Method and system for estimating building construction age based on breakpoint detection of remote sensing data

Abstract

The present invention relates to a method and system for estimating the construction age of a building based on breakpoint detection of remote sensing data, comprising: obtaining medium-resolution remote sensing time series data of a target building and using a CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data; for each pixel, calculating the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each initial detection breakpoint, and for each initial detection breakpoint, determining the target breakpoint representing the structural change of the building according to the absolute difference of the first-order amplitude and the absolute difference of the intercept; obtaining a building vector map of the current target building, and estimating the construction age of the target building according to the target breakpoint time values ​​of all pixels in the building vector frame in the building vector map. The present invention performs breakpoint detection based on medium-resolution remote sensing time series data, which can not only accurately estimate the construction age of a building, but also estimate the construction age of a large range of buildings.

Description

Technical Field

[0001] The invention relates to the technical field of remote sensing image processing, and in particular to a method and system for estimating the construction age of a building based on breakpoint detection of medium-resolution (10-60m) remote sensing long time series data. Background Art

[0002] The construction year (or age) of a building is one of the most important attribute parameters of a building, which is of great significance for assessing building energy consumption and building disaster vulnerability. However, the data on the construction age of buildings has not been well recorded and summarized for a long time around the world, resulting in the lack of such data in many areas. Therefore, some studies have begun to try to use geographic information data to estimate the construction age of buildings. For example, Tooke et al. (2014) used lidar data and random forest algorithm to estimate the construction age of buildings. Li et al. (2018) and Zeppelzaueret al. (2018) used deep learning technology to interpret street view photos or field photos to estimate the construction age of buildings. However, these methods are essentially indirect estimation methods that estimate the age of buildings by mining the relationship between building morphology and construction age. The relationship between building morphology and construction age may vary greatly in different regions. The existing methods cannot accurately estimate the construction age of buildings. At the same time, the field images or lidar data required by the existing methods are also very scarce in many areas, resulting in the inability to complete the estimation of the construction age of buildings in large areas. Therefore, the present invention proposes a method and system for estimating the construction age of a building based on breakpoint detection of medium-resolution (10-60m) remote sensing long time series data. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for estimating the construction age of a building based on breakpoint detection of remote sensing data, which obtains the time series breakpoints in the time series data by analyzing medium-resolution (10-60m) remote sensing long time series data, and accurately determines the estimated construction age of the target building in combination with the vector diagram of the target building.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A method for estimating the construction age of a building based on breakpoint detection of remote sensing data, comprising:

[0006] Obtaining medium-resolution remote sensing time series data of the target building and using a CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data; the medium-resolution remote sensing time series data refers to remote sensing time series data with a resolution of 10-60m;

[0007] For each of the pixel, the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints are calculated, and for each of the initial detection breakpoints, the target breakpoint representing the change of the building structure is determined according to the absolute difference of the first-order amplitude and the absolute difference of the intercept;

[0008] A building vector diagram of the current target building is obtained, and the construction age of the target building is estimated according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector diagram.

[0009] Optionally, the using of a CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data specifically includes:

[0010] For each of the pixels, remote sensing data in a first time period of the medium-resolution remote sensing time series data is obtained and data fitting is performed to obtain a harmonic function;

[0011] Calculating the fitting reflectivity value corresponding to each observation time point in the first time period according to the harmonic function;

[0012] Determining whether each of the observation time points is a potential breakpoint according to the fitted reflectance value and the actual reflectance value;

[0013] When there are a preset number of consecutive observation time points that are all potential breakpoints, the first potential breakpoint in the current sequence of consecutive potential breakpoints is determined to be the initial detection breakpoint; and after confirming the current initial detection breakpoint, a second time period is reselected and the second time period is set as the first time period, and the process returns to the step of "obtaining remote sensing data in the first time period of the medium-resolution remote sensing time series data and performing data fitting"; the second time period is the time period after the first time period;

[0014] When the initial detection breakpoint is not detected within the first time period, the first time period is expanded to obtain a third time period, and the third time period is set as the first time period, and the step of "obtaining remote sensing data within the first time period in the medium-resolution remote sensing time series data and performing data fitting to obtain a harmonic function" is returned until the initial detection breakpoint is detected within the current first time period.

