Hydro-junction construction range extraction method and system based on remote sensing image

Abstract

The invention relates to the technical field of remote sensing images, in particular to a hydro-junction construction range extraction method and system based on a remote sensing image, and the method comprises the steps: obtaining a remote sensing image of a hydro-junction region before construction, and determining a foundation construction range; obtaining a remote sensing image in construction, performing feature extraction in combination with the remote sensing image before construction, and determining a dynamic change range of each time sequence point; identifying a temporary change range according to the dynamic change range, and further obtaining a dynamic construction range; calculating a temporary environment influence index and a permanent environment influence index of each time sequence point to obtain a construction environment influence index sequence; extracting a non-foundation construction range according to the dynamic construction range and the foundation construction range, and calculating a non-foundation range index sequence; and monitoring the construction environment influence index sequence and the non-basic range index sequence in real time, and performing construction planning. According to the method, the problem of influence of temporary construction facilities and areas can be solved, and the precision of hydraulic engineering range extraction is improved.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing image technology, and in particular to a method and system for extracting the construction scope of a water conservancy project based on remote sensing images. Background Art

[0002] Water conservancy projects are a vital component of infrastructure development, playing a key role in flood control, power generation, and irrigation. With the rapid development of the water conservancy industry, the scale of large-scale water conservancy projects has continued to expand, with construction areas often spanning several to tens of square kilometers. During construction, accurately defining and monitoring the construction scope is crucial for protecting the ecological environment, ensuring construction safety, and controlling project costs. Especially for water conservancy projects in ecologically sensitive areas, strictly controlling the scope of construction activities to avoid unnecessary damage to the surrounding environment has become a key task in project management.

[0003] Currently, the construction scope of water conservancy projects relies primarily on two technical methods: manual field measurement and remote sensing image analysis. While manual field measurement is highly accurate, it is time-consuming and labor-intensive, and it is difficult to promptly reflect dynamic changes in the construction scope. Remote sensing image analysis, on the other hand, utilizes high-resolution imagery acquired by satellites or drones to identify construction areas through image processing and feature extraction. This method offers the advantages of wide coverage and a short acquisition cycle.

[0004] However, the construction of water conservancy projects often involves numerous temporary facilities, such as construction camps, temporary material yards, and access roads. The location and scope of these facilities fluctuate frequently as the construction progresses. This dynamic nature makes it difficult for traditional construction scope extraction methods to accurately distinguish between permanent projects and temporary facilities, and it also makes it difficult to effectively assess the environmental impact of construction activities. This leads to significant errors in construction scope definition results, compromising the scientific nature of construction management decisions. Therefore, accurately identifying and monitoring the dynamic changes in construction scope, particularly those affecting temporary facilities, has become a pressing technical challenge in the construction management of water conservancy projects.

[0005] Therefore, a method and system for extracting the construction scope of a water conservancy hub based on remote sensing images are proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for extracting the construction scope of a water conservancy hub based on remote sensing images. By obtaining pre-construction remote sensing images of the water conservancy hub area, the basic construction scope is determined; remote sensing images during construction are obtained, and feature extraction is performed in combination with the pre-construction remote sensing images to determine the dynamic change range of each time point; the temporary change range is identified based on the dynamic change range, and the dynamic construction range is further obtained; the temporary environmental impact index and the permanent environmental impact index of each time point are calculated to obtain a construction environmental impact index sequence; the non-basic construction scope is extracted based on the dynamic construction scope and the basic construction scope, and a non-basic scope index sequence is calculated; the construction environmental impact index sequence and the non-basic scope index sequence are monitored in real time to carry out construction planning. The present invention can solve the problem of the impact of temporary construction facilities and areas, and improve the accuracy of water conservancy project scope extraction.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for extracting the construction scope of a water conservancy project based on remote sensing images includes:

[0009] Obtain pre-construction remote sensing images of the water conservancy hub area to determine the scope of foundation construction;

[0010] Acquire the in-construction remote sensing image of the water conservancy project area at each time point according to a preset period, perform feature extraction on the in-construction remote sensing image and the pre-construction remote sensing image, and determine the dynamic change range of each time point;

