Route planning method, device, storage medium and electronic equipment

Abstract

The present invention discloses a route planning method, device, storage medium, and electronic device. The method comprises: obtaining multi-source remote sensing information of a target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; inputting the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; mapping the multiple preset planned routes to the target remote sensing image to obtain a target remote sensing image containing the preset planned routes; and screening each preset planned route in the target remote sensing image containing the preset planned routes to determine a target planning scheme. Through the above technical solution, it is possible to screen the planned routes in the target remote sensing image, avoid on-site exploration, and improve the efficiency of route planning.

Description

Technical Field

[0001] The present invention relates to the technical field of resource allocation, and in particular to a line planning method, device, storage medium and electronic equipment. Background Art

[0002] In recent years, the growth rate of social electricity consumption has accelerated, and the power supply pressure of the existing power grid structure is relatively large. It is necessary to continue to add new power grid lines, or optimize existing power grid lines, and further develop power grid construction.

[0003] In existing technologies, manual planning of power grid lines is usually adopted. Power grid line planning is based on load forecasting and power supply planning. Power grid line planning determines when, where and what type of transmission lines and the number of circuits to be built to achieve the required transmission capacity within the planning period. It also ensures economical and safe operation while meeting various technical indicators.

[0004] However, in actual grid line planning, the manual planning method of grid lines often requires a large amount of manpower and material resources to be wasted for on-site surveys, and the grid line planning cycle is long, resulting in low grid line planning efficiency. Summary of the Invention

[0005] Embodiments of the present invention provide a route planning method, device, storage medium, and electronic device to improve the efficiency of route planning.

[0006] In a first aspect, an embodiment of the present invention provides a route planning method, including:

[0007] Acquiring multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image;

[0008] Inputting the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area;

[0009] The plurality of preset planned routes are respectively mapped to the target remote sensing map to obtain a target remote sensing map containing the preset planned routes, and each preset planned route is screened in the target remote sensing map containing the preset planned routes to determine a target planning scheme.

[0010] In a second aspect, an embodiment of the present invention further provides a route planning device, comprising:

[0011] An information acquisition module is used to acquire multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image;

[0012] a target image generation module, configured to input the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area;

[0013] The target scheme determination module is used to map the multiple preset planning routes to the target remote sensing map respectively, obtain the target remote sensing map containing the preset planning routes, and screen each preset planning route in the target remote sensing map containing the preset planning routes to determine the target planning scheme.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising:

[0015] one or more processors;

[0016] a storage device for storing one or more programs,

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement any route planning method described in any one of the embodiments of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer-executable instructions, which, when executed by a computer processor, are used to execute any of the route planning methods described in the embodiments of the present invention.

[0019] The present invention obtains multi-source remote sensing information of a target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; inputs the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; maps the multiple preset planned routes onto the target remote sensing image to obtain a target remote sensing image containing the preset planned routes; and screens each preset planned route within the target remote sensing image containing the preset planned routes to determine a target planning solution. This technical solution enables screening of planned routes within the target remote sensing image, avoiding on-site exploration and improving route planning efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings introduced here only illustrate some of the embodiments to be described by the present invention, and are not exhaustive. A person skilled in the art can derive other drawings based on these drawings without inventive effort.

[0021] Figure 1 This is a flow chart of a route planning method provided in the first embodiment of the present invention;

[0022] Figure 2 This is a flow chart of a route planning method provided in the second embodiment of the present invention;

[0023] Figure 3 This is a flow chart of a route planning method provided in the third embodiment of the present invention;

[0024] Figure 4 This is a schematic structural diagram of a route planning device provided by a fourth embodiment of the present invention;

[0025] Figure 5 This is a structural diagram of an electronic device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention.