[0015] Optionally, the expression of the harmonic function is:

[0016]

[0017] in, It represents the observed reflectance value of the i-th band on the x-th day; x represents the observation time point; i represents the remote sensing image band number; T represents the number of days in a year; It represents the fitted reflectivity value of the i-th band on the x-th day; the coefficient a 0,i , a 1,i , a 2,i , b 1,i , b 2,i , c i Represents the fitting coefficients of the harmonic function.

[0018] Optionally, the confirming whether each of the observation time points is a potential breakpoint according to the fitted reflectance value and the actual reflectance value specifically includes:

[0019] Calculating the difference between the fitted reflectance value and the actual reflectance value;

[0020] Calculate the root mean square error of the harmonic function fitted in the i-th band;

[0021] Determine whether the observed time point is the potential breakpoint according to the difference and the root mean square error.

[0022] Optionally, the formula used to confirm whether the observed time point is the potential breakpoint is:

[0023]

[0024] Where k represents the total number of remote sensing image bands; ρ(i, x) represents the actual reflectance value of the i-th band on the x-th day; RMSE i It represents the root mean square error of the harmonic function fitted in the i-th band.

[0025] Optionally, the calculating of the absolute difference between the first-order amplitude of each of the initial detection breakpoints in the visible light band and the absolute difference between the intercepts in the short-wave infrared band specifically includes:

[0026]

[0027] Among them, a 1，i1 (j+) and b 1，i1 (j+) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period before the time period where the j-th initial detection breakpoint is located in the visible light band i1; a 1，i1 (j-) and b 1，i1 (j-) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period after the time period where the j-th initial detection breakpoint is located in the visible light band i1; ΔA i It represents the absolute difference of the first-order amplitude between the time period before and after the time period of the jth initial detection breakpoint in the visible light band i1; a 0，swir (j+) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the time period before the time period where the j-th initial detection breakpoint is located; a 0，swir(j-) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the next time period after the time period where the j-th initial detection breakpoint is located; ΔI swir Represents the absolute difference of the intercepts of the shortwave infrared band.

[0028] Optionally, for each of the initial detection breakpoints, determining a target breakpoint representing a change in the building structure according to the first-order amplitude absolute difference and the intercept absolute difference specifically includes:

[0029] For each of the initially detected breakpoints, all candidate breakpoints representing changes in the building structure are determined according to the first-order amplitude absolute difference and the intercept absolute difference;

[0030] All the candidate breakpoints are sorted in chronological order and the candidate breakpoint which is the latest in time is obtained as the target breakpoint.

[0031] Optionally, determining all candidate breakpoints characterizing the structural change of the building according to the first-order amplitude absolute difference and the intercept absolute difference specifically includes:

[0032] Calculating the mean of the absolute difference of the first-order amplitude for each of the initial detection breakpoints;

[0033] When the mean is greater than a first threshold and / or the intercept absolute difference is greater than a second threshold, the current initial detection breakpoint is the candidate breakpoint.

[0034] Optionally, estimating the construction age of the target building according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map specifically includes:

[0035] Calculate the median of the target breakpoint time values ​​of all the pixels in the building vector frame;

[0036] determining the median value as the estimated construction time of the target building;

[0037] The construction year is determined based on the estimated construction time.