[0011] Identifying a temporary change range according to the dynamic change range, and defining a difference between the temporary change range and the dynamic change range as a dynamic construction range;

[0012] Calculate the temporary environmental impact index according to the temporary change range, calculate the permanent environmental impact index according to the dynamic construction range, and calculate the construction environmental impact index sequence according to the temporary environmental impact index and the permanent environmental impact index of all time points;

[0013] The difference between the dynamic construction range and the basic construction range at each time point is defined as the non-basic construction range, and a non-basic range index sequence is calculated based on the non-basic construction ranges at all time points;

[0014] The construction environment impact index sequence and the non-basic range index sequence are monitored in real time to carry out construction planning.

[0015] Furthermore, determining the dynamic change range of each time point includes: extracting spectral feature vectors and texture feature vectors from the during-construction remote sensing image of each time point to obtain a first comprehensive feature vector; extracting pre-construction spectral feature vectors and pre-construction texture feature vectors from the pre-construction remote sensing image to obtain a second comprehensive feature vector; calculating the feature difference measure of each pixel based on the first comprehensive feature vector and the second comprehensive feature vector to obtain a difference image; calculating an adaptive difference threshold based on the difference image, and determining the dynamic change range of each time point based on the feature difference measure of each pixel and the adaptive difference threshold.

[0016] Furthermore, identifying the temporary change range according to the dynamic change range includes: using a connected domain analysis method to divide the dynamic change range of each time sequence point into regions to obtain a set of sub-regions of the dynamic change range; calculating the frequency of each sub-region in the dynamic change range within a time window; constructing a temporary construction range feature library, and obtaining the probability that each sub-region in the dynamic change range is a temporary change region based on a support vector machine classifier; defining the temporary change range T of the tth time sequence point t ={r∈C t |f t (r)<τ f and p t (r)>τ p}, where f t (r) represents the dynamic change range C at the tth time point t The frequency of the rth sub-region in f and τ p Represent the frequency threshold and probability threshold respectively, p t (r) indicates C t The probability that the rth sub-region is a temporary change region.

[0017] Furthermore, the calculation formula for the temporary environmental impact index at the tth time point is:

[0018]

[0019] Wherein, EI(t) represents the temporary environmental impact index at the t-th time point, T t represents the temporary change range of the t-th time point, A(i) represents T t where ω1, ω2 and ω3 are the normalized area of the ith subregion, V(i) represents the change intensity of the ith subregion, D(i) represents the environmental sensitivity of the ith subregion, and ω1, ω2 and ω3 represent the weight coefficients.

[0020] Furthermore, the calculation formula for the permanent environmental impact index at the tth time point is:

[0021]

[0022] Wherein, EI′(t) represents the permanent environmental impact index at the t-th time point, P t represents the dynamic construction range of the t-th time point, A(j) represents P t where ω4, ω5, ω6 and ω7 are weight coefficients.

[0023] Furthermore, the calculation formula of the non-basic range index at the t-th time point is:

[0024] NBI(t)=γ1*|N t | / |B|+γ2*S(N t ,B)+γ3*Dist(N t ,B);

[0025] Wherein, NBI(t) represents the non-basic range index at the t-th time point, |N t | represents the non-foundation construction scope N at the tth time point t The total number of pixels in the foundation construction area B, |B| represents the total number of pixels in the foundation construction area B, S(N t ,B) represents the spatial dispersion index, Dist(N t ,B) represents the distance index, γ1, γ2 and γ3 represent the weight coefficients.

