Road disaster early warning system and method utilizing satellite remote sensing and unmanned aerial vehicle data complementation

Abstract

The invention relates to a road disaster early warning system and method utilizing satellite remote sensing and unmanned aerial vehicle data complementation, and relates to the field of road disaster early warning, and the method comprises the steps: collecting a remote sensing image of a target road region through satellite remote sensing, and carrying out the recognition to obtain a snow coverage region; according to the remote sensing image and the snow coverage area, analyzing to obtain the average significance of the remote sensing image, and configuring an acquisition range for performing unmanned aerial vehicle image acquisition on the target road area; the method comprises the following steps: optimizing acquisition positions of unmanned aerial vehicle image acquisition to obtain optimal acquisition position distribution, analyzing the repeatability of the acquisition positions according to an acquisition range during optimization, and merging the acquisition positions; and performing unmanned aerial vehicle image acquisition according to the optimal acquisition position distribution and the acquisition range, and performing road disaster early warning in combination with remote sensing images. The technical problem of insufficient road disaster prediction accuracy caused by incomplete ground data acquisition and insufficient reliability in road disaster prediction in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of road disaster warning, and in particular to a road disaster warning system and method utilizing satellite remote sensing and unmanned aerial vehicle data complementarity. Background Art

[0002] At present, traditional road disaster warning systems, such as snow disaster warning systems, mainly rely on meteorological data and ground sensors, and have problems such as slow response speed and limited coverage. Satellite remote sensing technology can provide road information over a large area, but due to its low resolution, it is difficult to accurately identify local disaster situations. Although drone technology has high-resolution image acquisition capabilities, it is limited by flight time and area and cannot cover a wide area. In addition, the location where drones collect ground data is currently set manually, and the collection location may be redundant or insufficient.

[0003] The existing technology lacks effective ground data collection methods, resulting in an incomplete and insufficiently reliable data basis for disaster risk prediction, which in turn leads to insufficient accuracy in road disaster prediction and makes it difficult to cope with road disaster prediction in complex environments. Summary of the invention

[0004] The present invention aims to solve the technical problem that ground data collection in road disaster prediction in the prior art is incomplete and unreliable, resulting in insufficient accuracy in road disaster prediction, and provides a road disaster early warning system and method that utilizes satellite remote sensing and drone data to complement each other.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a road disaster early warning system using satellite remote sensing and drone data complementarity, comprising:

[0007] The remote sensing recognition module is used to collect remote sensing images of the target road area using satellite remote sensing and preliminarily identify the snow-covered area;

[0008] A collection range configuration module is used to analyze and obtain the average visibility of the remote sensing image according to the remote sensing image and the snow-covered area, and configure the collection range for drone image collection of the target road area according to the average visibility;

[0009] A collection position optimization module, used for optimizing the collection positions of drone image collection in the target road area to obtain an optimal collection position distribution, wherein the repetition of the collection positions is analyzed according to the collection range during the optimization to merge the collection positions;

[0010] The road risk warning module is used to collect drone images according to the optimal collection position distribution and collection range, and to predict road disaster risks in combination with the remote sensing images, obtain the road disaster risk level, and issue warnings.

[0011] In a second aspect, the present invention provides a road disaster early warning method using satellite remote sensing and drone data complementarity, comprising:

[0012] Using satellite remote sensing, collect remote sensing images of the target road area and preliminarily identify the snow-covered area;

[0013] Analyze and obtain the average visibility of the remote sensing image according to the remote sensing image and the snow-covered area, and configure a collection range for drone image collection of the target road area according to the average visibility;

[0014] In the target road area, the acquisition positions of the drone image acquisition are optimized to obtain the optimal acquisition position distribution, wherein the duplication of the acquisition positions is analyzed according to the acquisition range during the optimization, and the acquisition positions are merged;

[0015] According to the optimal collection position distribution and collection range, drone image collection is carried out, and road disaster risk prediction is carried out in combination with the remote sensing image to obtain the road disaster risk level and issue an early warning.

[0016] The beneficial effect of the present invention is that by combining satellite remote sensing with the complementary data of drones, the accuracy and efficiency of the road disaster warning system are significantly improved. First, satellite remote sensing is used to widely collect images of target road areas, which can quickly identify snow-covered areas and provide basic data for subsequent disaster risk prediction. Drone image acquisition is introduced. By analyzing satellite remote sensing images and snow-covered areas, the acquisition range of drones is determined, the accuracy of the range of ground images collected by drones is improved, the efficiency and quality of data acquisition and disaster prediction are guaranteed, and the distribution of acquisition locations is optimized, avoiding the problem of redundant or insufficient acquisition of ground image data, improving the reliability and comprehensiveness of ground image data acquisition, and thus improving the reliability and timeliness of disaster prediction. The complementarity of high-definition images collected by drones and satellite images, and the optimization of the configuration of acquisition range and acquisition location, can help accurately identify disaster risks and ensure the timeliness and reliability of road snow disaster prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the structure of a road disaster early warning system using satellite remote sensing and UAV data complementation provided by the present invention;

[0018] Figure 2 A schematic flow chart of a road disaster early warning method using complementary satellite remote sensing and drone data provided by the present invention.

[0019] In the accompanying drawings, the components represented by the reference numerals are described as follows:

[0020] Remote sensing recognition module 11, acquisition range configuration module 12, acquisition position optimization module 13, road risk warning module 14. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0023] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0024] Embodiment 1, as Figure 1 As shown, an embodiment of the present invention provides a road disaster warning system that utilizes satellite remote sensing and UAV data to complement each other, specifically including: a remote sensing identification module 11, a collection range configuration module 12, a collection location optimization module 13, and a road risk warning module 14.

