Disaster analysis method and device based on image data

Abstract

The present invention discloses a disaster analysis method and device based on image data, which belongs to the technical field of image data processing. The method includes: obtaining first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated; obtaining second image information collected by a second remote sensing platform according to a preset second image information acquisition strategy and the disaster position information to be calibrated; analyzing the first image information and the second image information according to a preset analysis strategy to obtain calibrated disaster position information and disaster extent information. The method and device provided by the present invention can quickly obtain disaster position information and disaster extent information based on analysis of image information collected by multi-source remote sensing platforms, thereby helping rescue personnel to formulate rescue strategies in a targeted manner. This method has the advantages of fast speed, high recognition accuracy, and significant robustness.

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and in particular to a disaster analysis method and device based on image data. Background Art

[0002] With the advancement and development of science and technology, drones have been used in many aspects of life, especially in photography, where they have unique advantages and have played a major role in disaster rescue. However, relying solely on drone photography cannot provide a more accurate assessment of disasters. Summary of the invention

[0003] The purpose of the present invention is to provide a disaster analysis method and device based on image data, in order to solve the problem of single data and small information dimension caused by using drone photography alone in the prior art.

[0004] To solve the above problems, the first aspect of the present invention provides a disaster analysis method based on image data, comprising:

[0005] Acquire first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated;

[0006] Acquire second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated;

[0007] Analyze the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information;

[0008] Obtain information on disaster types;

[0009] Acquire a corresponding disaster rating strategy according to the disaster type information;

[0010] Disaster severity information is obtained based on the processed first image information, second image information and disaster rating strategy analysis.

[0011] Optionally, acquiring the second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated includes:

[0012] Acquire an image data information set corresponding to the disaster location information to be calibrated and collected by a second remote sensing platform in multiple periods, wherein the image data information set includes multiple image information and time information corresponding to each image information;

[0013] sorting the image information according to the time information corresponding to each image information;

[0014] Calculate the similarity information of two adjacent image information respectively;

[0015] The image information with earlier time information among the two image information with the lowest similarity is obtained as the second image information.

[0016] Optionally, analyzing the first image information and the second image information according to a preset analysis strategy includes:

[0017] Preprocessing the first image information and the second image information;

[0018] The processed first image information and the second image information are compared to obtain disaster location information and disaster severity information.

[0019] Optionally, the preprocessing the first image information and the second image information includes:

[0020] performing proportional adjustment on the second image information;

[0021] grayscale the first image information and the second image information after the ratio adjustment;

[0022] Performing denoising processing on the first image information and the second image information after the ratio adjustment;

[0023] The first image information and the scale-adjusted second image information are superimposed.

[0024] Optionally, superimposing the first image information and the second image information includes:

[0025] Determine a first reference point and a second reference point at a position other than a position corresponding to the disaster position information in the first image information;

[0026] identifying the first reference point and the second reference point in the second image information;

[0027] The first image information and the second image information are superimposed based on the first reference point and the second reference point.

[0028] Optionally, analyzing the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information includes:

[0029] The disaster location information to be calibrated is calibrated based on the superimposed first image information and second image information, thereby acquiring the calibrated disaster location information.

[0030] Optionally, obtaining disaster severity information according to the processed first image information, second image information and disaster rating strategy analysis includes:

[0031] Obtaining a disaster development speed according to the disaster type, the first image information, and the second image information;

[0032] Obtaining a disaster status according to the disaster type and the first image information;

[0033] Obtain disaster development prognosis based on disaster type, surrounding environment information and weather information.

[0034] Optionally, obtaining the disaster development speed according to the disaster type, the first image information and the second image information includes:

[0035] Obtain disaster assessment characteristics based on disaster type;

[0036] identifying disaster assessment features in the first image information and the second image information respectively;

[0037] The disaster development speed is obtained according to the shooting time difference between the first image information and the second image information and the disaster assessment feature in the first image information and the disaster assessment feature in the second image information.