[0026] A system for extracting the construction scope of a water conservancy project based on remote sensing images, comprising:

[0027] The foundation scope determination module obtains pre-construction remote sensing images of the water conservancy hub area and determines the scope of foundation construction;

[0028] a variation range extraction module, which obtains the in-construction remote sensing images of the water conservancy hub area at each time point according to a preset period, performs feature extraction on the in-construction remote sensing images and the pre-construction remote sensing images, and determines the dynamic variation range of each time point;

[0029] a construction range extraction module, which identifies a temporary change range according to the dynamic change range, and defines a difference set between the temporary change range and the dynamic change range as a dynamic construction range;

[0030] The construction impact analysis module calculates a temporary environmental impact index based on the temporary change range, calculates a permanent environmental impact index based on the dynamic construction range, and calculates a construction environmental impact index sequence based on the temporary environmental impact index and the permanent environmental impact index at all time points; defines the difference between the dynamic construction range and the basic construction range at each time point as a non-basic construction range, and calculates a non-basic range index sequence based on the non-basic construction range at all time points;

[0031] The construction impact monitoring module monitors the construction environmental impact index sequence and the non-basic range index sequence in real time to carry out construction planning.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. By extracting features and performing difference analysis on time-series remote sensing images, the overall dynamic change range is first identified. Subsequently, a time-window-based frequency analysis is introduced. By calculating the frequency of each sub-region of the dynamic change range within the time window, combined with a support vector machine classifier, it effectively identifies areas of temporary change. This method not only considers the similarity of spatial features but also the temporal variation characteristics, accurately distinguishing between temporary facilities and permanent works. This precise identification and differentiation of temporary change areas effectively addresses the difficulty in defining construction scopes caused by the frequent changes in temporary facilities.

[0034] 2. Quantitatively assess the environmental impact of construction from both temporary and permanent perspectives. For temporary environmental impacts, three factors are comprehensively considered: the area of the affected area, the intensity of the change, and environmental sensitivity. For permanent environmental impacts, a cumulative impact factor is also introduced to assess long-term impacts. By combining the weighted temporary and permanent environmental impacts, a comprehensive assessment of the environmental impact of construction is achieved, making the environmental impact assessment more comprehensive and scientific. This provides quantifiable monitoring indicators for the dynamically changing construction scope, effectively supporting the management of the precisely defined construction scope.

[0035] 3. Construction activities that exceed the planned scope are identified by calculating the difference between the dynamic construction scope and the basic construction scope. Spatial dispersion and distance indicators are introduced as evaluation dimensions to calculate the non-basic scope index. This not only quantitatively assesses the degree of deviation from the construction scope but also reflects the spatial distribution characteristics of the deviation, providing more decision-making basis for construction management. By monitoring the dynamic changes in the construction scope in real time, construction activities that exceed the basic scope are promptly identified, effectively ensuring the precise definition and control of the construction scope. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a method for extracting the construction scope of a water conservancy project based on remote sensing images according to the present invention;

[0037] Figure 2 This is a data flow diagram of a method for extracting the construction scope of a water conservancy project based on remote sensing images according to the present invention;

[0038] Figure 3 The figure is a schematic structural diagram of a system for extracting the construction scope of a water conservancy project based on remote sensing images according to the present invention. DETAILED DESCRIPTION

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0040] See also Figures 1 to 3 The present invention provides a method and system for extracting the construction scope of a water conservancy project based on remote sensing images. The technical solution is as follows:

[0041] Example 1:

[0042] This embodiment provides a method for extracting the construction scope of a water conservancy project based on remote sensing images. Figure 1 and Figure 2 , showing the method flow and data flow, including:

[0043] Obtain pre-construction remote sensing images of the water conservancy hub area to determine the scope of foundation construction;

[0044] Acquire the in-construction remote sensing image of the water conservancy project area at each time point according to a preset period, perform feature extraction on the in-construction remote sensing image and the pre-construction remote sensing image, and determine the dynamic change range of each time point;

[0045] Identifying a temporary change range according to the dynamic change range, and defining a difference between the temporary change range and the dynamic change range as a dynamic construction range;

[0046] Calculate the temporary environmental impact index according to the temporary change range, calculate the permanent environmental impact index according to the dynamic construction range, and calculate the construction environmental impact index sequence according to the temporary environmental impact index and the permanent environmental impact index of all time points;

[0047] The difference between the dynamic construction range and the basic construction range at each time point is defined as the non-basic construction range, and a non-basic range index sequence is calculated based on the non-basic construction ranges at all time points;

[0048] The construction environment impact index sequence and the non-basic range index sequence are monitored in real time to carry out construction planning.