[0025] The remote sensing identification module 11 is used to execute step S10: using satellite remote sensing to collect remote sensing images of the target road area and preliminarily identify the snow-covered area.

[0026] In the embodiment of the present application, satellite remote sensing technology is first used to collect remote sensing images of the target road area. The target road area is the area where the road to be warned of snow disasters is located, such as the area within 5 km of the road and the sides of the road. Satellite remote sensing is a method of obtaining ground information through satellite sensors, which can provide wide-area surface data and is particularly suitable for large-scale monitoring and disaster assessment. In this process, satellite sensors capture ground information through spectral data in different bands to generate multi-spectral or full-color images. For example, the surface image of the target road area is obtained by the Landsat series of satellites.

[0027] Furthermore, by analyzing the spectral characteristics in remote sensing images, we can preliminarily identify snow-covered areas. Snow-covered areas have higher reflectivity in the visible and near-infrared bands, while other objects on the ground, such as soil or roads, reflect less. They can be identified by threshold-based segmentation methods, that is, according to the near-infrared band reflectivity in the remote sensing image, a specific near-infrared band reflectivity value is set, and the image is classified using this value, thereby preliminarily marking possible snow-covered areas. Alternatively, snow-covered areas in remote sensing images can be identified through machine learning.

[0028] In the embodiment of the present application, step S10 performed by the remote sensing identification module 11 further includes:

[0029] Using satellite remote sensing to collect remote sensing images of the target road area;

[0030] The remote sensing image is input into a pre-constructed snow-covered area identification channel to identify the snow-covered area, wherein the snow-covered area identification channel is pre-constructed based on a convolutional neural network training, and the training data includes sample remote sensing images and sample snow-covered areas.

[0031] In the embodiment of the present application, firstly, the remote sensing image of the target road area is collected using satellite remote sensing technology, for example, using Landsat or Sentinel-2 satellites to collect the remote sensing image of the target road area.

[0032] Furthermore, the collected remote sensing images are input into the snow-covered area recognition channel for automatic recognition of snow-covered areas. The snow-covered area recognition channel is constructed based on pre-training of a convolutional neural network, and the training data includes sample remote sensing images and sample snow-covered areas.

[0033] For example, according to the snow-covered area recognition data records in the remote sensing data in the historical time, a sample remote sensing image set can be collected, and the snow-covered area in each sample remote sensing image can be marked for location to obtain a sample snow-covered area set. Then, based on the convolutional neural network (CNN), a network structure of the snow-covered area recognition channel is built, which includes structures such as convolutional layers and pooling layers, and can automatically extract useful features from the image for image classification and recognition. Specifically, the snow-covered area recognition channel is supervised and trained using sample remote sensing images and sample snow-covered areas as training data. During the training, the network weights and other parameters are adjusted to improve the recognition accuracy of the output snow-covered area. Training of multiple sets of training data is continued until the accuracy meets the requirements, for example, the accuracy reaches 95%, and the training is completed.

[0034] The currently acquired remote sensing image is input into the trained snow-covered area identification channel to identify the snow-covered area.

[0035] The above steps effectively improved the automatic identification capability of snow-covered areas, which not only greatly reduced manual intervention, but also improved the timeliness and accuracy of the monitoring system, providing a reliable data basis for subsequent drone-based data collection and disaster risk assessment.

[0036] The acquisition range configuration module 12 is used to execute step S20: according to the remote sensing image and the snow-covered area, analyze and obtain the average visibility of the remote sensing image, and according to the average visibility, configure the acquisition range for drone image acquisition of the target road area.

[0037] In the embodiment of the present application, the average visibility of the remote sensing image is analyzed based on the remote sensing image and the snow-covered area, wherein the average visibility reflects the difference between all pixels in the remote sensing image and the pixels in the snow-covered area. The greater the difference, the greater the average visibility, indicating that the difference between the pixels in the remote sensing image and the pixels in the snow-covered area is large, and the snowfall area is small, and the probability of a snowstorm is small. Then a larger acquisition range is configured for drone ground image acquisition, such as increasing the drone's flight altitude, improving the timeliness of snowstorm warnings, and losing a certain degree of accuracy. However, since the probability of snowstorms is small, the warning needs can still be met.

[0038] On the contrary, the smaller the difference, the smaller the average visibility, indicating that the difference between the pixels in the remote sensing image and the pixels in the snow-covered area is small, the snowfall area is large, and the probability of snowstorms is high. Configure a smaller acquisition range for drone ground image acquisition, such as lowering the drone's flight altitude. Small-range image acquisition can improve image resolution and quality, and improve the accuracy of snowstorm warnings.

[0039] Step S20 performed by the acquisition range configuration module 12 in the embodiment of the present application includes:

[0040] Grayscale processing is performed on the remote sensing image, and the mean grayscale of the remote sensing image of the snow-covered area is calculated to obtain the grayscale of the snow-covered area;

[0041] The grayscale values ​​of multiple pixels are randomly selected in the grayscale remote sensing image, and the mean is calculated to obtain the random grayscale;

[0042] The grayscale deviation ratio between the random grayscale and the grayscale of the snow-covered area is calculated as the average conspicuity.

