Remote sensing monitoring analysis method and system for risk area

Abstract

The invention discloses a remote sensing monitoring analysis method and system for a risk area, and the method comprises the steps: obtaining a risk reference image set, an image collection time length and a preset collection mode, and then obtaining an area collection time length according to a risk grade reference image and the image collection time length; secondly, obtaining a mode image and synthesis time according to the risk level reference image, the area acquisition duration and a preset acquisition mode, and obtaining a mode risk area and a mode risk area level according to the risk level reference image and the mode image; and obtaining a risk level composite image and a risk alarm according to the mode risk area and the mode risk area level, and finally updating the risk reference image set according to the risk level composite image and the composite time. According to the invention, the multi-shooting mode, the differentiated region acquisition time and the adaptive reference image are adopted, the risk alarm is output, and the monitoring reliability, stability and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method and system for remote sensing monitoring and analysis of risk areas. Background Art

[0002] With the frequent occurrence of natural disasters, remote sensing monitoring and management of key risk areas have become particularly important. Currently, using InSAR satellite system data for ground displacement monitoring has become an efficient and reliable monitoring method. However, in practical applications, we face a series of challenges.

[0003] Firstly, due to different shooting modes of satellite remote sensing images, there are significant differences in their clarity. Existing remote sensing image monitoring methods do not fully combine the risk levels of risk areas, but use a unified shooting mode for data collection, ignoring the requirements of risk level determination for image clarity and monitoring accuracy, resulting in inaccurate determination of risk areas. Secondly, the time allocation of satellite shooting is also an important issue in current remote sensing monitoring. Existing remote sensing image monitoring methods do not reasonably allocate shooting time for different risk areas according to their risk levels, not only wasting valuable satellite resources, but also affecting the stability and accuracy of monitoring due to insufficient or excessive shooting frequencies. Finally, the determination of risk areas requires accurate reference images, and existing remote sensing image monitoring methods cannot reasonably update reference images for risk areas, resulting in insufficient reliability of monitoring. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a method and system for remote sensing monitoring and analysis of risk areas. The technical problems to be solved by the present invention are realized through the following technical solutions:

[0005] A method for remote sensing monitoring and analysis of risk areas, comprising:

[0006] Obtaining a risk reference image set, an image acquisition duration, and a preset acquisition mode;

[0007] Obtaining a regional acquisition duration according to the risk level reference image and the image acquisition duration;

[0008] Obtaining a mode image and a synthesis time according to the risk level reference image, the regional acquisition duration, and the preset acquisition mode;

[0009] Obtaining a mode risk area and a mode risk area level according to the risk level reference image and the mode image;

[0010] Obtaining a risk level synthesis image and a risk warning according to the mode risk area and the mode risk area level;

[0011] Update the risk reference image set according to the synthesized image and the synthesis time corresponding to the risk level;

[0012] Among them, the risk reference image set includes at least one frame of risk level reference image, the risk level reference image includes a risk area, a diffusion area and a full area, the area acquisition duration includes the full area acquisition duration, the risk sub - area acquisition duration and the diffusion sub - area acquisition duration, and the preset acquisition mode includes the SM acquisition mode, the IW acquisition mode and the EWS acquisition mode.

[0013] In a specific embodiment, obtaining the area acquisition duration according to the risk level reference image and the image acquisition duration includes:

[0014] Use labeled connectivity on the risk level reference image to obtain the number of risk sub - areas, the number of diffusion sub - areas, the area of the risk sub - areas, the area of the diffusion sub - areas, the area of the full area, and the risk sub - area level;

[0015] Obtain the corresponding risk area acquisition duration, diffusion area acquisition duration, and full area acquisition duration according to the area of the risk sub - areas, the area of the diffusion sub - areas, the area of the full area, and the image acquisition duration;

[0016] Obtain the risk sub - area acquisition duration according to the risk area acquisition duration, the number of risk sub - areas, the area of the risk sub - areas, and the risk sub - area level;

[0017] Obtain the diffusion sub - area acquisition duration according to the diffusion area acquisition duration, the number of diffusion sub - areas, and the area of the diffusion sub - areas.

[0018] In a specific embodiment, obtaining the mode image and the synthesis time according to the risk level reference image, the area acquisition duration, and the preset acquisition mode includes:

[0019] Obtain the SM mode sampling time of the risk sub - area according to the SM acquisition mode duration and the risk sub - area acquisition duration;

[0020] Obtain the IW mode sampling time of the diffusion sub - area according to the IW acquisition mode duration and the diffusion sub - area acquisition duration;

[0021] Obtain the EWS mode sampling time of the full area according to the EWS acquisition mode duration and the full area acquisition duration;

[0022] Use nearest - neighbor matching on the SM mode sampling time, the IW mode sampling time, and the EWS mode sampling time to obtain the synthesis time and the corresponding time offset;

[0023] Obtain the SM mode image of the risk sub-region based on the synthesis time, the SM time offset, and the SM acquisition mode;

[0024] Obtain the IW mode image of the diffusion sub-region based on the synthesis time, the IW time offset, and the IW acquisition mode;

[0025] Obtain the EWS mode image of the entire region based on the synthesis time, the EWS time offset, and the EWS acquisition mode.