[0038] The present invention also provides a system for estimating the construction age of a building based on remote sensing data breakpoint detection, comprising:

[0039] The initial inspection breakpoint acquisition module is used to obtain the medium-resolution remote sensing time series data of the target building and use the CCD algorithm to detect the initial inspection breakpoint of each pixel in the medium-resolution remote sensing time series data; the medium-resolution remote sensing time series data refers to the remote sensing time series data with a resolution of 10-60m;

[0040] A target breakpoint determination module is used to calculate, for each pixel, the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints, and determine, for each of the initial detection breakpoints, a target breakpoint representing a change in the building structure according to the absolute difference of the first-order amplitude and the absolute difference of the intercept;

[0041] The construction age estimation module is used to obtain the building vector map of the current target building, and estimate the construction age of the target building according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] The present invention relates to a method and system for estimating the construction age of a building based on breakpoint detection of remote sensing data, comprising: obtaining medium-resolution remote sensing time series data of a target building and using a CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data; for each pixel, calculating the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each initial detection breakpoint, and for each initial detection breakpoint, determining the target breakpoint representing the structural change of the building according to the absolute difference of the first-order amplitude and the absolute difference of the intercept; obtaining a building vector map of the current target building, and estimating the construction age of the target building according to the target breakpoint time values ​​of all pixels in the building vector frame in the building vector map. The present invention performs breakpoint detection based on medium-resolution remote sensing time series data, and can accurately estimate the construction age of a building, and because the medium-resolution remote sensing time series data is data covering the global range, the present invention can also estimate the construction age of buildings in a large range. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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 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.

[0045] Figure 1 A flow chart of a method for estimating the construction age of a building based on breakpoint detection of remote sensing data provided in Example 1 of the present invention;

[0046] Figure 2 A block diagram showing the principle of estimating the construction age of a building provided in Example 1 of the present invention;

[0047] Figure 3 A schematic diagram of remote sensing data of a target building in a certain area provided by Example 1 of the present invention;

[0048] Figure 4 An example diagram of using the red light band amplitude difference and the short-wave infrared intercept difference to select breakpoints representing changes in building structure in step S2 provided in embodiment 1 of the present invention;

[0049] Figure 5 A diagram showing the estimated construction age of a target building in a certain area provided in Example 1 of the present invention;

[0050] Figure 6 A scatter plot of the estimated construction year and the actual construction year of the target building provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The purpose of the present invention is to provide a method and system for estimating the construction age of a building based on breakpoint detection of remote sensing data, which obtains the time series breakpoints in the time series data by analyzing medium-resolution (10-60m) remote sensing long time series data, and accurately determines the estimated construction age of the target building in combination with the vector diagram of the target building.

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Example 1

[0055] like Figure 1 and 2 As shown, this embodiment provides a method for estimating the construction age of a building based on breakpoint detection of remote sensing data, comprising:

[0056] S1: Obtain medium-resolution remote sensing time series data of the target building and use CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data. The medium-resolution remote sensing time series data refers to remote sensing time series data with a resolution of 10-60m.

[0057] Medium-resolution remote sensing data, represented by the Landsat series, have a long historical observation time series, with 30-meter resolution images dating back to 1985. Therefore, change detection based on remote sensing time series has the potential to be used to estimate the age of individual buildings built after 1985. Figure 3As shown, remote sensing time series data of target buildings in a certain area are shown, and the data are obtained from the 30-meter resolution Landsat remote sensing time series data that is freely available worldwide.

[0058] Among them, S1 specifically includes:

[0059] S11: For each of the pixels, partial remote sensing data in a first time period of the medium-resolution remote sensing time series data is obtained and data fitting is performed to obtain a harmonic function.

[0060] When fitting the harmonic function, only part of the data in the first time period can be selected. After fitting to the harmonic function, the observation time points in the first time period can be directly substituted into the harmonic function to obtain the fitted reflectivity value.