[0038] According to another aspect of the present invention, there is also provided a disaster analysis device based on image data, comprising:

[0039] A first image information acquisition module, which acquires first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated;

[0040] A second image information acquisition module, which acquires second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated;

[0041] A disaster location analysis module analyzes the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information;

[0042] Disaster type acquisition module, to obtain disaster type information;

[0043] A disaster rating strategy acquisition module is used to acquire a corresponding disaster rating strategy according to the disaster type information;

[0044] The disaster degree analysis module obtains disaster degree information based on the processed first image information, second image information and disaster rating strategy analysis.

[0045] The above technical solution of the present invention has the following beneficial technical effects:

[0046] The method and device for disaster analysis based on image data provided by the present application can quickly obtain disaster location information and disaster extent information by analyzing image information collected by drones and satellites. At the same time, the method combines the first image information and the second image information as well as the surrounding environment information and weather information for comprehensive analysis to obtain information in multiple dimensions such as the speed of disaster development, the state of disaster, and the prognosis of disaster development, thereby helping rescue personnel to formulate rescue strategies in a targeted manner. The method has the advantages of fast speed, high recognition accuracy, and significant robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of a disaster analysis method based on image data according to an embodiment of the present invention;

[0048] Figure 2 It is a schematic diagram of the location of image information.

[0049] Figure 3 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0051] Obviously, the described embodiments are only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] In the description of the present invention, it should be noted that the terms “first”, “second” and “third” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0053] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0054] The present invention is aimed at monitoring multiple elements of natural resources, including forest resources, marine resources, lake resources, etc., and specifically includes disaster analysis, and the disasters may be forest disasters, water disasters, etc.; more specifically, the present invention provides a disaster analysis method based on image data.

[0055] Reference Figure 1One aspect of the present invention provides a disaster analysis method based on image data, comprising:

[0056] Acquire first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated; for example, the first remote sensing platform may be a drone, a camera or other imaging equipment, in short, the first remote sensing platform acquires real-time image information;

[0057] Acquire second image information collected by a second remote sensing platform according to a preset second image information acquisition strategy and the first position information; the second remote sensing platform may be a satellite, or other platform with periodic shooting; the second image information acquisition strategy may be to acquire the second image information according to the position information carried by the first image information;

[0058] Analyze the first image information, the first location information, and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information;

[0059] Obtain information on disaster types;

[0060] Acquire a corresponding disaster rating strategy according to the disaster type information;

[0061] The disaster degree information is obtained by analyzing the processed first image information, the second image information and the disaster rating strategy. Accurate location information can be obtained by comparing the first image information with the second image information, and the disaster degree can be analyzed according to the surrounding environment and disaster area in the image.

[0062] The image data-based disaster analysis method provided in the present application performs analysis based on first image information collected by a first remote sensing platform and second image information collected by a second remote sensing platform, and can quickly obtain disaster location information and disaster severity information, thereby helping emergency rescue personnel to formulate rescue strategies in a targeted manner.

[0063] In one embodiment, acquiring the second image information collected by the second remote sensing platform according to the preset second image information acquisition strategy and the disaster location information to be calibrated includes:

[0064] Acquire an image data information set corresponding to the disaster location information to be calibrated and collected by the second remote sensing platform in multiple periods, the image data information set including multiple image information and time information corresponding to each image information;

[0065] sorting the image information according to the time information corresponding to each image information;

[0066] Calculate the similarity information of two adjacent image information respectively;

[0067] The image information with earlier time information among the two image information with the lowest similarity is obtained as the second image information.

[0068] In one embodiment, obtaining an image data information set corresponding to the disaster location information to be calibrated and collected by the second remote sensing platform in multiple cycles includes:

[0069] identifying disaster location information to be calibrated in the first image information;

[0070] According to the disaster location information to be calibrated, an image data information set collected by the second remote sensing platform in multiple cycles is obtained.

[0071] In one embodiment, identifying the disaster location information to be calibrated in the first image information includes:

[0072] Obtain disaster type data, which can be obtained through manual input;

[0073] Obtain the neural network model corresponding to the disaster type;

[0074] A neural network model is used to identify disaster location information to be calibrated in the first image information.