[0043] In the embodiment of the present application, the remote sensing image is grayed to convert the original multispectral remote sensing image into a single-channel gray image. Graying is the process of converting the color information in the image into gray levels of different intensities, thereby simplifying the image analysis process. The gray value of each pixel represents the brightness intensity of the location, and the gray value is usually between 0 and 255, where 0 represents black, 255 represents white, and the intermediate value represents different grays. The grayed remote sensing image also includes grayed remote sensing images in snow-covered areas.

[0044] Calculate the grayscale mean of the snow-covered area, which reflects the overall average grayscale of the snow-covered area in the remote sensing image, and obtain the grayscale of the snow-covered area. The reflectivity of the snow-covered area is high, usually appearing as a brighter area in the grayscale image, and the snow-covered area has a larger grayscale.

[0045] In order to further analyze the visibility of the snow-covered area in the image, the grayscale values ​​of multiple pixels in the grayscale remote sensing image are randomly extracted. For example, the grayscale values ​​of 100 pixels in the grayscale remote sensing image are randomly extracted, and the mean is calculated to obtain the random grayscale. Random extraction ensures randomness and globality. The random grayscale reflects the overall average grayscale of the grayscale remote sensing image. If there are more snow-covered areas in the remote sensing image, that is, more snowfall, the random grayscale is larger. On the contrary, if there are fewer snow-covered areas in the remote sensing image, that is, less snowfall, the random grayscale is smaller.

[0046] Calculate the grayscale deviation ratio between the random grayscale and the grayscale of the snow-covered area as the average visibility. For example: Grayscale deviation ratio = | grayscale of the snow-covered area - random grayscale | / grayscale of the snow-covered area, that is, the ratio of the absolute value of the difference between the random grayscale and the grayscale of the snow-covered area to the grayscale of the snow-covered area. The greater the average visibility, the less snowfall in the remote sensing image, and the smaller the average visibility, the more snowfall in the remote sensing image. For example, if the random grayscale is 100 and the snow-covered area grayscale is 200, the average visibility is 50%, that is, 0.5.

[0047] Through the above content, we can quantitatively analyze the average visibility of snow-covered areas in remote sensing images, reflect the amount of snowfall in the target road area in the remote sensing image, and provide an important basis for the subsequent configuration of the UAV image acquisition range.

[0048] In the embodiment of the present application, step S20 performed by the acquisition range configuration module 12 also includes:

[0049] Get the maximum acquisition range of the drone for image acquisition;

[0050] The average conspicuity is multiplied by the maximum acquisition range of the drone image acquisition to obtain the acquisition range.

[0051] In the embodiment of the present application, it is first necessary to determine the maximum acquisition range of the drone image acquisition. The drone image acquisition range refers to the maximum ground area that the drone can cover in one flight. The maximum acquisition range is determined by factors such as the drone's flight altitude, the camera's field of view (FOV), the camera's resolution, and the performance of the image acquisition device. For example, for a certain model of drone, the maximum range of a single snow image acquisition that meets the accuracy requirements is 10,000 m. 2 .

[0052] Furthermore, the average visibility is multiplied by the maximum acquisition range of the drone image acquisition to obtain the acquisition range, for example 0.5*10000m 2 =5000m 2 The drone's altitude and field of view can be adjusted to achieve a collection range of 5000m 2 Perform image acquisition.

[0053] The average visibility reflects the amount of snowfall in the remote sensing image. The greater the average visibility, the less snowfall, and the lower the probability of a snowstorm. A larger acquisition range is configured for drone ground image acquisition to improve the timeliness of snowstorm warnings, while losing a certain degree of accuracy. However, since the probability of snowstorms is small, the warning needs can still be met. The smaller the average visibility, the more snowfall, and the higher the probability of a snowstorm. A smaller acquisition range is configured for drone ground image acquisition, such as lowering the drone's flight altitude. Small-range image acquisition can improve image resolution and quality, and improve the accuracy of snowstorm warnings.

[0054] The acquisition position optimization module 13 is used to execute step S30: in the target road area, the acquisition position of the drone image acquisition is optimized to obtain the optimal acquisition position distribution, wherein the repetition of the acquisition position is analyzed according to the acquisition range during the optimization, and the acquisition positions are merged.

[0055] In the embodiment of the present application, when performing drone image acquisition, the optimization of the acquisition position is a key step to ensure data acquisition efficiency and quality. In order to avoid redundancy or incompleteness of the acquisition position, the duplication of the acquisition position is analyzed according to the acquisition range during the optimization, and the acquisition position is merged and optimized to reduce the overlap of the acquisition position, thereby improving the quality and value of the drone-acquired image.

[0056] Step S30 performed by the acquisition position optimization module 13 in the embodiment of the present application includes:

[0057] Dividing the area outside the snow-covered area in the remote sensing image to obtain a snow-free area;

[0058] In the target road area, randomly generate a plurality of first acquisition positions, and analyze first uniformity parameters of the plurality of first acquisition positions;

[0059] When the first uniformity parameter is greater than or equal to a uniformity threshold, dividing and obtaining a plurality of first acquisition ranges according to the plurality of first acquisition positions and acquisition ranges;

[0060] Calculating and obtaining first acquisition adaptability of the plurality of first acquisition ranges according to the repetition ratios of the plurality of first acquisition ranges and the coverage of the snow-covered boundary area;

[0061] Calculating the repetition ratio between each first acquisition range and an adjacent first acquisition range, merging the multiple first acquisition ranges to obtain a plurality of merged second acquisition ranges;

[0062] Generating a plurality of second acquisition positions in the target road area, combining the plurality of acquisition positions that merge the second acquisition range, calculating and obtaining a second uniformity parameter, and continuing the optimization when the second uniformity parameter is greater than or equal to the uniformity threshold;

[0063] Until the optimization converges, multiple collection positions with the largest collection fitness are output to obtain the optimal collection position distribution.