[0061] Wherein, the expression of the harmonic function is:

[0062]

[0063] Among them, x represents the observation time point; i represents the remote sensing image band number; T represents the number of days in a year; It represents the fitted reflectivity value of the i-th band on the x-th day; the coefficient a 0,i , a 1,i , a 2,i , b 1,i , b 2,i , c i represents the fitting coefficient of the harmonic function. Specifically, a 0,i is the intercept term, representing the long-term average value of reflectivity; a 1,i , a 2,i , b 1,i , b 2,i is the amplitude coefficient, which represents the amplitude of different frequencies; c i It is a trend term, which represents the long-term trend change of reflectivity.

[0064] The observation time point x represents the superimposed number of all observation time points in different years. For example, there are 10 observation time points in 1999 and 20 observation time points in 2000, so the value range of x is 1-30.

[0065] S12: Calculating the fitting reflectivity value corresponding to each observation time point in the first time period according to the harmonic function.

[0066] S13: confirming whether each of the observed time points is a potential breakpoint according to the fitted reflectance value and the actual reflectance value.

[0067] Wherein, step S13 specifically includes:

[0068] S131: Calculate the difference between the fitted reflectivity value and the actual reflectivity value.

[0069] S132: Calculate the root mean square error of the harmonic function fitted in the i-th band.

[0070] S133: Confirm whether the observation time point is the potential breakpoint according to the difference and the root mean square error.

[0071] The formula used to confirm whether the observed time point is the potential breakpoint is:

[0072]

[0073] Where k represents the total number of remote sensing image bands; ρ(i, x) represents the actual reflectance value of the i-th band on the x-th day; RMSE i It represents the root mean square error of the harmonic function fitted in the i-th band.

[0074] S14: When there are a preset number of consecutive observation time points that are all potential breakpoints, the first potential breakpoint in the current continuous sequence of potential breakpoints is identified as the initial detection breakpoint; and after confirming the current initial detection breakpoint, a second time period is reselected and the second time period is set as the first time period, and the process returns to step S11; the second time period is the time period after the first time period.

[0075] For example, when six consecutive observation time points are all potential breakpoints, the first of the six potential breakpoints is identified as the initial detection breakpoint.

[0076] S15: When the initial detection breakpoint is not detected within the first time period, the first time period is expanded to obtain a third time period, and the third time period is set as the first time period, and the process returns to step S11 until the initial detection breakpoint is detected within the current first time period.

[0077] When the initial detection breakpoint is not detected in the first time period, it means that there is no time breakpoint in the current first time period, and it is necessary to expand the range of the first time period, that is, add new observations to the original observations to update the fitted harmonic function and RMSE.

[0078] After step S1, the corresponding initial detection breakpoint is detected for each pixel in the remote sensing time series image.

[0079] S2: For each of the pixel, calculate the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints, and for each of the initial detection breakpoints, determine the target breakpoint representing the change of the building structure according to the absolute difference of the first-order amplitude and the absolute difference of the intercept.

[0080] In addition to characterizing the construction of the building, the initial inspection breakpoints detected in step S1 may also characterize the color changes caused by roof painting or renovation. Therefore, it is necessary to screen the initial inspection breakpoints to remove the breakpoints caused by roof painting or renovation. Although roof painting will change the color of visible light, it does not change the seasonal changes in the building structure and its shadows, and therefore will not change the first-order amplitude of the seasonal changes in the visible light band time series (mainly caused by changes in the sun angle); at the same time, due to the common spectral characteristics of building materials, roof painting or renovation has little effect on the long-term average reflectivity of the short-wave infrared band. Therefore, it is necessary to calculate the mean (ΔA) of the absolute difference of the first-order amplitude of the visible light band (i.e., the red, green, and blue RGB bands) of each initial inspection breakpoint. visual ) and the absolute difference between the intercepts in the short-wave infrared band (ΔI swir ).

[0081] Specifically, in step S2, for each of the initial detection breakpoints, determining the target breakpoint representing the change in the building structure according to the first-order amplitude absolute difference and the intercept absolute difference includes:

[0082] S21: For each of the initially detected breakpoints, all candidate breakpoints characterizing the structural changes of the building are determined according to the first-order amplitude absolute difference and the intercept absolute difference.