[0075] For example, a neural network model is trained for each disaster type according to its own characteristics, and the trained neural network model is used to identify the distance of the boundary of the disaster location to be calibrated in the first image information relative to the boundary of the first image information, and the disaster location information to be calibrated is obtained according to the first location information and the scale of the first image information. For example, the size of the first image is 5CM*5CM, and the corresponding scale is 1 kilometer per centimeter. At the same time, the four boundaries of the emcee image correspond to different longitudes and latitudes, respectively. Then, the disaster location information (longitude and latitude information) to be calibrated can be obtained according to the above information.

[0076] Preferably, the neural network model may use a Mask R-CNN model;

[0077] A trained neural network model is used to identify the boundary of the disaster location to be calibrated in the first image information. According to the boundary of the disaster location information to be calibrated in the first image information, a variety of methods can be used to calculate the distance from the boundary of the disaster location information to be calibrated to the boundary of the original image, including calculating the distance between edge pixels and the edge of the original image, using a foreground mask to calculate the distance to the edge of the water surface, etc.

[0078] The following steps are used to train the neural network:

[0079] Data collection and preprocessing: Collect image data containing the corresponding disaster type, for example, for fire, obtain a variety of fire images and other disaster images. At the same time, preprocess the satellite images, such as resizing, mirror flipping, rotating, filling blank areas, etc., to expand the diversity and quantity of the training set.

[0080] Labeling and classification: Label and classify the collected satellite images, and classify fire and other disasters as two categories. At the same time, divide the data into training set and test set, and use some professional remote sensing software or programming languages ​​such as Python for data processing.

[0081] Network design: Select a suitable neural network structure and set corresponding hyperparameters, such as network depth, convolution kernel size, learning rate, etc. For example, the network structure includes convolutional neural network (CNN), residual neural network (ResNet), etc. Preferably, the Mask R-CNN (Mask Region-based Convolutional Neural Network) model can be used.

[0082] For example, the main model parameters are as follows:

[0083] 1. Backbone network: Use advanced convolutional neural networks such as ResNet (residual neural network) 50 / ResNet101 / ResNeXt101 as the backbone network of the model.

[0084] 2. RoIPooling (Region of Interest Pooling) Size: By default, the output size of RoIPooling is 7x7.

[0085] 3. Anchor (target detection network) size and ratio: By default, the size of Anchor ranges from 32x32 to 256x256, and the ratio ranges from 0.5 to 1.0.

[0086] 4. Maximum number of iterations: The maximum number of iterations during training, generally 100,000 to 200,000 times.

[0087] 5. Initial learning rate: The initial value of the learning rate, generally a smaller value is used, such as 0.001.

[0088] 6. Learning rate descent strategy: You can adopt a common learning rate descent strategy, such as dividing the learning rate by 10 every certain number of times.

[0089] 7. Loss function weights: Different weights can be assigned to object classification loss, bounding box regression loss, mask loss, etc. to adjust the performance of the model on different tasks.

[0090] Training network: Define the various components of the model, including RPN (region generation network), RoIPooling (region of interest pooling), Classification head, Regression head, and Segmentation head, etc. Input the processed satellite images into the neural network for training, use the cross entropy loss function to calculate the error, and update the network weights and biases through the back propagation algorithm. GPU can be used to accelerate the calculation to improve the training speed and efficiency.

[0091] Model testing: The testing process includes inputting test data, forward propagation, and outputting prediction results.

[0092] The neural network model constructed through the above steps can identify the disaster location information to be calibrated according to the first image information and the first location information.

[0093] The image data information set obtained based on the disaster location information includes:

[0094] Obtain a set of satellite image data covering disaster location information.

[0095] For example, the image information is sorted according to the time information corresponding to each image information;

[0096] Calculate the similarity information of two adjacent images respectively;

[0097] Acquire the image information with earlier time information among the two image information with the lowest similarity as the second image information;

[0098] For example, a disaster (fire, mudslide, earthquake, etc.) occurs at a certain location. At this time, a drone is used to photograph the location to generate first image information. At the same time, the drone has location information acquisition modules such as GPS and Beidou, which can collect location information (latitude and longitude) corresponding to the photographed first image information. It can be known that the shooting cycle of the satellite is fixed (1 day / 3 days / 5 days, etc.). Here we take the shooting cycle of 1 day as an example, and obtain the image information of the location photographed by the satellite in 5 cycles, namely A1 (1 day ago), A2 (2 days ago), A3 (3 days ago), A4 (4 days ago), and A5 (5 days ago). At this time, the similarity between A1 and A2, the similarity between A2 and A3, the similarity between A3 and A4, and the similarity between A4 and A5 are calculated; by comparison, it is found that the similarity between A1 and A2 is the lowest (that is, the disaster occurred during this period), and A2 is obtained as the second image information. This part can be set according to the actual situation. If the satellite shooting cycle is long (for example, 1 month or 3 months), and the last shooting time is a long time ago (greater than the threshold), then you can directly select the most recent shooting image information as an example without similarity comparison.