[0026] In a specific embodiment, the obtaining the mode risk region and the mode risk region level based on the risk level reference image and the mode image includes:

[0027] Obtain the SM mode risk region and the SM mode risk region level based on the SM mode image and multiple risk level reference images;

[0028] Obtain the IW mode risk region and the IW mode risk region level based on the IW mode image and the risk level reference image with the closest time;

[0029] Obtain the EWS mode risk region and the EWS mode risk region level based on the EWS mode image and the risk level reference image with the closest time.

[0030] In a specific embodiment, the obtaining the SM mode risk region and the SM mode risk region level based on the SM mode image and multiple risk level reference images includes

[0031] Perform position matching on the SM mode image and multiple risk level reference images to obtain corresponding risk region reference images;

[0032] Perform pixel difference matching on the SM mode image and multiple risk region reference images to obtain the SM mode risk image;

[0033] Perform threshold truncation on the SM mode risk image to obtain the SM mode risk region;

[0034] Perform median filtering on the SM mode risk region to obtain the SM mode risk region level.

[0035] In a specific embodiment, the obtaining the IW mode risk region and the IW mode risk region level based on the IW mode image and the risk level reference image with the closest time includes

[0036] Perform position matching on the IW mode image and the risk level reference image with the closest time to obtain the diffusion region reference image;

[0037] Performing differential filtering on the IW mode image and the diffusion area reference image to obtain an IW mode difference image;

[0038] Using an interpretation model for the IW mode image to obtain IW mode type weights;

[0039] Performing distance calculation on the IW mode image to obtain IW mode distance weights;

[0040] Obtaining an IW mode risk image based on the IW mode difference image, the IW mode type weights, and the IW mode distance weights;

[0041] Performing threshold truncation on the IW mode risk image to obtain an IW mode risk area;

[0042] Performing extreme value filtering on the IW mode risk area to obtain the IW mode risk area level.

[0043] In a specific embodiment, the obtaining of the EWS mode risk area and the EWS mode risk area level according to the EWS mode image and the risk level reference image with the closest time includes

[0044] Performing differential filtering on the EWS mode image and the risk level reference image with the closest time to obtain an EWS mode difference image;

[0045] Using an interpretation model for the EWS mode image to obtain EWS mode type weights;

[0046] Obtaining an EWS mode marked image based on the EWS mode difference image and the EWS mode type weights;

[0047] Setting a preset EWS mode risk area level for the EWS mode risk area.

[0048] In a specific embodiment, the obtaining of the risk level composite image and the risk warning according to the mode risk area and the mode risk area level includes

[0049] Obtaining a fusion risk area, a fusion risk area level, and a risk warning according to the SM mode risk area, the IW mode risk area, the EWS mode risk area, the SM mode risk area level, the IW mode risk area level, and the EWS mode risk area level;

[0050] Obtaining a fusion diffusion area according to the fusion risk area and the fusion risk area level;

[0051] Obtaining a risk level composite image according to the fusion risk area, the fusion risk area level, and the fusion diffusion area.

[0052] In a specific embodiment, updating the risk reference image set according to the risk-level composite image and the synthesis time includes:

[0053] Performing dual-weight calculation on the risk-level composite image and the risk-level reference images in the risk reference image set to obtain a difference data set;

[0054] Performing threshold determination on the difference data set to obtain an update strategy;

[0055] Using the update strategy for the risk-level composite image and the risk reference image set to obtain the updated risk reference image set.

[0056] In a specific embodiment, a remote sensing monitoring and analysis system for a risk area includes:

[0057] A data acquisition unit, configured to obtain a risk reference image set, an image acquisition duration, and a preset acquisition mode;

[0058] A duration analysis unit, configured to obtain a regional acquisition duration according to the risk-level reference image and the image acquisition duration;

[0059] An image acquisition unit, configured to obtain a mode image and a synthesis time according to the risk-level reference image, the regional acquisition duration, and the preset acquisition mode;

[0060] A risk calculation unit, configured to obtain a mode risk area and a mode risk area level according to the risk-level reference image and the mode image;

[0061] A synthesis and alarm unit, configured to obtain a risk-level composite image and a risk alarm according to the mode risk area and the mode risk area level;

[0062] A reference update unit, configured to update the risk reference image set according to the risk-level composite image and the synthesis time;

[0063] Wherein, the risk reference image set includes at least one frame of risk-level reference image, the risk-level reference image includes a risk area, a diffusion area, and a full area, the regional acquisition duration includes a full-area acquisition duration, a risk sub-area acquisition duration, and a diffusion sub-area acquisition duration, and the preset acquisition mode includes an SM acquisition mode, an IW acquisition mode, and an EWS acquisition mode.