[0083] The calculation formula for the absolute difference of the first-order amplitude and the absolute difference of the intercept in the short-wave infrared band is:

[0084]

[0085] Among them, a 1，i1 (j+) and b 1，i1 (j+) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period before the time period where the j-th initial detection breakpoint is located in the visible light band i1; a 1，i1 (j-) and b 1，i1 (j-) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period after the time period where the j-th initial detection breakpoint is located in the visible light band i1; ΔA i It represents the absolute difference of the first-order amplitude between the time period before and after the time period of the jth initial detection breakpoint in the visible light band i1; a 0，swir (j+) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the time period before the time period where the j-th initial detection breakpoint is located; a 0，swir (j-) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the next time period after the time period where the j-th initial detection breakpoint is located; ΔI swir Represents the absolute difference of the intercepts of the shortwave infrared band.

[0086] Among them, S21 specifically includes:

[0087] S211: Calculate the mean of the absolute difference of the first-order amplitude for each of the initial detection breakpoints using the following formula:

[0088]

[0089] In the formula, RGB represents the red, green and blue RGB bands.

[0090] S212: When the mean is greater than the first threshold value T1 and / or the intercept absolute difference is greater than the second threshold value T2, the current initial detection breakpoint is the candidate breakpoint.

[0091] If the two differences corresponding to the initial detection breakpoints (ΔA visual With ΔI swir )When one of them exceeds a certain threshold, it is marked as a breakpoint representing the change of building structure:

[0092] ΔA visual >T 1 or ΔI swir >T 2 .

[0093] S22: Sort all the candidate breakpoints in chronological order and obtain the largest candidate breakpoint in time as the target breakpoint.

[0094] For the most commonly used Landsat long-term data, the visible light band is the 1st to 3rd band (Landsat8OLI takes the 2nd to 4th band), the short-wave infrared band is the 7th band (Landsat 8OLI takes the 8th band), and the threshold T 1 With T 2 After investigation and experimentation, the recommended values ​​are 0.0065 and 0.16 respectively. Figure 4 Comparing the spectral time series characteristics of roof painting and building reconstruction, it can be found that for roof reconstruction, although the intercept of the visible light band time series changes greatly (taking the red light band as an example in the figure), the seasonal amplitude changes very little, and the intercept change of the short-wave infrared band time series is also relatively small; while building reconstruction will greatly change the seasonal amplitude of the red light band time series and the intercept of the short-wave infrared band time series. Therefore, by setting two thresholds (ΔA visual >T 1 or ΔI swir >T 2 ) can filter out the breakpoints representing the changes in the building structure pixel by pixel, and then output the time of the most recent breakpoint.

[0095] S3: Obtain a building vector map of the current target building, and estimate the construction age of the target building according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map.

[0096] Overlay the vector map of a single building with the raster data of the latest breakpoint time (target breakpoint), extract the latest breakpoint time of all pixels intersecting with its vector frame for each building, then take the median of these latest breakpoint time values, which is the estimated construction time of the building, and finally convert it into the construction year output. Each pixel corresponds to a target breakpoint, and the target breakpoints of all pixels are combined to form raster data. Specifically, step S3 includes:

[0097] Calculate the median of the target breakpoint time values ​​of all the pixels in the building vector frame;

[0098] determining the median value as the estimated construction time of the target building;

[0099] The construction year is determined based on the estimated construction time.

[0100] Figure 5 A map showing the estimated construction dates of the buildings in the case area shows that the construction dates gradually become later from the city center outward, which is in line with the actual characteristics.

[0101] Figure 6 By comparing the estimated construction age of 149 building samples with the actual construction age, it can be found that the root mean square error of the estimated construction age is 7.1 years, and the estimated construction age error of 82% of the buildings is within 5 years. This shows that the method of the present invention can effectively estimate the construction age of single buildings in a large area.