[0049] Specifically, the scope of foundation construction is determined based on the water conservancy project plan and combined with field investigations. Taking the construction of a hydropower station dam as an example, the construction area is located in a valley area and involves the main dam project, temporary roads, and material yards. The scope of foundation construction covers an area of approximately 2.5 square kilometers. The monitoring period is from March 202X to August 202X, with a preset cycle of one month. High-resolution remote sensing images are collected once a month for monitoring. The image collection time is from 10:00 am to 2:00 pm local time to ensure consistent lighting conditions. The image data comes from the domestically produced high-resolution satellite Gaofen-2, with a panchromatic band spatial resolution of 0.8 meters and a multispectral band resolution of 3.2 meters. A color composite image with a resolution of 1 meter is generated by fusing the panchromatic and multispectral bands. Data preprocessing includes radiation correction, geometric correction, and atmospheric correction.

[0050] Furthermore, determining the dynamic change range of each time point includes: extracting spectral feature vectors and texture feature vectors from the during-construction remote sensing image of each time point to obtain a first comprehensive feature vector; extracting pre-construction spectral feature vectors and pre-construction texture feature vectors from the pre-construction remote sensing image to obtain a second comprehensive feature vector; calculating the feature difference measure of each pixel based on the first comprehensive feature vector and the second comprehensive feature vector to obtain a difference image; calculating an adaptive difference threshold based on the difference image, and determining the dynamic change range of each time point based on the feature difference measure of each pixel and the adaptive difference threshold.

[0051] Specifically, the spectral feature vector and the texture feature vector are weightedly fused to obtain the first comprehensive feature vector, and the pre-construction spectral feature vector and the pre-construction texture feature vector are weightedly fused using the same weight distribution to obtain the second comprehensive feature vector; the first comprehensive feature vector is subtracted from the second comprehensive feature vector and then the Euclidean norm is calculated to obtain the feature difference measure. The adaptive difference threshold θ at the t-th time point t The calculation formula is θ t =μ t +κ*σ t , where μ t and σ t They are the difference images D t The mean and standard deviation of , κ is the adjustment coefficient; the dynamic range of the t-th time point C t Represented as C t ={(x,y)|D t (x,y)>θ t}, (x, y) is the pixel position in the remote sensing image during construction.

[0052] By extracting the spectral and texture features of remote sensing images before and after construction and combining them with adaptive difference thresholds, we can efficiently identify dynamically changing areas during the construction process, eliminating the limitations of traditional fixed threshold methods. This makes the extraction of dynamic change ranges more accurate and reliable in different time periods and complex environmental conditions, providing accurate basic data for subsequent temporary change range identification.

[0053] Further, identifying a temporary change range sequence according to the dynamic change range sequence includes:

[0054] Using a connected domain analysis method to divide the dynamic change range of each time sequence point into regions, and obtain a sub-region set of the dynamic change range;

[0055] Calculate the frequency of each subregion in the dynamic range within the time window:

[0056]

[0057] Among them, f t (r) represents the dynamic change range C at the tth time point t The frequency of the rth sub-region in the , w represents the time window size, min(·) represents the minimum function, max(·) represents the maximum function, C i Indicates the dynamic range of the i-th time window, I(r,C i ) represents the characteristic function, when C i When there is a similar subregion of the rth subregion in I(r,C i )=1, otherwise I(r,C i )=0; the similar sub-region of the rth sub-region is: i In, with C t The spatial overlap of the rth sub-region is greater than the spatial overlap threshold and the feature similarity is greater than the feature similarity threshold; the spatial overlap is the number of pixels with the same coordinates in the two sub-regions in C t The feature similarity is the cosine similarity of the comprehensive feature vectors of the two sub-regions. The calculation method of the sub-region comprehensive feature vector refers to the first comprehensive feature vector;

[0058] Construct a temporary construction range feature library and obtain the probability that each sub-area in the dynamic change range is a temporary change area based on the support vector machine classifier;

[0059] Define the temporary change range T of the tth time point t ={r∈C t |f t (r)<τ f and p t(r)>τ p}, where τ f and τ p Represent the frequency threshold and probability threshold respectively, p t (r) indicates C t The probability that the rth sub-region is a temporary change region.