[0064] Advantages of the present invention:

[0065] A remote sensing monitoring and analysis method and system for risk areas of the present invention first adopts a multi - shooting mode to differentially collect different areas, then allocates differential collection times to different areas according to the satellite shooting time, and then adaptively updates the reference image, and finally provides a risk level image and a risk warning, improving the reliability, stability and accuracy of monitoring.

[0066] The following will further elaborate on the present invention in conjunction with the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 is a flowchart of a remote sensing monitoring and analysis method for risk areas provided by an embodiment of the present invention;

[0068] Figure 2 is a schematic diagram of the synthesis time of a remote sensing monitoring and analysis method for risk areas provided by an embodiment of the present invention;

[0069] Figure 3 is a schematic diagram of the fused risk area of a remote sensing monitoring and analysis method for risk areas provided by an embodiment of the present invention;

[0070] Figure 4 is a block diagram of a module of a remote sensing monitoring and analysis method for risk areas provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] The following further describes the present invention in detail with specific embodiments, but the implementation manners of the present invention are not limited thereto.

[0072] Embodiment 1

[0073] In a specific implementation manner, please refer to Figure 1 , Figure 1 which is a flowchart of a remote sensing monitoring and analysis method for risk areas, and the specific steps are as follows:

[0074] S1: To start the remote sensing monitoring management process, obtain a risk reference image set, an image acquisition duration, and a preset acquisition mode. The risk reference image set includes at least one frame of risk level reference images, which not only clearly depict the risk area and the diffusion area corresponding to the risk area, but also depict the specific image content. The image acquisition duration is an important reference for subsequent steps.

[0075] S2: To more reasonably allocate shooting resources, ensure that high - risk areas can be more intensively and accurately monitored, thereby improving the monitoring efficiency and accuracy, obtain the regional acquisition duration according to the risk level reference image and the image acquisition duration.

[0076] S21: Apply labeled connectivity to the risk level reference image to obtain the number of risk sub-regions, the number of diffusion sub-regions, the area of risk sub-regions, the area of diffusion sub-regions, the area of the entire region, and the risk sub-region level. In a specific embodiment, the calculation steps are as follows:

[0077] 1. Obtain the latest risk level reference image: From the set of risk reference images, in this embodiment, a risk level reference image is selected as the basis for analysis according to the principle of the latest time;

[0078] 2. Determine the number of risk sub-regions: Apply the eight-point region connectivity algorithm to process the risk level reference image to identify and count the number of risk sub-regions;

[0079] 3. Calculate the area of risk sub-regions: Traverse all risk sub-regions. Using the method of coordinate pixel determination, in this embodiment, the area of each risk sub-region can be obtained. Here, the area is measured by the number of pixels, which reflects the occupation degree of the risk sub-region in the image;

[0080] 4. Determine the number of diffusion sub-regions: Similar to the identification of risk sub-regions, to identify relatively scattered or irregular regions in the image, in this embodiment, the four-point region connectivity algorithm is used to identify the number of diffusion sub-regions;

[0081] 5. Calculate the area of diffusion sub-regions: Traverse all diffusion sub-regions and calculate the area of each diffusion sub-region using the method of coordinate pixel determination;

[0082] 6. Determine the area of the entire region: The area of the risk level reference image is the area of the entire region, which reflects the scope of the entire monitoring region.

[0083] S22: Obtain the corresponding acquisition duration of the risk region, the acquisition duration of the diffusion region, and the acquisition duration of the entire region according to the area of the risk sub-region, the area of the diffusion sub-region, the area of the entire region, and the image acquisition duration. In a specific embodiment, the calculation steps are as follows:

[0084] 1. Calculate the acquisition duration of the risk region: Use the proportion formula to calculate the acquisition duration of the risk region according to the proportion of the area of the risk sub-region in the area of the entire region and the image acquisition duration;

[0085] 2. Calculate the acquisition duration of the diffusion region: Use the proportion formula to calculate the acquisition duration of the diffusion region according to the proportion of the area of the diffusion sub-region in the area of the entire region and the image acquisition duration;

[0086] 3. Determine the acquisition duration of the entire region: After allocating the acquisition durations of the risk region and the diffusion region, we can obtain the acquisition duration of the entire region by subtracting the acquisition durations of these two regions from the image acquisition duration;

[0087] The proportion formula is as follows:

[0088]

[0089] Where time_sub and time_all are the acquisition durations of the region and the image respectively, Ssub and Sall are the areas of the sub-region and the entire region respectively, i is the sub-region index, and N is the number of sub-regions.