[0102] In this embodiment, a continuous change detection (CCD) algorithm is used to fit harmonic functions pixel by pixel to medium-resolution remote sensing time series data and preliminarily detect time series breakpoints; based on the difference in amplitude of the harmonic function in the visible light band (red, green, and blue) before and after the initial detection breakpoint and the difference in intercept terms in the short-wave infrared band, the most recent time series breakpoints that characterize changes in building structures are screened, and the most recent breakpoint time is extracted pixel by pixel; the raster image of the most recent breakpoint time is superimposed with the vector layer of the individual building, and the most recent breakpoint time of all pixels intersecting with its vector frame is extracted for each building and the median is taken, which is the construction age of the building. This method uses breakpoint detection of medium-resolution remote sensing long time series data for the first time to accurately estimate the construction age of each building, providing a new technical means for the census of building age information in a large area.

[0103] Example 2

[0104] This embodiment provides a system for estimating the construction age of a building based on remote sensing data breakpoint detection, comprising:

[0105] The initial detection breakpoint acquisition module M1 is used to acquire the medium-resolution remote sensing time series data of the target building and detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data using a CCD algorithm;

[0106] The target breakpoint determination module M2 is used to calculate, for each pixel, the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints, and determine, for each of the initial detection breakpoints, the target breakpoint representing the change of the building structure according to the absolute difference of the first-order amplitude and the absolute difference of the intercept;

[0107] The construction age estimation module M3 is used to obtain the building vector map of the current target building, and estimate the construction age of the target building according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map.

[0108] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0109] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for estimating the construction age of a building based on breakpoint detection of remote sensing data, characterized in that: include: Acquire medium-resolution remote sensing time series data of the target building and use a CCD algorithm to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data; The medium-resolution remote sensing time series data refers to remote sensing time series data with a resolution of 10-60m; For each of the pixel, the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints are calculated, and for each of the initial detection breakpoints, the target breakpoint representing the change of the building structure is determined according to the absolute difference of the first-order amplitude and the absolute difference of the intercept; Obtaining a building vector diagram of the current target building, and estimating the construction age of the target building according to the target breakpoint time values ​​of all the pixels within the building vector frame in the building vector diagram; The CCD algorithm is used to detect the initial detection breakpoint of each pixel in the medium-resolution remote sensing time series data, specifically including: For each of the pixels, remote sensing data in a first time period of the medium-resolution remote sensing time series data is obtained and data fitting is performed to obtain a harmonic function; Calculating the fitting reflectivity value corresponding to each observation time point in the first time period according to the harmonic function; Determining whether each of the observation time points is a potential breakpoint according to the fitted reflectance value and the actual reflectance value; When there are a preset number of consecutive observation time points that are all potential breakpoints, the first potential breakpoint in the current sequence of consecutive potential breakpoints is determined to be the initial detection breakpoint; and after confirming the current initial detection breakpoint, a second time period is reselected and the second time period is set as the first time period, and the process returns to the step of "obtaining remote sensing data in the first time period of the medium-resolution remote sensing time series data and performing data fitting"; the second time period is the time period after the first time period; When the initial detection breakpoint is not detected in the first time period, the first time period is extended to obtain a third time period, the third time period is set as the first time period, and the process returns to the step of "obtaining remote sensing data in the first time period of the medium-resolution remote sensing time series data and performing data fitting to obtain a harmonic function", until the initial detection breakpoint is detected in the current first time period; Wherein, confirming whether each of the observation time points is a potential breakpoint according to the fitted reflectivity value and the actual reflectivity value specifically includes: Calculating the difference between the fitted reflectance value and the actual reflectance value; Calculate the root mean square error of the harmonic function fitted in the i-th band; confirming whether the observation time point is the potential breakpoint according to the difference and the root mean square error; The construction age of the target building is estimated according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map, specifically including: Calculate the median of the target breakpoint time values ​​of all the pixels in the building vector frame; determining the median value as the estimated construction time of the target building; The construction year is determined based on the estimated construction time.