[0099] In one embodiment, analyzing according to the analysis strategy, the first image information, the first location information, and the second image information includes:

[0100] preprocessing the first image information and the second image information;

[0101] The processed first image information is compared with the second image information to obtain calibrated disaster location information and disaster extent information.

[0102] In one embodiment, preprocessing the first image information and the second image information includes:

[0103] Performing proportional adjustment on the second image information;

[0104] grayscale the first image information and the second image information after the ratio adjustment;

[0105] Performing denoising processing on the first image information and the second image information after the ratio adjustment;

[0106] The first image information and the second image information after the ratio adjustment are superimposed.

[0107] In one embodiment, scaling the second image information includes:

[0108] Acquire first scale information corresponding to the first image information;

[0109] Acquire second scale information corresponding to the second image information;

[0110] adjusting the scale of the second image information according to the first scale information and the second scale information, wherein adjusting the scale includes enlarging or reducing the second image information;

[0111] grayscale the first image information and the scaled second image information;

[0112] De-noising is performed on the first image information and the scaled second image information.

[0113] For example, the first scale of the first image information acquired by the drone is 1 / 500, and the second scale of the second image information acquired by the satellite is 1 / 1000. Then, the satellite image is enlarged according to the proportional relationship.

[0114] In one embodiment, superimposing the first image information and the second image information includes:

[0115] Determine a first reference point and a second reference point at a position other than the position corresponding to the disaster position information in the first image information; for example, the first reference point and the second reference point may be manually marked or identified by an algorithm. When the algorithm is used for identification, the first reference point and the second reference point are the first point and the second point with the highest contrast at a position other than the position corresponding to the disaster position information in the first image information; the first reference point and the second reference point both represent position information;

[0116] identifying the first reference point and the second reference point in the second image information;

[0117] The first image information and the second image information are superimposed based on the first reference point and the second reference point.

[0118] Superimposing the first image information and the second image information based on the first reference point and the second reference point means:

[0119] Recognize the first reference point and the second reference point in the first image information and the second image information respectively, align the first reference point in the first image information with the first reference point in the second image information, and align the second reference point in the first image information with the second reference point in the second image information; before this step, the two image data need to be adjusted to the same accuracy (same scale);

[0120] The first image information is used as the bottom layer, and the second image information is used as the first layer for superposition, and the opacity of the second image information is adjusted according to the situation, for example, it can be adjusted to 80%;

[0121] Identify the first reference point and the second reference point in the second image information; the identification in this step can be performed by manual annotation or by using a neural network model, and the method of using the neural network model is the same as described above.

[0122] The higher the contrast, the clearer the image. Therefore, using the point with the highest contrast as a landmark for identification can effectively improve the recognition accuracy. This method can effectively improve the matching accuracy by performing a secondary match when the longitude and latitude are initially matched. In terms of landmark selection, this method selects landmarks at locations that have not been damaged by disasters to avoid situations where the landmark cannot be matched due to damage. Therefore, it has the advantages of fast speed and significant robustness.

[0123] For example, refer to Figure 2 , the second image information without proportional adjustment is A, the second image information after proportional adjustment is B, the first image information is C, and the image information after superposition is D; wherein, the first reference point is Z1, the second reference point is Z2, the areas corresponding to the first image information and the second image information after preprocessing are both X, the area corresponding to the disaster location information is Y, and the area outside the area corresponding to the disaster location information in the first image information refers to the area belonging to X and not belonging to Y, that is, the location of Z1 and Z2.

[0124] For example, superimposing the first image information and the second image information includes:

[0125] Image grayscale processing: Grayscale processing refers to processing RGB images into grayscale images to reduce algorithm pressure. It can be obtained by weighted average method.