[0060] Through the existence frequency analysis within the time window, the temporal characteristics of temporary facilities are effectively captured. Combined with the support vector machine classifier, a scientific temporary change discrimination model is established, which takes into account both spatial and temporal characteristics, significantly improving the accuracy of temporary change area identification, and helping to solve the problem of construction scope ambiguity caused by frequent changes of temporary facilities during the construction process.

[0061] Furthermore, the calculation formula for the temporary environmental impact index at the tth time point is:

[0062]

[0063] Wherein, EI(t) represents the temporary environmental impact index at the t-th time point, T t represents the temporary change range of the t-th time point, A(i) represents T t where ω1, ω2 and ω3 are the normalized area of the ith subregion, V(i) represents the change intensity of the ith subregion, D(i) represents the environmental sensitivity of the ith subregion, and ω1, ω2 and ω3 represent the weight coefficients.

[0064] A comprehensive temporary impact assessment system was established by comprehensively considering the normalized area, change intensity and spatial distribution weights; the quantitative calculation was performed using a weighted combination method, making the temporary environmental impact assessment more objective and facilitating construction management personnel to promptly identify and deal with environmental problems caused by temporary changes.

[0065] Furthermore, the calculation formula for the permanent environmental impact index at the tth time point is:

[0066]

[0067] Wherein, EI′(t) represents the permanent environmental impact index at the t-th time point, P t represents the dynamic construction range of the t-th time point, A(j) represents P t where ω4, ω5, ω6 and ω7 are weight coefficients.

[0068] P t The calculation formula for the cumulative impact factor of the jth sub-region is:

[0069]

[0070] Among them, α represents the cumulative effect coefficient, P k represents the dynamic construction range of the kth time point, I(r,P k ) represents the characteristic function, when P k When there is a similar subregion of the jth subregion in I(j,P k )=1, otherwise I(j,P k )=0; the similar sub-region of the jth sub-region is: k In, with P t The sub-regions whose spatial overlap of the j-th sub-region is greater than the spatial overlap threshold and whose feature similarity is greater than the feature similarity threshold.

[0071] The introduction of cumulative impact factors enables a dynamic assessment of permanent environmental impacts. By tracking and analyzing historical data, it accurately reflects the long-term cumulative effects of construction activities, providing a scientific basis for construction planning and environmental protection decision-making. Furthermore, a similar sub-region identification mechanism ensures the continuity and reliability of assessment results.

[0072] Furthermore, the change intensity of the sub-region is calculated by calculating the characteristic difference between the temporary change range or dynamic construction range and the remote sensing image before construction and normalizing it; the environmental sensitivity of the sub-region is calculated by determining the environmental elements in the sub-region based on the remote sensing image before construction, including water bodies, vegetation, and soil, etc., determining the sensitivity score of each type of environmental element and performing a weighted combination to obtain the environmental sensitivity; the sensitivity score of each type of environmental element is determined according to the distance between the environmental element and the center of the sub-region.

[0073] Table 1 shows the changing trends of the temporary environmental impact index, permanent environmental impact index and construction environmental impact index. The temporary environmental impact index fluctuates with the changes in temporary facilities, while the permanent environmental impact index rises steadily.

[0074] Table 1 Environmental impact index assessment results

[0075] Timing Point Temporary Environmental Impact Index Permanent Environmental Impact Index Construction Environmental Impact Index 202X-03 0.32 0.45 0.385 202X-04 0.38 0.48 0.430 202X-05 0.31 0.52 0.415 202X-06 0.35 0.54 0.445 202X-07 0.28 0.56 0.420 202X-08 0.34 0.58 0.460

[0076] Furthermore, the calculation formula of the non-basic range index at the t-th time point is:

[0077] NBI(t)=γ1*|N t | / |B|+γ2*S(N t ,B)+γ3*Dist(N t ,B);

[0078] Wherein, NBI(t) represents the non-basic range index at the t-th time point, |Nt | represents the non-foundation construction scope N at the tth time point t The total number of pixels in the construction area B, |B| represents the total number of pixels in the construction area B, S(N t ,B) represents the spatial dispersion index, Dist(N t ,B) represents the distance index, γ1, γ2 and γ3 represent the weight coefficients.