[0090] S23: Obtain the acquisition duration of the risk sub-region based on the acquisition duration of the risk region, the number of risk sub-regions, the area of the risk sub-region, and the risk sub-region level. In a specific embodiment, the calculation formula for the acquisition duration of the risk sub-region is:

[0091]

[0092] Where time_dg and time_subdg are the acquisition durations of the risk region and the risk sub-region respectively, Sdg is the area of the risk sub-region, Ldg is the risk sub-region level weight, i is the risk sub-region index, and N is the number of risk sub-regions.

[0093] S24: Obtain the acquisition duration of the diffusion sub-region based on the acquisition duration of the diffusion region, the number of diffusion sub-regions, and the area of the diffusion sub-region. In a specific embodiment, the calculation formula for the acquisition duration of the diffusion sub-region is:

[0094]

[0095] Where time_sp and time_subsp are the acquisition durations of the diffusion region and the diffusion sub-region respectively, Ssp is the area of the diffusion sub-region, i is the diffusion sub-region index, and N is the number of diffusion sub-regions.

[0096] S3: Since the selection of the shooting mode will directly affect the clarity of the image and the monitoring accuracy, and the synthesis time determines when the final monitoring result can be obtained in this embodiment, the mode image and the synthesis time are obtained according to the risk level reference image, the acquisition duration of the region, and the preset acquisition mode.

[0097] S31: Obtain the SM mode sampling time of the risk sub-region based on the acquisition duration of the SM acquisition mode and the acquisition duration of the risk sub-region. Among them, the SM mode usually has a relatively high resolution and is suitable for scenarios that require high-resolution ground coverage. In a specific embodiment, the 24 hours are equally divided using the average distribution principle to obtain multiple SM mode sampling times with the same time interval.

[0098] S32: Obtain the IW mode sampling time of the diffusion sub-region based on the IW acquisition mode duration and the diffusion sub-region acquisition duration. Among them, although the IW mode is widely used in the detailed monitoring of landforms and terrains, its resolution is slightly lower than that of the SM mode. In a specific implementation, the central concentration principle is adopted to perform continuous image acquisition within a preset time period.

[0099] S33: Obtain the EWS mode sampling time of the entire region based on the EWS acquisition mode duration and the entire region acquisition duration. Among them, the EWS mode provides a very large area coverage at the cost of sacrificing spatial resolution and is suitable for scenarios that require rapid large-area coverage but have low requirements for resolution. In a specific implementation, the principle of average distribution of brightness differences is adopted. The sampling time during the day is half an hour, that is, 1 frame of image is acquired every half an hour, and the sampling time at night is 10 minutes, that is, 1 frame of image is acquired every 10 minutes.

[0100] S34: Adopt proximity matching for the SM mode sampling time, the IW mode sampling time, and the EWS mode sampling time to obtain the synthesis time and the corresponding time offset. Please refer to Figure 2 , Figure 2 which is a schematic diagram of the synthesis time of a remote sensing monitoring and analysis method for risk regions. In a specific implementation, the calculation steps are as follows:

[0101] 1. Adopt time proximity for multiple SM mode sampling times of each risk sub-region to obtain the SM synthesis time and the first SM time offset;

[0102] 2. Adopt time proximity for multiple IW mode sampling times of each diffusion sub-region to obtain the IW synthesis time and the first IW time offset;

[0103] 3. Adopt time proximity for the SM synthesis time, the IW synthesis time, and the EWS mode sampling time to obtain the synthesis time, the second SM time offset, and the second IW time offset;

[0104] 4. Obtain the SM time offset based on the first SM time offset and the second SM time offset;

[0105] 5. Obtain the IW time offset based on the first IW time offset and the second IW time offset.

[0106] S35: Obtain the SM mode image of the risk sub-region based on the synthesis time, the SM time offset, and the SM acquisition mode.

[0107] S36: Obtain the IW mode image of the diffusion sub-region based on the synthesis time, the IW time offset, and the IW acquisition mode.

[0108] S37: Obtain the EWS mode image of the entire region based on the synthesis time, the EWS time offset, and the EWS acquisition mode.

[0109] S4: By comparing the risk level reference image and the captured mode image, in this embodiment, the mode risk region can be identified and its risk level can be evaluated. This step is directly related to the accuracy of subsequent risk assessment and alarm. Through precise risk region identification and evaluation, this embodiment can more accurately grasp the risk situation and provide strong support for coping with potential natural disasters. Therefore, obtain the mode risk region and the mode risk region level based on the risk level reference image and the mode image.