[0064] In an embodiment of the present application, first, after the snow-covered area is identified from the remote sensing image, the remaining area of ​​the remote sensing image is divided into the snow-free covered area to obtain the snow-free covered area.

[0065] In the target road area, a plurality of first collection positions are randomly generated. The plurality of first collection positions are completely randomly generated and may be any position in the target road area.

[0066] In order to ensure the uniform distribution of multiple first acquisition positions in the target road area and avoid concentration in a local area to affect the globality and accuracy of subsequent acquisition position analysis, the uniform distribution of multiple first acquisition positions is analyzed to obtain a first uniformity parameter.

[0067] The step of "randomly generating a plurality of first collection positions in the target road area and analyzing the first uniformity parameters of the plurality of first collection positions" performed by the collection position optimization module 13 in the embodiment of the present application includes:

[0068] Calculating the distance between each first acquisition position and an adjacent first acquisition position to obtain a plurality of first distances;

[0069] The variance of the plurality of first distances is calculated and the inverse of the variance is calculated to obtain a first uniformity parameter.

[0070] In the embodiment of the present application, each first acquisition position refers to a point predetermined for collecting images in the target road area, specifically including a coordinate. For each first acquisition position, the spatial distance between it and the other multiple first acquisition positions is calculated, for example, the distance is calculated according to the coordinates, and then the first acquisition position with the smallest distance is determined as the adjacent first acquisition position. The distance between each first acquisition position and the adjacent first acquisition position is recorded to obtain multiple first distances.

[0071] Furthermore, the variance of the multiple first distances is calculated to quantify the uniformity of the distribution of the acquisition positions. The larger the variance, the greater the deviation of the multiple first distances, the more uneven the distribution of the multiple first acquisition positions, and the worse the uniformity. Further, the reciprocal of the variance is calculated and the reciprocal is used as the first uniformity parameter. The smaller the variance, the larger the first uniformity parameter, and the more uniform the distribution of the multiple first acquisition positions.

[0072] For example, the first distances are 50m, 60m, 40m and 70m, and the calculated inverse of the variance is 0.008, that is, the first uniformity parameter is 0.008.

[0073] When the first uniformity parameter is greater than or equal to the uniformity threshold, the multiple first acquisition positions are relatively uniform, and the subsequent steps are continued. The uniformity threshold can be set by those skilled in the art according to the size of the target road area and the requirement for uniform distribution of acquisition positions, for example, 0.01.

[0074] If the first uniformity parameter is less than the uniformity threshold, multiple first acquisition positions are randomly generated again in the target road area, and the first uniformity parameter is calculated until the first uniformity parameter is greater than or equal to the uniformity threshold, thereby ensuring the uniformity and globality of the initial population in the optimization.

[0075] When the first uniformity parameter is greater than or equal to the uniformity threshold, the acquisition range configured in the above steps is used according to the multiple first acquisition positions, and the multiple first acquisition positions are used as center points and the acquisition range is used as the division range to obtain multiple first acquisition ranges. In particular, in combination with the scale of the remote sensing image, the area sizes of the multiple first acquisition ranges and the coordinates of the multiple first acquisition positions, multiple first acquisition ranges can be obtained by division within the remote sensing image.

[0076] Furthermore, according to the repetition ratio of the multiple first collection ranges and the coverage of the snow-covered boundary area, the first collection fitness of the multiple first collection ranges is calculated for the purpose of reducing the repetition ratio and improving the coverage.

[0077] The step of "calculating and obtaining the first acquisition fitness of the plurality of first acquisition ranges according to the repetition ratio of the plurality of first acquisition ranges and the coverage of the snow-covered boundary area" performed by the acquisition position optimization module 13 in the embodiment of the present application includes:

[0078] Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios;

[0079] According to the snow-covered area and the snow-free covered area, a snow-covered boundary area is obtained by dividing, and a coverage ratio of the plurality of first acquisition ranges to the snow-covered boundary area is obtained as a coverage degree;

[0080] According to the multiple first repetition ratios and coverages, the first acquisition fitness of the multiple first acquisition ranges is calculated as follows:

[0081]

[0082] Among them, FIT cov is the acquisition fitness, w1 and w2 are the repetition weight and coverage weight, the sum of w1 and w2 is 1, M is the number of multiple first acquisition ranges, M is an integer greater than 1, is the repetition ratio of the ith first acquisition range to the adjacent first acquisition range, K c For coverage.

[0083] In the embodiment of the present application, the range repetition ratio between each first acquisition range and the adjacent first acquisition range is calculated, and the first acquisition range adjacent to each first acquisition range can be determined according to the first acquisition position adjacent to each first acquisition position in the aforementioned content.

[0084] Furthermore, the range repetition ratio of each first acquisition range and the adjacent first acquisition range is calculated. For example, according to the range areas divided by multiple first acquisition ranges in the remote sensing image, according to the scale of the remote sensing image, the area where each first acquisition range and the adjacent first acquisition range intersect is measured as the intersection area, and then the ratio of the intersection area to the area of ​​the first acquisition range is calculated as the range repetition ratio. For example, the intersection area is 500m 2 , the first collection range is 5000m 2 , then the range repetition ratio is 10%. If the first acquisition range does not have an intersection with the adjacent first acquisition range, then the range repetition ratio is 0. In this way, multiple first repetition ratios are calculated.