[0079] The calculation formula for spatial dispersion is:

[0080]

[0081] Among them, σ(N t ) represents the non-foundation construction scope N t σ(B) represents the standard deviation of all pixels within the foundation construction range B to its centroid.

[0082] The distance index calculation formula is:

[0083]

[0084] Among them, D max Indicates the preset maximum allowable distance, u and v represent the pixels in the non-foundation construction range and the foundation construction range respectively, min represents the minimum value, and ||uv||2 represents the Euclidean distance between two pixels.

[0085] Through the dual evaluation of spatial dispersion and distance indicators, accurate quantification of behaviors beyond the scope of foundation construction is achieved; combining pixel statistics and spatial analysis methods, it can not only evaluate the degree of deviation, but also reflect the spatial distribution characteristics of the deviation, providing a powerful tool for the refined management of the construction scope.

[0086] Table 2 reflects the situation beyond the scope of foundation construction, and the non-foundation scope index is all lower than 0.4.

[0087] Table 2 Monitoring results of non-foundation construction scope

[0088] Timing Point Non-basic area (hectares) Spatial dispersion Distance indicator Non-basic range index 202X-03 12.5 0.28 0.15 0.215 202X-04 15.8 0.32 0.18 0.245 202X-05 18.4 0.35 0.22 0.285 202X-06 20.2 0.38 0.25 0.315 202X-07 22.7 0.42 0.28 0.345 202X-08 25.3 0.45 0.32 0.385

[0089] This embodiment provides a method for extracting a dynamic construction range based on remote sensing images. First, through feature extraction and adaptive threshold methods, accurate identification of the dynamic change range of the construction area is achieved. Secondly, the temporary change range and dynamic construction range are identified based on the dynamic change range, and assessment systems for temporary environmental impact and permanent environmental impact are established respectively, making the assessment of the construction environmental impact more accurate and comprehensive. Thirdly, through the calculation of the non-basic range index, real-time monitoring of construction activities beyond the planned scope is achieved. This not only solves the problem of difficulty in defining the construction scope caused by frequent changes in temporary facilities, but also realizes the scientific assessment and effective management of the environmental impact of construction activities.

[0090] Example 2:

[0091] This embodiment provides a system for extracting the construction scope of a water conservancy project based on remote sensing images. Figure 3 Shown, including:

[0092] The foundation scope determination module obtains pre-construction remote sensing images of the water conservancy hub area and determines the scope of foundation construction;

[0093] a variation range extraction module, which obtains the in-construction remote sensing images of the water conservancy hub area at each time point according to a preset period, performs feature extraction on the in-construction remote sensing images and the pre-construction remote sensing images, and determines the dynamic variation range of each time point;

[0094] a construction range extraction module, which identifies a temporary change range according to the dynamic change range, and defines a difference set between the temporary change range and the dynamic change range as a dynamic construction range;

[0095] The construction impact analysis module calculates a temporary environmental impact index based on the temporary change range, calculates a permanent environmental impact index based on the dynamic construction range, and calculates a construction environmental impact index sequence based on the temporary environmental impact index and the permanent environmental impact index at all time points; defines the difference between the dynamic construction range and the basic construction range at each time point as a non-basic construction range, and calculates a non-basic range index sequence based on the non-basic construction range at all time points;

[0096] The construction impact monitoring module monitors the construction environmental impact index sequence and the non-basic range index sequence in real time to carry out construction planning.

[0097] Table 3 shows the changes in the dynamic change range, temporary change range, and dynamic construction range within the construction area over the six months. The temporary change range showed fluctuations, which was mainly due to the addition or reduction of temporary material yards and the construction or removal of temporary construction access roads.