[0110] S41: Obtain the SM mode risk region and the SM mode risk region level based on the SM mode image and multiple risk level reference images.

[0111] S411: To ensure that this embodiment can accurately correspond the risk region in the SM mode image with the risk level in the risk level reference image, perform position matching on the SM mode image and multiple risk level reference images to obtain the corresponding risk region reference image.

[0112] S412: To identify the risk region in the SM mode image by comparing the pixel differences between two images, perform pixel difference matching on the SM mode image and multiple risk region reference images to obtain the SM mode risk image.

[0113] S413: To clearly mark the risk region in the image, perform threshold truncation on the SM mode risk image to obtain the SM mode risk region.

[0114] S414: To remove the isolated points in the image while retaining the edge information of the image, so as to obtain a more accurate risk level evaluation result, perform median filtering on the SM mode risk region to obtain the SM mode risk region level.

[0115] S42: Obtain the IW mode risk region and the IW mode risk region level based on the IW mode image and the risk level reference image with the closest time.

[0116] S421: To identify the risk diffusion region in the IW mode image, perform position matching on the IW mode image and the risk level reference image with the closest time to obtain the diffusion region reference image.

[0117] S422: By comparing the differences between two images to identify the risk change region in the IW mode image, perform difference filtering on the IW mode image and the diffusion region reference image to obtain the IW mode difference image.

[0118] S423: Apply the interpretation model to the IW mode image to obtain the IW mode type weight. The interpretation model is an algorithm that can identify different risk types based on image features and can provide important reference information for subsequent risk level assessment.

[0119] S424: Apply distance calculation to the IW mode image to obtain the IW mode distance weight.

[0120] S425: In order to comprehensively consider multiple factors to obtain a more accurate risk area identification result, obtain the IW mode risk image based on the IW mode difference image, the IW mode type weight, and the IW mode distance weight.

[0121] S426: In order to clearly mark the risk area in the image, apply threshold truncation to the IW mode risk image to obtain the IW mode risk area.

[0122] S427: In order to further reflect the risk, apply extreme value filtering to the IW mode risk area to obtain the IW mode risk area level.

[0123] S43: Obtain the EWS mode risk area and the EWS mode risk area level based on the EWS mode image and the risk level reference image with the closest time.

[0124] S431: In order to identify the risk change area in the EWS mode image, apply difference filtering to the EWS mode image and the risk level reference image with the closest time to obtain the EWS mode difference image.

[0125] S432: In order to identify different risk types based on image features, apply the interpretation model to the EWS mode image to obtain the EWS mode type weight.

[0126] S433: Obtain the EWS mode marked image based on the EWS mode difference image and the EWS mode type weight.

[0127] S434: Set the preset EWS mode risk area level for the EWS mode risk area according to experience or relevant regulations.

[0128] S5: Based on the mode risk area and the risk level, in this embodiment, a risk level composite image is generated, and an alarm is issued according to the risk situation. The output of this step is the core of the monitoring and management process, providing an intuitive and comprehensive display of the risk situation for monitoring, helping relevant departments take timely measures to reduce the losses caused by natural disasters. Therefore, obtain the risk level composite image and the risk alarm based on the mode risk area and the mode risk area level.

[0129] S51: Obtain the fused risk region, the fused risk region level, and the risk warning based on the SM mode risk region, the IW mode risk region, the EWS mode risk region, the SM mode risk region level, the IW mode risk region level, and the EWS mode risk region level. In a specific embodiment, the risk warning calculation steps are as follows:

[0130] 1. Determination of the risk level index: To ensure that the most severe risk situation is concerned in this embodiment, compare the SM mode risk region level, the IW mode risk region level, and the EWS mode risk region level, and find the maximum risk region level among them as the risk level index;

[0131] 2. Determination of the warning level: To quickly determine the warning level corresponding to the current risk situation, look up the corresponding warning level in the risk level list according to the risk level index, where the risk level list is a preset table constructed based on the correspondence between the risk level and the warning level;

[0132] 3. Issuance of the risk warning indication: To ensure that relevant departments can receive risk information in a timely manner and take corresponding countermeasures, issue the corresponding risk warning indication according to the determined warning level;

[0133] In a specific embodiment, the risk level calculation formula for each pixel in the fused risk region is:

[0134]

[0135] where \(v_{dg}\), \(v_{sp}\), and \(v_{all}\) are the pixel risk levels of the SM mode risk region, the IW mode risk region, and the EWS mode risk region respectively, \(w_{dg}\), \(w_{sp}\), and \(w_{all}\) are the pixel weights of the SM mode risk region, the IW mode risk region, and the EWS mode risk region respectively, and \(dg\) is the risk level of the pixel in the fused risk region;

[0136] In a specific embodiment, please refer to Figure 3 , Figure 3 is a schematic diagram of the fused risk region of a remote sensing monitoring and analysis method for risk regions. If \(dg\) is not 0, then the coordinate position where \(dg\) is located is the fused risk region.