2. The method according to claim 1, characterized in that The expression of the harmonic function is: in, It represents the observed reflectivity value of the i-th band on the x-th day; x represents the observation time point; i represents the remote sensing image band number; T represents the number of days in a year; It represents the fitted reflectivity value of the i-th band on the x-th day; the coefficient a 0,i , a 1,i , a 2,i , b 1,i , b 2,i , c i Represents the fitting coefficients of the harmonic function.

3. The method according to claim 1, characterized in that The formula used to confirm whether the observed time point is the potential breakpoint is: Where k represents the total number of remote sensing image bands; ρ(i, x) represents the actual reflectance value of the i-th band on the x-th day; RMSE i It represents the root mean square error of the harmonic function fitted in the i-th band.

4. The method according to claim 1, characterized in that: The calculating of the absolute difference between the first-order amplitude of each of the initial detection breakpoints in the visible light band and the absolute difference between the intercepts in the short-wave infrared band specifically includes: Among them, a 1，i1 (j+) and b 1，i1 (j+) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period before the time period where the j-th initial detection breakpoint is located in the visible light band i1; a 1，i1 (j-) and b 1，i1 (j-) respectively represent the first-order harmonic fitting coefficient of the harmonic function corresponding to the time period after the time period where the j-th initial detection breakpoint is located in the visible light band i1; ΔA i It represents the absolute difference of the first-order amplitude between the time period before and after the time period of the jth initial detection breakpoint in the visible light band i1; a 0，swir (j+) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the time period before the time period where the j-th initial detection breakpoint is located; a 0，swir (j-) represents the intercept fitting coefficient of the harmonic function in the short-wave infrared band in the next time period after the time period where the j-th initial detection breakpoint is located; ΔI swir Represents the absolute difference of the intercepts of the shortwave infrared band.

5. The method according to claim 1, characterized in that For each of the initial detection breakpoints, determining a target breakpoint representing a change in the building structure according to the first-order amplitude absolute difference and the intercept absolute difference specifically includes: For each of the initially detected breakpoints, all candidate breakpoints representing changes in the building structure are determined according to the first-order amplitude absolute difference and the intercept absolute difference; All the candidate breakpoints are sorted in chronological order and the candidate breakpoint which is the latest in time is obtained as the target breakpoint.

6. The method according to claim 5, characterized in that The step of determining all candidate breakpoints representing changes in the building structure according to the first-order amplitude absolute difference and the intercept absolute difference specifically includes: Calculating the mean of the absolute difference of the first-order amplitude for each of the initial detection breakpoints; When the mean is greater than a first threshold and / or the intercept absolute difference is greater than a second threshold, the current initial detection breakpoint is the candidate breakpoint.

7. A system for estimating the age of a building based on breakpoint detection of remote sensing data, used to execute the method for estimating the age of a building based on breakpoint detection of remote sensing data as claimed in any one of claims 1 to 6, characterized in that: include: The initial inspection breakpoint acquisition module is used to obtain the medium-resolution remote sensing time series data of the target building and use the CCD algorithm to detect the initial inspection breakpoint of each pixel in the medium-resolution remote sensing time series data; the medium-resolution remote sensing time series data refers to the remote sensing time series data with a resolution of 10-60m; A target breakpoint determination module is used to calculate, for each pixel, the absolute difference of the first-order amplitude in the visible light band and the absolute difference of the intercept in the short-wave infrared band of each of the initial detection breakpoints, and determine, for each of the initial detection breakpoints, a target breakpoint representing a change in the building structure according to the absolute difference of the first-order amplitude and the absolute difference of the intercept; The construction age estimation module is used to obtain the building vector map of the current target building, and estimate the construction age of the target building according to the target breakpoint time values ​​of all the pixels in the building vector frame in the building vector map.

Images