[0098] Table 3 Statistics on changes in construction area

[0099] Timing Point Dynamic range (hectares) Temporary change area (hectares) Dynamic construction range (hectares) 202X-03 282.5 85.3 197.2 202X-04 295.8 98.7 197.1 202X-05 308.4 82.5 225.9 202X-06 315.2 90.8 224.4 202X-07 321.7 75.4 246.3 202X-08 328.3 88.6 239.7

[0100] Furthermore, determining the dynamic change range of each time point includes: extracting spectral feature vectors and texture feature vectors from the during-construction remote sensing image of each time point to obtain a first comprehensive feature vector; extracting pre-construction spectral feature vectors and pre-construction texture feature vectors from the pre-construction remote sensing image to obtain a second comprehensive feature vector; calculating the feature difference measure of each pixel based on the first comprehensive feature vector and the second comprehensive feature vector to obtain a difference image; calculating an adaptive difference threshold based on the difference image, and determining the dynamic change range of each time point based on the feature difference measure of each pixel and the adaptive difference threshold.

[0101] Furthermore, identifying the temporary change range according to the dynamic change range includes: using a connected domain analysis method to divide the dynamic change range of each time sequence point into regions to obtain a set of sub-regions of the dynamic change range; calculating the frequency of each sub-region in the dynamic change range within a time window; constructing a temporary construction range feature library, and obtaining the probability that each sub-region in the dynamic change range is a temporary change region based on a support vector machine classifier; defining the temporary change range T of the tth time sequence point t ={r∈C t |f t (r)<τ f and p t (r)>τ p}, where f t (r) represents the dynamic change range C at the tth time point t The frequency of the rth sub-region in f and τ p Represent the frequency threshold and probability threshold respectively, p t (r) indicates C t The probability that the rth sub-region is a temporary change region.

[0102] Furthermore, the calculation formula for the temporary environmental impact index at the tth time point is:

[0103]

[0104] Wherein, EI(t) represents the temporary environmental impact index at the t-th time point, T t represents the temporary change range of the t-th time point, A(i) represents T t where ω1, ω2 and ω3 are the normalized area of the ith subregion, V(i) represents the change intensity of the ith subregion, D(i) represents the environmental sensitivity of the ith subregion, and ω1, ω2 and ω3 represent the weight coefficients.

[0105] Furthermore, the calculation formula for the permanent environmental impact index at the tth time point is:

[0106]

[0107] Wherein, EI′(t) represents the permanent environmental impact index at the t-th time point, P t represents the dynamic construction range of the t-th time point, A(j) represents P t where ω4, ω5, ω6 and ω7 are weight coefficients.

[0108] Furthermore, the calculation formula of the non-basic range index at the t-th time point is:

[0109] NBI(t)=γ1*|N t | / |B|+γ2*S(N t ,B)+γ3*Dist(N t ,B);

[0110] Wherein, NBI(t) represents the non-basic range index at the t-th time point, |N t | represents the non-foundation construction scope N at the tth time point t The total number of pixels in the foundation construction area B, |B| represents the total number of pixels in the foundation construction area B, S(N t ,B) represents the spatial dispersion index, Dist(N t ,B) represents the distance index, γ1, γ2 and γ3 represent the weight coefficients.

[0111] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for extracting the construction scope of a water conservancy project based on remote sensing images, characterized in that: include: Obtain pre-construction remote sensing images of the water conservancy hub area to determine the scope of foundation construction; Acquire the in-construction remote sensing image of the water conservancy project area at each time point according to a preset period, perform feature extraction on the in-construction remote sensing image and the pre-construction remote sensing image, and determine the dynamic change range of each time point; Identifying a temporary change range according to the dynamic change range, and defining a difference between the temporary change range and the dynamic change range as a dynamic construction range; Calculate the temporary environmental impact index according to the temporary change range, calculate the permanent environmental impact index according to the dynamic construction range, and calculate the construction environmental impact index sequence according to the temporary environmental impact index and the permanent environmental impact index of all time points; The difference between the dynamic construction range and the basic construction range at each time point is defined as the non-basic construction range, and a non-basic range index sequence is calculated based on the non-basic construction ranges at all time points; The construction environment impact index sequence and the non-basic range index sequence are monitored in real time to carry out construction planning.