[0085] According to the snow-covered areas and snow-free areas divided in the above content, the snow-covered boundary area can be divided in the remote sensing image, that is, the area at the junction of the snow-covered area and the snow-free area. Among them, the junction of the snow-covered area and the snow-free area is often the early sign of a snowstorm. For example, when snowfall begins, snow accumulation initially appears in the boundary area, and over time the snow will expand to the surrounding areas. Therefore, these boundary parts are often the areas where snow changes most significantly, and monitoring these areas can help identify the risk of road snowstorms early.

[0086] Exemplarily, the snow-covered boundary area includes the boundary line between the snow-covered area and the snow-free area, specifically including the area within 1 m on both sides of the boundary line. In this way, the snow-covered boundary area in the remote sensing image is obtained by division, and the area of ​​the snow-covered boundary area can be calculated.

[0087] Furthermore, according to the ranges of the multiple first acquisition ranges divided in the remote sensing image, the intersection area of ​​each first acquisition range and the snow-covered boundary area can be obtained, and the area of ​​the intersection area can be calculated according to the scale of the remote sensing image. The sum of the areas of the intersections of all the multiple first acquisition ranges and the snow-covered boundary area is calculated, and the ratio of the sum of the areas to the area of ​​the snow-covered boundary area is calculated as the coverage, that is, the ratio of the areas of the intersections of the multiple first acquisition ranges and the snow-covered boundary area to the area of ​​the snow-covered boundary area. Exemplarily, the sum of the areas of the intersections of the multiple first acquisition ranges and the snow-covered boundary area is 200m 2 The snow-covered boundary area is 400m 2 , then the coverage is 0.5.

[0088] The greater the coverage, the more the multiple first acquisition ranges can cover the junction of snow-covered areas and non-snow-covered areas, thereby improving the reliability and accuracy of snow disaster road disaster prediction and warning.

[0089] Finally, according to the multiple first repetition ratios and coverages, the first acquisition fitness of the multiple first acquisition ranges is calculated as follows:

[0090]

[0091] Among them, FIT cov is the acquisition fitness. The greater the acquisition fitness, the better the multiple first acquisition ranges are and the higher the value of the acquired images. w1 and w2 are the repetition weight and coverage weight. The sum of w1 and w2 is 1. They can be set according to the importance of reducing the repetition ratio of the acquisition range and improving the coverage, for example, 0.5 and 0.5 respectively. M is the number of the multiple first acquisition ranges. M is an integer greater than 1, for example, 10. The number of the multiple first acquisition positions is also M. is the repetition ratio of the i-th first acquisition range to the adjacent first acquisition range. As the repetition ratio increases, the fitness decreases. K c is the coverage. As the coverage increases, the fitness increases.

[0092] In this way, the first acquisition fitness of multiple first acquisition ranges is calculated. The higher the first acquisition fitness, the better the multiple first acquisition ranges are, the quality of the images of the target road area within the multiple first acquisition ranges collected by the drone is higher, and the reliability of predicting the snow disaster risk is higher.

[0093] Furthermore, during the optimization process, if the repetition ratio between a first acquisition range and an adjacent first acquisition range is large, the two first acquisition ranges are close, and the contents of the drone images collected in the closer acquisition range are similar, the first acquisition range with a larger repetition ratio and the adjacent first acquisition range can be merged to retain one acquisition range, thereby avoiding wasting computing power in similar acquisition ranges in subsequent optimization.

[0094] Specifically, the repetition ratio between each first acquisition range and an adjacent first acquisition range is calculated, and multiple groups of first acquisition ranges with a larger repetition ratio are merged to obtain multiple merged second acquisition ranges.

[0095] The step of "calculating the repetition ratio of each first acquisition range to the adjacent first acquisition range, merging the multiple first acquisition ranges, and obtaining the merged multiple merged second acquisition ranges" performed by the acquisition position optimization module 13 in the embodiment of the present application includes:

[0096] Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios;

[0097] Select the first acquisition ranges corresponding to the largest N first repetition ratios as N combined first acquisition ranges, where N is an integer greater than 1 and less than M;

[0098] Calculate the center points of the first acquisition positions of the N merged first acquisition ranges and the acquisition positions adjacent to the first acquisition range respectively as the N merged acquisition positions;

[0099] According to the N combined acquisition positions and the acquisition range, a plurality of combined second acquisition ranges are obtained by division.

[0100] In the embodiment of the present application, based on the steps in the aforementioned content, the range repetition ratio between each first acquisition range and an adjacent first acquisition range is calculated to obtain a plurality of first repetition ratios.

[0101] The first acquisition ranges corresponding to the largest N first repetition ratios are selected as N merged first acquisition ranges, where N is an integer greater than 1 and less than M, for example, 5. The larger the repetition ratio, the greater the similarity between two adjacent first acquisition ranges, and the merge is performed. The selected N merged first acquisition ranges are respectively merged with their adjacent first acquisition ranges.

[0102] Specifically, the center point of the first acquisition positions of N merged first acquisition ranges and the acquisition positions adjacent to the first acquisition range is calculated, that is, the center point between the two acquisition positions in each group of merged first acquisition ranges and adjacent first acquisition ranges is calculated as the acquisition position of the merged acquisition range, to obtain N merged acquisition positions.