[0126] Image denoising,Image denoising can be processed by median filtering.

[0127] The SURF (Speed-Up Robust Features) algorithm is used to identify the first reference point and the second reference point in the preprocessed first image information and the second image information to complete the superposition.

[0128] In one embodiment, the first image information and the second image information are analyzed according to a preset disaster location analysis strategy to obtain calibrated disaster location information including:

[0129] The disaster location information to be calibrated is calibrated based on the superimposed first image information and second image information, thereby acquiring the calibrated disaster location information.

[0130] For example, superimposing the first image information and the second image information based on the first reference point and the second reference point to obtain the calibrated disaster location information includes:

[0131] Superimposing the first image information and the second image information based on the first reference point and the second reference point means:

[0132] Recognize the first reference point and the second reference point in the first image information and the second image information respectively, align the first reference point in the first image information with the first reference point in the second image information, and align the second reference point in the first image information with the second reference point in the second image information; before this step, the two image data need to be adjusted to the same accuracy (same scale);

[0133] The first image information is used as the bottom layer, and the second image information is used as the first layer for superposition, and the opacity of the second image information is adjusted according to the situation, for example, it can be adjusted to 80%;

[0134] Based on the above steps, the first image information and the second image information are superimposed;

[0135] Calibrating the disaster location information to be calibrated based on the superimposed first image information and second image information includes:

[0136] Identifying a position where the superimposed first image information and the second image information have differences;

[0137] The boundary of the distinguished position is obtained to obtain calibrated disaster location information.

[0138] Since the second image information is taken before the disaster occurs, the position where the first image information and the second image information are distinguished is the disaster position. At this time, the boundary of the distinguished position is identified as the calibrated disaster position information.

[0139] In one embodiment, comparing the processed first image information with the second image information to obtain disaster location information and disaster severity information further includes:

[0140] Identify the disaster type and obtain a disaster rating strategy based on a preset disaster identification strategy and the first image information;

[0141] Disaster severity information is obtained based on the processed first image information, second image information, disaster type and disaster rating strategy.

[0142] In one embodiment, each disaster type corresponds to a disaster rating strategy;

[0143] Each disaster rating strategy consists of different characteristics depending on the type of disaster.

[0144] For example, the disaster extent information is obtained according to the first image information and the second image information:

[0145] An evaluation model composed of different features is established for each disaster. For example, the characteristics of a fire can be the proportion of red and yellow in the graphics of the first image part corresponding to the calibrated disaster location information, the proportion of combustibles other than the disaster location information, whether the disaster location information is in direct contact with the combustibles, the wind direction, the direction of the surrounding combustibles, etc.

[0146] For example, in the first image information, the red color accounts for 1%, and at the same time, there are no combustibles outside the disaster location, and the disaster level is low; in short, the disaster rating strategy can set different conditions according to different disaster types, different weather factors, etc.

[0147] In one embodiment, the disaster degree information obtained by analyzing the processed first image information, the second image information and the disaster rating strategy includes:

[0148] Obtaining a disaster development speed according to the disaster type, the first image information, and the second image information;

[0149] Obtaining a disaster status according to the disaster type and the first image information;

[0150] Obtain disaster development prognosis based on disaster type, surrounding environment information and weather information.

[0151] In one embodiment, obtaining the disaster development speed according to the disaster type, the first image information and the second image information includes:

[0152] Obtain disaster assessment characteristics based on disaster type;

[0153] identifying disaster assessment features in the first image information and the second image information respectively;

[0154] The disaster development speed is obtained according to the shooting time difference between the first image information and the second image information and the disaster assessment feature in the first image information and the disaster assessment feature in the second image information.

[0155] It is worth noting that the disaster assessment features vary according to the type of disaster. For example, the disaster assessment features of earthquake disasters can be the proportion of the collapsed area of ​​the building to the total building area, the length of the cracks on the ground, the overall disaster area, etc.; the disaster assessment features of fire disasters can be the color and area of ​​the flames, etc.; the disaster assessment features of rainstorms can be the light value, the texture features of the image, etc. The above disaster feature recognition can be performed using a trained neural network model;

[0156] Taking earthquake disasters as an example, the disaster development speed can be obtained based on the shooting time difference between the first image information and the second image information and the disaster assessment features in the first image information and the disaster assessment features in the second image information. For example, the disaster development speed can be obtained by taking the difference between the length of the cracks on the ground in the first image and the length of the cracks on the ground in the second image, and dividing the difference by time.