2. The method for extracting the construction scope of a water conservancy project based on remote sensing images according to claim 1 is characterized in that: Determining the dynamic change range of each time point includes: extracting the spectral feature vector and the texture feature vector of the during-construction remote sensing image of each time point to obtain a first comprehensive feature vector; extracting the pre-construction spectral feature vector and the pre-construction texture feature vector of the pre-construction remote sensing image to obtain a second comprehensive feature vector; calculating the feature difference measure of each pixel based on the first comprehensive feature vector and the second comprehensive feature vector to obtain a difference image; calculating an adaptive difference threshold based on the difference image, and determining the dynamic change range of each time point based on the feature difference measure of each pixel and the adaptive difference threshold.

3. The method for extracting the construction scope of a water conservancy project based on remote sensing images according to claim 1 is characterized in that: Identifying the temporary change range according to the dynamic change range includes: using a connected domain analysis method to divide the dynamic change range of each time sequence point into regions to obtain a set of sub-regions of the dynamic change range; calculating the existence frequency of each sub-region in the dynamic change range within a time window; constructing a temporary construction range feature library, and obtaining the probability that each sub-region in the dynamic change range is a temporary change region based on a support vector machine classifier; defining the temporary change range T of the t-th time sequence point t ={r∈C t |f t (r)<τ f and p t (r)>τ p }, where f t (r) represents the dynamic change range C at the tth time point t The frequency of the rth sub-region in f and τ p Represent the frequency threshold and probability threshold respectively, p t (r) indicates C t The probability that the rth sub-region is a temporary change region.

4. The method for extracting the construction scope of a water conservancy project based on remote sensing images according to claim 1 is characterized in that: The calculation formula for the temporary environmental impact index at the tth time point is: Wherein, EI(t) represents the temporary environmental impact index at the t-th time point, T t represents the temporary change range of the t-th time point, A(i) represents T t where ω1, ω2 and ω3 are the normalized area of the ith subregion, V(i) represents the change intensity of the ith subregion, D(i) represents the environmental sensitivity of the ith subregion, and ω1, ω2 and ω3 represent the weight coefficients.

5. The method for extracting the construction scope of a water conservancy project based on remote sensing images according to claim 1 is characterized in that: The calculation formula for the permanent environmental impact index at the tth time point is: Wherein, EI′(t) represents the permanent environmental impact index at the t-th time point, P t represents the dynamic construction range of the t-th time point, A(j) represents P t where ω4, ω5, ω6 and ω7 are weight coefficients.

6. The method for extracting the construction scope of a water conservancy project based on remote sensing images according to claim 1 is characterized in that: The calculation formula for the non-basic range index at the tth time point is: NBI(t)=γ1*|N t | / |B|+γ2*S(N t ,B)+γ3*Dist(N t ,B); Wherein, NBI(t) represents the non-basic range index at the t-th time point, |N t | represents the non-foundation construction scope N at the tth time point t The total number of pixels in the foundation construction area B, |B| represents the total number of pixels in the foundation construction area B, S(N t ,B) represents the spatial dispersion index, Dist(N t ,B) represents the distance index, γ1, γ2 and γ3 represent the weight coefficients.

7. A system for extracting the construction scope of a water conservancy project based on remote sensing images, characterized in that: include: The foundation scope determination module obtains pre-construction remote sensing images of the water conservancy hub area and determines the scope of foundation construction; a variation range extraction module, which obtains the in-construction remote sensing images of the water conservancy hub area at each time point according to a preset period, performs feature extraction on the in-construction remote sensing images and the pre-construction remote sensing images, and determines the dynamic variation range of each time point; a construction range extraction module, which identifies a temporary change range according to the dynamic change range, and defines a difference set between the temporary change range and the dynamic change range as a dynamic construction range; The construction impact analysis module calculates a temporary environmental impact index based on the temporary change range, calculates a permanent environmental impact index based on the dynamic construction range, and calculates a construction environmental impact index sequence based on the temporary environmental impact index and the permanent environmental impact index at all time points; defines the difference between the dynamic construction range and the basic construction range at each time point as a non-basic construction range, and calculates a non-basic range index sequence based on the non-basic construction range at all time points; The construction impact monitoring module monitors the construction environmental impact index sequence and the non-basic range index sequence in real time to carry out construction planning.

Images