[0137] S52: Obtain the fused diffusion region based on the fused risk region and the fused risk region level. In a specific embodiment, the calculation steps are as follows:

[0138] 1. Setting of dilation radius and erosion radius: To ensure that this embodiment can accurately simulate the spread of risks, corresponding dilation radius and erosion radius are set according to the level of the fused risk area;

[0139] 2. Calculation of dilation area: To simulate the process of risk spreading outward, this embodiment adopts a dilation filtering algorithm to calculate the dilation area according to the fused risk area and the corresponding dilation radius;

[0140] 3. Optimization of dilation filtering area: To further optimize the dilation area to make it more conform to the actual risk spreading situation, this embodiment adopts a neighborhood filtering algorithm for the dilation area to obtain the dilation filtering area;

[0141] 4. Calculation of fused diffusion area: To simulate the attenuation situation during the risk diffusion process, this embodiment adopts an erosion filtering algorithm to calculate the fused diffusion area according to the dilation filtering area and the corresponding erosion radius.

[0142] S53: Obtain a risk level composite image based on the fused risk area, the fused risk area level, and the fused diffusion area. In a specific embodiment, calculate using a fusion formula according to the fused risk area level, the fused risk area, and the fused diffusion area. The fusion formula is:

[0143]

[0144] where max is the maximum pixel value of the risk level composite image, wdg and wsp are the risk weight and the diffusion weight respectively, dgk is the risk level of the pixels in the risk area, sp is the identifier of the diffusion area, and i and j are the coordinates of the risk level composite image.

[0145] S6: Update the risk reference image set according to the risk level composite image and the synthesis time. The completion of this step marks the end of a cycle of the monitoring and management process, and at the same time provides more accurate and rich data support for the start of the next cycle. Through continuous optimization, the monitoring method of this embodiment can continuously adapt to the changing risk environment and improve the accuracy and reliability of monitoring.

[0146] S61: Obtain a difference data set by performing double-weight calculation on the risk level composite image and the risk level reference images in the risk reference image set. In a specific embodiment, the calculation steps are as follows:

[0147] 1. Calculation of time difference weight: To consider the influence of time factors on image differences and ensure a more reasonable update process, calculate the time difference weight corresponding to each risk level reference image according to the synthesis time of the risk level composite image and the acquisition time of each image in the risk reference image set;

[0148] 2. Calculation of the average risk level difference: To quantify the degree of difference between images and provide data support for the construction of subsequent difference datasets, each image in the risk level composite image is compared one by one with the images in the risk reference image set, and the average value of the frame-level risk level differences between them is calculated.

[0149] 3. Calculation of difference data: To comprehensively consider the time factor and the risk level factor, combining the time difference weight and the average value of the frame-level risk level differences, the difference data for each risk level reference image is calculated.

[0150] 4. Construction of the difference dataset: To provide data support for subsequent difference analysis and the formulation of update strategies, the difference data of all risk level reference images is statistically analyzed to construct a difference dataset.

[0151] S62: An update strategy is obtained by using threshold determination for the difference dataset. In a specific embodiment, the calculation steps are as follows:

[0152] 1. Determination of the maximum difference data: To determine the image in the current risk reference image set with the largest difference from the risk level composite image, the maximum value principle is used to find the maximum difference data and the corresponding risk level reference image in the difference dataset.

[0153] 2. Formulation of the update strategy: The maximum difference data is compared with a preset threshold. If the maximum difference data is greater than the preset threshold, it indicates that the difference between the current risk reference image set and the risk level composite image is relatively large, and an insert update strategy needs to be used for updating; if the maximum difference data is not greater than the preset threshold, it indicates that the difference between the current risk reference image set and the risk level composite image is relatively small, and a self-update strategy can be used for fine-tuning.

[0154] S63: The updated risk reference image set is obtained by using the update strategy for the risk level composite image and the risk reference image set. In a specific embodiment, the calculation steps are as follows:

[0155] 1. Implementation of the insert update strategy: If the update strategy is the insert update strategy, to ensure that the risk reference image set can accurately reflect the current risk environment and improve the accuracy of monitoring, in this embodiment, the risk level composite image is inserted into the risk reference image set and replaces the risk level reference image corresponding to the maximum difference data.