[0157] In one embodiment, obtaining the disaster status according to the disaster type and the first image information includes:

[0158] Obtain disaster assessment characteristics and evaluation indicators of disaster assessment characteristics according to disaster types;

[0159] A disaster assessment feature is identified in the first image information, and the identified disaster assessment feature is evaluated according to an evaluation index of the disaster assessment feature.

[0160] Among them, the disaster assessment features are the same as above. The evaluation indicators of disaster assessment features can be constructed according to the actual scene. For example, the color of the flame corresponds to different disaster levels according to different colors, and the fire area corresponds to different disaster levels according to the size of the area.

[0161] In one embodiment, obtaining the disaster development prognosis according to the disaster type, surrounding environment information and weather information includes:

[0162] Obtain surrounding environment information evaluation indicators and weather information evaluation indicators according to disaster categories;

[0163] The surrounding environment information and weather information are evaluated using the surrounding environment information evaluation index and weather information evaluation index to obtain the disaster development prognosis.

[0164] Taking fire as an example, the surrounding environment information evaluation indicators include the area of ​​flammable materials. In case of heavy rain, the surrounding environment information evaluation indicators include the surrounding terrain, altitude, etc.; weather information includes wind speed, temperature, rainfall, etc.

[0165] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes the following process of the above-mentioned embodiment of each method. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0166] In another aspect of the present application, there is also provided a disaster analysis device based on image data, comprising: a first image information acquisition module, which acquires first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated;

[0167] A second image information acquisition module, which acquires second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated;

[0168] A disaster location analysis module analyzes the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information;

[0169] Disaster type acquisition module, to obtain disaster type information;

[0170] A disaster rating strategy acquisition module is used to acquire a corresponding disaster rating strategy according to the disaster type information;

[0171] The disaster degree analysis module obtains disaster degree information based on the processed first image information, second image information and disaster rating strategy analysis.

[0172] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The modules in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs.

[0173] The present invention is intended to protect a disaster analysis method and device based on image data, the method comprising:

[0174] Acquire first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated;

[0175] Acquire second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated;

[0176] Analyze the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information;

[0177] Obtain information on disaster types;

[0178] Acquire a corresponding disaster rating strategy according to the disaster type information;

[0179] Disaster severity information is obtained based on the processed first image information, second image information and disaster rating strategy analysis.

[0180] The method and device for disaster analysis based on image data provided by the present application can quickly obtain disaster location information and disaster severity information based on image information collected by drones and satellites, thereby helping rescue personnel to formulate rescue strategies in a targeted manner. The method has the advantages of fast speed, high recognition accuracy, and significant robustness.

[0181] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the above-mentioned disaster analysis method based on image data when executing the computer program.

[0182] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the disaster analysis method based on image data as described above can be implemented.

[0183] Figure 3 is an exemplary structural diagram of an electronic device capable of implementing a disaster analysis method based on image data provided according to an embodiment of the present invention.

[0184] like Figure 3 As shown, the electronic device includes an input device 501, an input interface 502, a central processor 503, a memory 504, an output interface 505, and an output device 506. The input interface 502, the central processor 503, the memory 504, and the output interface 505 are interconnected through a bus 507, and the input device 501 and the output device 506 are connected to the bus 507 through the input interface 502 and the output interface 505, respectively, and then connected to other components of the electronic device. Specifically, the input device 501 receives input information from the outside, and transmits the input information to the central processor 503 through the input interface 502; the central processor 503 processes the input information based on the computer executable instructions stored in the memory 504 to generate output information, temporarily or permanently stores the output information in the memory 504, and then transmits the output information to the output device 506 through the output interface 505; the output device 506 outputs the output information to the outside of the electronic device for use by the user.

[0185] That is to say, Figure 3 The electronic device shown may also be implemented as comprising: a memory storing computer executable instructions; and one or more processors, which may implement the combination of Figure 1 Described is a disaster analysis method based on image data.

