 is characterized in that: Before calculating the area ratio of each key landscape element of each city in each pixel in the time series image, the following steps are also included: Get the city boundary dataset; Based on the boundary dataset, the city boundary range is generated by multi-source image superposition and maximum voting method; Removing irrelevant land types within the city boundary to determine the area of ​​interest, wherein the irrelevant land types include roads, hills and small water bodies; The calculating the area ratio of each key urban landscape element in each pixel in the time series image includes: calculating the area ratio of each key urban landscape element in each pixel in the area of ​​interest in the time series image.

3. The urban renewal monitoring method based on multi-factor temporal interaction characteristics according to claim 1 is characterized in that: The calculating the area ratio of each key urban landscape element in each pixel in the time series image, and constructing a time series of the area ratio of each key urban landscape element for the corresponding pixel based on the calculation result, includes: Calculating the area ratio of each basic urban landscape element in each pixel in the time series image by a regression model or a mixed pixel decomposition model; wherein the basic urban landscape elements include high-reflectivity landscape, low-reflectivity landscape, vegetation and bare soil; Based on the area ratio of each basic urban landscape element in each pixel in the time series image, the area ratio of each key urban landscape element in each pixel in the time series image is determined, wherein the high-reflectivity landscape and the low-reflectivity landscape correspond to the impermeable surface, the vegetation corresponds to the green vegetation, and the bare soil corresponds to the bare land / vacant land.

4. The urban renewal monitoring method based on multi-factor temporal interaction characteristics according to claim 3 is characterized in that: The determining of the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information and the preset judgment criteria further includes: Based on the time information of the rapid increase in the area ratio of bare land / vacant land and the sharp decrease in the area ratio of impervious surface in the time series dynamic curve of the target time series corresponding to the urban renewal area, the renewal year of the urban renewal area is determined.

5. The urban renewal monitoring method based on multi-factor temporal interaction characteristics according to claim 4 is characterized in that: The determining of the spatiotemporal distribution range and the renewal stage of urban renewal based on the characteristic information and the preset judgment criteria further includes: If the terminal value of the impervious surface in the time series dynamic curve of the target time series corresponding to the urban renewal area is less than the second threshold, it is determined that the urban renewal area is in the demolition stage at the end time node of this target time series; if it is greater than or equal to the second threshold, it is determined that the urban renewal area is in the reconstruction stage at the end time node of this target time series.

6. The urban renewal monitoring method based on multi-factor temporal interaction characteristics according to claim 1 is characterized in that: The trend, breakpoint and peak-valley analysis of the time series dynamic curve based on the preset time series detection algorithm includes: Calculate the rate of change of the target time series and identify whether the trend of the target time series is increasing or decreasing; Identify the breakpoints of the target time series by using the LandTrendr time series breakpoint detection method; Based on a preset discriminant function, the peaks and valleys of the target time series are identified.

7. An urban renewal monitoring device based on multi-factor temporal interaction characteristics, characterized in that: Used to perform the method according to any one of claims 1 to 6, comprising: The first generation module is used to obtain a time series image of the city, calculate the area ratio of each key landscape element of the city in each pixel in the time series image, and construct a time series of the area ratio of each key landscape element of the city for the corresponding pixel based on the calculation result; wherein the key landscape elements of the city include impervious surface, green vegetation and bare land / vacant land; A second generating module is used to generate a time series dynamic curve of the area ratio of each key landscape element in each city based on the time series of the area ratio of each key landscape element in each city; An analysis and extraction module, used to perform trend, breakpoint and peak-valley analysis on the time series dynamic curve based on a preset time series detection algorithm, and extract feature information; The judgment and determination module is used to determine the spatiotemporal distribution range and renewal stage of urban renewal based on the characteristic information of the interaction of multiple key elements and preset judgment criteria.

8. An urban renewal monitoring device based on multi-factor temporal interaction characteristics, characterized in that: The invention comprises a processor and a memory, wherein the processor is connected to the memory: Wherein, the processor is used to call and execute the program stored in the memory; The memory is used to store the program, and the program is at least used to execute the urban renewal monitoring method based on multi-factor time series interaction characteristics as described in any one of claims 1-6.

Images