[0156] 2. Implementation of the self-update strategy: If the update strategy is the self-update strategy, in order to fine-tune the images with relatively small differences from the risk-level composite images in the risk reference image set while maintaining the stability of the risk reference image set, so as to improve the reliability of monitoring, this embodiment will sort the difference data set and sort the corresponding risk-level composite images in ascending order of the difference data.

[0157] In a specific embodiment, please refer to Figure 4 , Figure 4 which is a module block diagram of a method for analyzing the regional spatio-temporal coverage rate of remote sensing images. A remote sensing monitoring and analysis system for risk areas is characterized by including:

[0158] A data acquisition unit for acquiring a risk reference image set and the image acquisition duration;

[0159] A duration analysis unit for obtaining the regional acquisition duration according to the risk-level reference image and the image acquisition duration;

[0160] An image acquisition unit for obtaining a pattern image and a synthesis time according to the risk-level reference image and the regional acquisition duration;

[0161] A risk calculation unit for obtaining a pattern risk area and a pattern risk area level according to the regional risk-level reference image and the pattern image;

[0162] A synthesis alarm unit for obtaining a risk-level composite image and a risk alarm according to the pattern risk area and the pattern risk area level;

[0163] A reference update unit for updating the risk reference image set according to the risk-level composite image and the synthesis time;

[0164] Among them, the risk reference image set includes at least one frame of risk-level reference image, and the risk-level reference image includes a risk area and a diffusion area.

[0165] A method and system for remote sensing monitoring and analysis of a risk area in this embodiment first adopts a multi-shooting mode to perform differential acquisition on different regions, then allocates differential acquisition times to different regions according to the satellite shooting time, and then adaptively updates the reference image, and finally provides a risk-level image and a risk alarm, improving the reliability, stability and accuracy of monitoring.

[0166] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0167] Although the present application has been described in conjunction with various embodiments, however, in the process of implementing the claimed present application, those skilled in the art can understand and achieve other variations of the disclosed embodiments by viewing the accompanying drawings, the disclosure content, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality of cases.

[0168] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A remote sensing monitoring and analysis method for risk areas, characterized in that: include: Obtain risk reference image set, image acquisition duration and preset acquisition mode; Obtaining a region acquisition duration according to the risk level reference image and the image acquisition duration; Obtaining a pattern image and a synthesis time according to the risk level reference image, the area acquisition time and a preset acquisition mode; Obtaining a model risk area and a model risk area level according to the risk level reference image and the model image; Obtaining a risk level composite image and a risk warning according to the model risk area and the model risk area level; updating the risk reference image set according to the risk level synthesized image and the synthesis time; Among them, the risk reference image set includes at least one frame of risk level reference image, the risk level reference image includes risk area, diffusion area and full area, the area acquisition time includes full area acquisition time, risk sub-area acquisition time and diffusion sub-area acquisition time, and the preset acquisition mode includes SM acquisition mode, IW acquisition mode and EWS acquisition mode.

2. The remote sensing monitoring and analysis method for risk areas according to claim 1 is characterized in that: The obtaining the area acquisition duration according to the risk level reference image and the image acquisition duration includes: Using marked connectivity on the risk level reference image, the number of risk sub-regions, the number of diffusion sub-regions, the area of ​​risk sub-regions, the area of ​​diffusion sub-regions, the area of ​​the entire region and the level of risk sub-regions are obtained; Obtaining corresponding risk area acquisition time, diffusion area acquisition time, and full area acquisition time according to the risk sub-area area, the diffusion sub-area area, the full area area, and the image acquisition time; Obtaining the risk sub-region collection time according to the risk region collection time, the number of risk sub-regions, the area of ​​the risk sub-region and the risk sub-region level; The diffusion sub-region collection time length is obtained according to the diffusion region collection time length, the number of the diffusion sub-regions and the area of ​​the diffusion sub-regions.

3. The remote sensing monitoring and analysis method for risk areas according to claim 2 is characterized in that: The obtaining of the mode image and the synthesis time according to the risk level reference image, the area acquisition duration and the preset acquisition mode includes: Obtaining the SM mode sampling time of the risk sub-area according to the SM acquisition mode duration and the risk sub-area acquisition duration; Obtaining an IW mode sampling time of a diffusion sub-region according to the IW acquisition mode duration and the diffusion sub-region acquisition duration; Obtaining the EWS mode sampling time of the entire area according to the EWS acquisition mode duration and the entire area acquisition duration; Adopting proximity matching to obtain the composite time and the corresponding time offset for the SM mode sampling time, the IW mode sampling time and the EWS mode sampling time; Obtaining an SM mode image of the risk sub-area according to the synthesis time, the SM time offset and the SM acquisition mode; Obtaining an IW mode image of the diffusion sub-region according to the synthesis time, the IW time offset and the IW acquisition mode; The EWS mode image of the entire area is obtained according to the synthesis time, the EWS time offset and the EWS acquisition mode.