[0103] For example, the coordinates of the merged first acquisition range and the two first acquisition positions adjacent to the first acquisition range are (x1, y1) and (x2, y2) respectively, and the center point, i.e. the merged acquisition position, is ((x1+x2) / 2, (y1+y2) / 2). In this way, N merged acquisition positions are obtained by merging.

[0104] Furthermore, according to the above-mentioned collection range, the collection range is divided with the coordinates of the N merged collection positions as the center point to obtain the merged N merged collection ranges, which are combined with the M-2N first collection ranges that have not been merged to obtain multiple merged second collection ranges, thus completing the merging in a round of optimization.

[0105] In the embodiment of the present application, a second round of optimization is continued to generate multiple second acquisition positions in the target road area, and the number of the multiple second acquisition positions generated is equal to the number of the merged first acquisition ranges, for example, N second acquisition positions are generated.

[0106] Based on the same method, multiple second acquisition positions and multiple acquisition positions that merge the second acquisition range are combined to continue calculating to obtain the second uniformity parameter. When it is greater than or equal to the uniformity threshold, continue to optimize. If it is less than the uniformity threshold, the second acquisition position is regenerated.

[0107] The optimization is continued until the optimization converges, for example, when a preset number of optimization rounds, such as 50 rounds, is reached, multiple acquisition positions with the largest acquisition fitness are output, and multiple optimal acquisition positions, i.e., optimal acquisition position distribution, are obtained.

[0108] Through the above steps, not only the locations of repeated collection are reduced, but also the complete coverage of the collection area can be ensured. The merged collection range is more spatially uniform, and more images of the snow-covered boundary area can be collected, which can improve the efficiency and quality of drone data collection and avoid unnecessary waste of resources. It can improve the overall performance of the road disaster warning system and ensure that the image data obtained through drone image collection has higher quality and more comprehensive geographical coverage.

[0109] The road risk warning module 14 is used to execute step S40: according to the optimal collection position distribution and collection range, drone image collection is performed, and road disaster risk prediction is performed in combination with the remote sensing image, and the road disaster risk level is obtained to issue a warning.

[0110] In an embodiment of the present application, a plurality of optimal collection positions within the optimal collection position distribution are taken as center points, the drone is controlled to fly to the plurality of center points, and then drone images are collected according to the collection range to obtain ground images at the plurality of optimal collection positions, which include information on snowfall at the plurality of optimal collection positions. Combined with remote sensing images, road disaster risk prediction is performed to obtain the road disaster risk level and to issue an early warning.

[0111] Exemplarily, based on the convolutional neural network in the machine learning technology in the prior art, sample drone images and remote sensing images can be collected, annotated according to the actual snowfall, and used as training data to train the snowfall recognition model, identify the snowfall in the drone images and remote sensing images, classify and obtain the corresponding road disaster risk level, and issue an early warning. It is also possible to manually identify the snowfall in the image based on the technical personnel in this field, and combine the snowfall area to evaluate and classify the corresponding road disaster risk level, such as blue, yellow, and red snowfall warning levels, and issue an early warning. For example, a snowfall of 0mm-5cm is a blue snowfall warning level, 5cm-15cm is a yellow snowfall warning level, and 15cm or more is a red snowfall warning level.

[0112] The road disaster warning system using satellite remote sensing and UAV data complementation provided by the embodiment of the present invention has at least the following technical effects:

[0113] By combining satellite remote sensing with the complementary data of drones, the accuracy and efficiency of the road disaster warning system have been significantly improved. First, satellite remote sensing is used to widely collect images of target road areas, which can quickly identify snow-covered areas and provide basic data for subsequent disaster risk prediction. Drone image acquisition is introduced. By analyzing satellite remote sensing images and snow-covered areas, the acquisition range of drones is determined, the accuracy of the range of ground images collected by drones is improved, the efficiency and quality of data collection and disaster prediction are guaranteed, and the distribution of acquisition locations is optimized to avoid the problem of redundant or insufficient ground image data collection, improve the reliability and comprehensiveness of ground image data collection, and thus improve the reliability and timeliness of disaster prediction. The high-definition images collected by drones complement satellite images, and optimize the configuration of acquisition range and acquisition location, which can help accurately identify disaster risks and ensure the timeliness and reliability of road snow disaster prediction.

[0114] Embodiment 2, as Figure 2 As shown, based on the same inventive concept of the road disaster warning system using satellite remote sensing and drone data complementation provided in Example 1, the embodiment of the present invention also provides a road disaster warning method using satellite remote sensing and drone data complementation, including:

[0115] Using satellite remote sensing, collect remote sensing images of the target road area and preliminarily identify the snow-covered area;

[0116] Analyze and obtain the average visibility of the remote sensing image according to the remote sensing image and the snow-covered area, and configure a collection range for drone image collection of the target road area according to the average visibility;

[0117] In the target road area, the acquisition positions of the drone image acquisition are optimized to obtain the optimal acquisition position distribution, wherein the duplication of the acquisition positions is analyzed according to the acquisition range during the optimization, and the acquisition positions are merged;

[0118] According to the optimal collection position distribution and collection range, drone image collection is carried out, and road disaster risk prediction is carried out in combination with the remote sensing image to obtain the road disaster risk level and issue an early warning.