[0186] In one embodiment, Figure 3The electronic device shown can be implemented to include: a memory 504 configured to store executable program codes; and one or more processors 503 configured to run the executable program codes stored in the memory 504 to execute the disaster analysis method based on image data in the above-mentioned embodiment.

[0187] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0188] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0189] Computer readable media include permanent and non-permanent, removable and non-removable, and the media can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), data versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0190] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

[0191] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

[0192] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0193] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

Claims

1. A disaster analysis method based on image data, characterized in that: include: Acquire first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated; Acquire second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated; Analyze the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information; Obtain information on disaster types; Acquire a corresponding disaster rating strategy according to the disaster type information; Obtaining disaster severity information based on the processed first image information, second image information and disaster rating strategy analysis; The disaster degree information obtained by analyzing the processed first image information, the second image information and the disaster rating strategy includes: Obtaining a disaster development speed according to the disaster type, the first image information, and the second image information; Obtaining a disaster status according to the disaster type and the first image information; Obtain disaster development prognosis based on disaster type, surrounding environment information and weather information; The method of obtaining the disaster development speed according to the disaster type, the first image information and the second image information comprises: Obtain disaster assessment characteristics based on disaster type; identifying disaster assessment features in the first image information and the second image information respectively; Obtaining a disaster development speed according to a shooting time difference between the first image information and the second image information and a disaster assessment feature in the first image information and a disaster assessment feature in the second image information; The step of acquiring the second image information collected by the second remote sensing platform according to the preset second image information acquisition strategy and the disaster location information to be calibrated includes: Acquire an image data information set corresponding to the disaster location information to be calibrated and collected by a second remote sensing platform in multiple periods, wherein the image data information set includes multiple image information and time information corresponding to each image information; sorting the image information according to the time information corresponding to each image information; Calculate the similarity information of two adjacent image information respectively; Acquire the image information with earlier time information among the two image information with the lowest similarity as the second image information; Analyzing the first image information and the second image information according to a preset analysis strategy includes: Preprocessing the first image information and the second image information; Comparing the processed first image information with the second image information to obtain disaster location information and disaster severity information; The preprocessing of the first image information and the second image information comprises: performing proportional adjustment on the second image information; grayscale the first image information and the second image information after the ratio adjustment; Performing denoising processing on the first image information and the second image information after the ratio adjustment; superimposing the first image information and the second image information after the ratio is adjusted; The superimposing the first image information and the second image information comprises: Determine a first reference point and a second reference point at a position other than a position corresponding to the disaster position information in the first image information; identifying the first reference point and the second reference point in the second image information; superimposing the first image information and the second image information based on the first reference point and the second reference point; The step of analyzing the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information includes: The disaster location information to be calibrated is calibrated based on the superimposed first image information and second image information, thereby acquiring the calibrated disaster location information.

2. A disaster analysis device based on image data, characterized in that: Performing disaster analysis using the disaster analysis method based on image data as claimed in claim 1 comprises: A first image information acquisition module, which acquires first image information collected by a first remote sensing platform and first position information corresponding to the first image information, wherein the first position information includes disaster position information to be calibrated; A second image information acquisition module, which acquires second image information collected by the second remote sensing platform according to a preset second image information acquisition strategy and the disaster location information to be calibrated; A disaster location analysis module analyzes the first image information and the second image information according to a preset disaster location analysis strategy to obtain calibrated disaster location information; Disaster type acquisition module, to obtain disaster type information; A disaster rating strategy acquisition module is used to acquire a corresponding disaster rating strategy according to the disaster type information; A disaster degree analysis module, which obtains disaster degree information according to the processed first image information, second image information and disaster rating strategy analysis; The disaster degree analysis module obtains the disaster development speed according to the disaster type, the first image information and the second image information, obtains the disaster state according to the disaster type and the first image information, and obtains the disaster development prognosis according to the disaster category, the surrounding environment information and the weather information; The disaster degree analysis module obtains disaster assessment features according to the disaster type, identifies the disaster assessment features in the first image information and the second image information respectively, and obtains the disaster development speed according to the shooting time difference between the first image information and the second image information and the disaster assessment features in the first image information and the second image information.

Images