4. The remote sensing monitoring and analysis method for risk areas according to claim 3 is characterized in that: The obtaining the model risk area and the model risk area level according to the risk level reference image and the model image includes: Obtaining an SM mode risk area and an SM mode risk area level according to the SM mode image and a plurality of risk level reference images; Obtaining an IW mode risk area and an IW mode risk area level according to the IW mode image and the risk level reference image that is closest in time; The EWS mode risk area and the EWS mode risk area level are obtained according to the EWS mode image and the risk level reference image that is closest in time.

5. The remote sensing monitoring and analysis method for risk areas according to claim 4 is characterized in that: The step of obtaining the SM mode risk area and the SM mode risk area level according to the SM mode image and the plurality of risk level reference images includes: Using position matching on the SM mode image and the plurality of risk level reference images to obtain a corresponding risk area reference image; Using pixel difference matching on the SM mode image and a plurality of risk area reference images to obtain an SM mode risk image; Using a threshold to truncate the SM mode risk image to obtain a SM mode risk area; The SM mode risk area is subjected to median filtering to obtain the SM mode risk area level.

6. The remote sensing monitoring and analysis method for risk areas according to claim 4 is characterized in that: The IW mode risk area and the IW mode risk area level are obtained according to the IW mode image and the risk level reference image closest in time, including Using position matching between the IW mode image and the risk level reference image that is closest in time to obtain a diffusion area reference image; Applying difference filtering to the IW mode image and the diffusion area reference image to obtain an IW mode difference image; Using an interpretation model on the IW mode image to obtain an IW mode type weight; Using distance calculation on the IW mode image to obtain an IW mode distance weight; Obtaining an IW mode risk image according to the IW mode difference image, the IW mode type weight, and the IW mode distance weight; Using a threshold to truncate the IW mode risk image to obtain an IW mode risk area; The IW mode risk area is subjected to extreme value filtering to obtain the IW mode risk area level.

7. The remote sensing monitoring and analysis method for risk areas according to claim 4 is characterized in that: The EWS mode risk area and the EWS mode risk area level are obtained according to the EWS mode image and the risk level reference image closest in time, including Applying difference filtering to the EWS mode image and the risk level reference image closest in time to obtain an EWS mode difference image; Using an interpretation model on the EWS mode image to obtain an EWS mode type weight; Obtaining an EWS mode label image according to the EWS mode difference image and the EWS mode type weight; A preset EWS mode risk area level is set for the EWS mode risk area.

8. The remote sensing monitoring and analysis method for risk areas according to claim 4 is characterized in that: The risk level composite image and risk warning are obtained according to the model risk area and the model risk area level, including Obtaining a fused risk area, a fused risk area level and a risk warning according to the SM mode risk area, the IW mode risk area, the EWS mode risk area, the SM mode risk area level, the IW mode risk area level and the EWS mode risk area level; Obtaining a fusion diffusion area according to the fusion risk area and the fusion risk area level; A risk level composite image is obtained according to the fusion risk area, the fusion risk area level and the fusion diffusion area.

9. The remote sensing monitoring and analysis method for risk areas according to claim 1, characterized in that: The updating of the risk reference image set according to the risk level synthesized image and the synthesis time includes: A difference data set is obtained by performing a double weight calculation on the risk level synthetic image and the risk level reference image in the risk reference image set; Using a threshold value to determine the difference data set to obtain an update strategy; The update strategy is applied to the risk level composite image and the risk reference image set to obtain an updated risk reference image set.

10. A remote sensing monitoring and analysis system for risk areas, characterized in that: include: A data acquisition unit, used to obtain a risk reference image set, image acquisition duration, and a preset acquisition mode; A duration analysis unit, configured to obtain a region acquisition duration according to the risk level reference image and the image acquisition duration; An image acquisition unit, configured to obtain a pattern image and a synthesis time according to the risk level reference image, the area acquisition time and a preset acquisition mode; a risk calculation unit, configured to obtain a pattern risk area and a pattern risk area level according to the risk level reference image and the pattern image; A synthetic warning unit, which obtains a risk level synthetic image and a risk warning according to the model risk area and the model risk area level; a reference updating unit, configured to update the risk reference image set according to the risk level composite image and the composite time; Among them, the risk reference image set includes at least one frame of risk level reference image, the risk level reference image includes risk area, diffusion area and full area, the area acquisition time includes full area acquisition time, risk sub-area acquisition time and diffusion sub-area acquisition time, and the preset acquisition mode includes SM acquisition mode, IW acquisition mode and EWS acquisition mode.

Images