[0119] Furthermore, satellite remote sensing is used to collect remote sensing images of the target road area and initially identify the snow-covered area, including:

[0120] Using satellite remote sensing to collect remote sensing images of the target road area;

[0121] The remote sensing image is input into a pre-constructed snow-covered area identification channel to identify the snow-covered area, wherein the snow-covered area identification channel is pre-constructed based on a convolutional neural network training, and the training data includes sample remote sensing images and sample snow-covered areas.

[0122] Further, according to the remote sensing image and the snow-covered area, the average visibility of the remote sensing image is obtained by analysis, including:

[0123] Grayscale processing is performed on the remote sensing image, and the mean grayscale of the remote sensing image of the snow-covered area is calculated to obtain the grayscale of the snow-covered area;

[0124] The grayscale values ​​of multiple pixels are randomly selected in the grayscale remote sensing image, and the mean is calculated to obtain the random grayscale;

[0125] The grayscale deviation ratio between the random grayscale and the grayscale of the snow-covered area is calculated as the average conspicuity.

[0126] Further, according to the average visibility, the acquisition range of the drone image acquisition of the target road area is configured, including:

[0127] Get the maximum acquisition range of the drone for image acquisition;

[0128] The average conspicuity is multiplied by the maximum acquisition range of the drone image acquisition to obtain the acquisition range.

[0129] Furthermore, in the target road area, the acquisition position of the drone image acquisition is optimized to obtain the optimal acquisition position distribution, including:

[0130] Dividing the area outside the snow-covered area in the remote sensing image to obtain a snow-free area;

[0131] In the target road area, randomly generate a plurality of first acquisition positions, and analyze first uniformity parameters of the plurality of first acquisition positions;

[0132] When the first uniformity parameter is greater than or equal to a uniformity threshold, dividing and obtaining a plurality of first acquisition ranges according to the plurality of first acquisition positions and acquisition ranges;

[0133] Calculating and obtaining first acquisition adaptability of the plurality of first acquisition ranges according to the repetition ratios of the plurality of first acquisition ranges and the coverage of the snow-covered boundary area;

[0134] Calculating the repetition ratio between each first acquisition range and an adjacent first acquisition range, merging the multiple first acquisition ranges to obtain a plurality of merged second acquisition ranges;

[0135] Generating a plurality of second acquisition positions in the target road area, combining the plurality of acquisition positions that merge the second acquisition range, calculating and obtaining a second uniformity parameter, and continuing the optimization when the second uniformity parameter is greater than or equal to the uniformity threshold;

[0136] Until the optimization converges, multiple collection positions with the largest collection fitness are output to obtain the optimal collection position distribution.

[0137] Furthermore, in the target road area, a plurality of first acquisition positions are randomly generated, and first uniformity parameters of the plurality of first acquisition positions are analyzed, including:

[0138] Calculating the distance between each first acquisition position and an adjacent first acquisition position to obtain a plurality of first distances;

[0139] The variance of the plurality of first distances is calculated and the inverse of the variance is calculated to obtain a first uniformity parameter.

[0140] Further, according to the repetition ratio of the multiple first collection ranges and the coverage of the snow-covered boundary area, the first collection adaptability of the multiple first collection ranges is calculated, including:

[0141] Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios;

[0142] According to the snow-covered area and the snow-free covered area, a snow-covered boundary area is obtained by dividing, and a coverage ratio of the plurality of first acquisition ranges to the snow-covered boundary area is obtained as a coverage degree;

[0143] According to the multiple first repetition ratios and coverages, the first acquisition fitness of the multiple first acquisition ranges is calculated as follows:

[0144]

[0145] Among them, FIT cov is the acquisition fitness, w1 and w2 are the repetition weight and coverage weight, the sum of w1 and w2 is 1, M is the number of multiple first acquisition ranges, M is an integer greater than 1, is the repetition ratio of the ith first acquisition range to the adjacent first acquisition range, K c For coverage.

[0146] Further, calculating the repetition ratio between each first acquisition range and an adjacent first acquisition range, merging the multiple first acquisition ranges to obtain multiple merged second acquisition ranges, including:

[0147] Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios;

[0148] Select the first acquisition ranges corresponding to the largest N first repetition ratios as N combined first acquisition ranges, where N is an integer greater than 1 and less than M;

[0149] Calculate the center points of the first acquisition positions of the N merged first acquisition ranges and the acquisition positions adjacent to the first acquisition range respectively as the N merged acquisition positions;

[0150] According to the N combined acquisition positions and the acquisition range, a plurality of combined second acquisition ranges are obtained by division.

[0151] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] 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.

[0153] 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 computer, or other programmable data processing device to generate 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.

[0154] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0156] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0157] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A road disaster warning system using satellite remote sensing and drone data complements each other, characterized by: The system comprises: The remote sensing recognition module is used to collect remote sensing images of the target road area using satellite remote sensing and preliminarily identify the snow-covered area; A collection range configuration module is used to analyze and obtain the average visibility of the remote sensing image according to the remote sensing image and the snow-covered area, and configure the collection range for drone image collection of the target road area according to the average visibility; A collection position optimization module, used for optimizing the collection positions of drone image collection in the target road area to obtain an optimal collection position distribution, wherein the repetition of the collection positions is analyzed according to the collection range during the optimization to merge the collection positions; The road risk warning module is used to collect drone images according to the optimal collection position distribution and collection range, and to predict road disaster risks in combination with the remote sensing images, obtain the road disaster risk level, and issue warnings.

2. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 1 is characterized in that: Using satellite remote sensing, collect remote sensing images of the target road area and preliminarily identify the snow-covered area, including: Using satellite remote sensing to collect remote sensing images of the target road area; The remote sensing image is input into a pre-constructed snow-covered area identification channel to identify the snow-covered area, wherein the snow-covered area identification channel is pre-constructed based on a convolutional neural network training, and the training data includes sample remote sensing images and sample snow-covered areas.

3. The road disaster early warning system using satellite remote sensing and UAV data complementation according to claim 1 is characterized in that: According to the remote sensing image and the snow-covered area, the average visibility of the remote sensing image is obtained by analysis, including: Grayscale processing is performed on the remote sensing image, and the mean grayscale of the remote sensing image of the snow-covered area is calculated to obtain the grayscale of the snow-covered area; The grayscale values ​​of multiple pixels are randomly selected in the grayscale remote sensing image, and the mean is calculated to obtain the random grayscale; The grayscale deviation ratio between the random grayscale and the grayscale of the snow-covered area is calculated as the average conspicuity.

4. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 1 is characterized in that: According to the average visibility, the acquisition range of the drone image acquisition of the target road area is configured, including: Get the maximum acquisition range of the drone for image acquisition; The average conspicuity is multiplied by the maximum acquisition range of the drone image acquisition to obtain the acquisition range.

5. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 1 is characterized in that: In the target road area, the acquisition position of the drone image acquisition is optimized to obtain the optimal acquisition position distribution, including: Dividing the area outside the snow-covered area in the remote sensing image to obtain a snow-free area; In the target road area, randomly generate a plurality of first acquisition positions, and analyze first uniformity parameters of the plurality of first acquisition positions; When the first uniformity parameter is greater than or equal to a uniformity threshold, dividing and obtaining a plurality of first acquisition ranges according to the plurality of first acquisition positions and acquisition ranges; Calculating and obtaining first acquisition adaptability of the plurality of first acquisition ranges according to the repetition ratios of the plurality of first acquisition ranges and the coverage of the snow-covered boundary area; Calculating the repetition ratio between each first acquisition range and an adjacent first acquisition range, merging the multiple first acquisition ranges to obtain a plurality of merged second acquisition ranges; Generating a plurality of second acquisition positions in the target road area, combining the plurality of acquisition positions that merge the second acquisition range, calculating and obtaining a second uniformity parameter, and continuing the optimization when the second uniformity parameter is greater than or equal to the uniformity threshold; Until the optimization converges, multiple collection positions with the largest collection fitness are output to obtain the optimal collection position distribution.

6. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 5 is characterized in that: In the target road area, a plurality of first acquisition positions are randomly generated, and first uniformity parameters of the plurality of first acquisition positions are analyzed, including: Calculating the distance between each first acquisition position and an adjacent first acquisition position to obtain a plurality of first distances; The variance of the plurality of first distances is calculated and the inverse of the variance is calculated to obtain a first uniformity parameter.

7. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 5 is characterized in that: Calculating and obtaining the first acquisition adaptability of the plurality of first acquisition ranges according to the repetition ratios of the plurality of first acquisition ranges and the coverage of the snow-covered boundary area includes: Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios; According to the snow-covered area and the snow-free covered area, a snow-covered boundary area is obtained by dividing, and a coverage ratio of the plurality of first acquisition ranges to the snow-covered boundary area is obtained as a coverage degree; According to the multiple first repetition ratios and coverages, the first acquisition fitness of the multiple first acquisition ranges is calculated as follows: Among them, FIT cov is the acquisition fitness, w1 and w2 are the repetition weight and coverage weight, the sum of w1 and w2 is 1, M is the number of multiple first acquisition ranges, M is an integer greater than 1, is the repetition ratio of the ith first acquisition range to the adjacent first acquisition range, K c For coverage.

8. The road disaster early warning system using satellite remote sensing and drone data complementation according to claim 7 is characterized in that: Calculating the repetition ratio of each first acquisition range to an adjacent first acquisition range, merging the multiple first acquisition ranges to obtain multiple merged second acquisition ranges, including: Calculating a range repetition ratio between each first acquisition range and an adjacent first acquisition range to obtain a plurality of first repetition ratios; Select the first acquisition ranges corresponding to the largest N first repetition ratios as N combined first acquisition ranges, where N is an integer greater than 1 and less than M; Calculate the center points of the first acquisition positions of the N merged first acquisition ranges and the acquisition positions adjacent to the first acquisition range respectively as the N merged acquisition positions; According to the N combined acquisition positions and the acquisition range, a plurality of combined second acquisition ranges are obtained by division.

9. A road disaster early warning method using satellite remote sensing and drone data complementarity, characterized in that: The method is applied to the road disaster early warning system using satellite remote sensing and drone data complementation as described in any one of claims 1 to 8, comprising: Using satellite remote sensing, collect remote sensing images of the target road area and preliminarily identify the snow-covered area; Analyze and obtain the average visibility of the remote sensing image according to the remote sensing image and the snow-covered area, and configure a collection range for drone image collection of the target road area according to the average visibility; In the target road area, the acquisition positions of the drone image acquisition are optimized to obtain the optimal acquisition position distribution, wherein the duplication of the acquisition positions is analyzed according to the acquisition range during the optimization, and the acquisition positions are merged; According to the optimal collection position distribution and collection range, drone image collection is carried out, and road disaster risk prediction is carried out in combination with the remote sensing image to obtain the road disaster risk level and issue an early warning.

Images