[0027] It should also be noted that, for ease of description, only the part relevant to the present invention, rather than all of the content, is shown in the accompanying drawings. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processing or methods depicted as flow charts. Although flow charts describe various operations (or steps) as sequential processing, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. When its operation is completed, the processing can be terminated, but can also have additional steps not included in the accompanying drawings. The processing can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0028] Example 1

[0029] Figure 1 This is a flowchart of a route planning method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where a planned route scheme is automatically determined in a remote sensing image. The method can be performed by a route planning device provided in an embodiment of the present invention. The device can be implemented by software and / or hardware and can be configured on an electronic computing device, such as a terminal and / or server. Specifically, the method includes the following steps:

[0030] S110 , obtaining multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image.

[0031] The target area may be an area requiring line planning, and the line planning may be distribution network line planning or road construction line planning, which is not limited in this embodiment. Multi-source remote sensing information may be a variety of remote sensing image data, and the multi-source remote sensing information may include but is not limited to panchromatic remote sensing images and multispectral images. Panchromatic remote sensing images refer to black and white images of the panchromatic bands in the radiation of ground objects, and are high spatial resolution images; multispectral images contain spectral information of multiple bands and are high spectral resolution images. For example, if a multispectral image contains spectral information of three RGB bands, the multispectral image is an RGB color image.

[0032] A pre-defined planned route can be composed of multiple target points, which can be power load devices or construction locations in the distribution network. For example, the distribution load can be used as the overall target. Multiple pre-defined planned routes can be derived based on the outage frequency, number of outages, duration, and power loss of the original distribution lines, taking into account the load points and power supply connectivity.

[0033] The method for obtaining multi-source remote sensing information may include but is not limited to: in some embodiments, real-time detection of the ground can be performed by satellite to obtain multi-source remote sensing information; in some embodiments, multi-source remote sensing information can be retrieved from a preset information storage device. This embodiment does not limit the method for obtaining multi-source remote sensing information.

[0034] S120: Input the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area.

[0035] The pre-trained route planning model can be a neural network model. Specifically, the panchromatic remote sensing image and the multispectral image are input as input data into the pre-trained route planning model. The route planning model extracts features from the panchromatic remote sensing image and the multispectral image, fuses them based on the extracted features, and thus obtains and outputs a target remote sensing image of the target area. It is understood that the target remote sensing image is a fusion image of the panchromatic remote sensing image and the multispectral image. The target remote sensing image has the advantages of both the high spatial resolution of the panchromatic remote sensing image and the high spectral resolution of the multispectral image, and can accurately identify the ground.

[0036] The route planning model can be pre-trained using a large number of sample panchromatic remote sensing images and sample multispectral images. Feature extraction is performed on these sample panchromatic remote sensing images and sample multispectral images in the trained route planning model. The model parameters are then trained based on these extracted features. The trained route planning model is then obtained by continuously adjusting the model parameters.

[0037] S130, mapping the plurality of preset planned routes to the target remote sensing map respectively to obtain a target remote sensing map containing the preset planned routes, and screening each preset planned route in the target remote sensing map containing the preset planned routes to determine a target planning scheme.

[0038] In an embodiment of the present invention, multiple target points in a preset planned route may be mapped to a target remote sensing map to obtain a target remote sensing map including the target points in the preset planned route. It is understandable that the target remote sensing map includes longitude and latitude coordinates. By mapping multiple target points in the preset planned route to the target remote sensing map, the longitude and latitude coordinates of each target point can be obtained to facilitate the selection and implementation of subsequent target plans. The mapping method may specifically be to manually draw the preset planned route to the target remote sensing map, or to automatically map the preset planned route to the target remote sensing map by matching key information with the target remote sensing map. For example, the preset planned route may include key information Factory A. If Factory A is also included in the target remote sensing map, Factory A will be used as the target point, and Factory A will be marked in the target remote sensing map.

[0039] The target planning scheme may be a planned route obtained by screening the preset planned routes. In some embodiments, the planned route with the lowest economic cost among the preset planned routes may be selected as the target planning scheme; in some embodiments, the planned route with the shortest road among the preset planned routes may be selected as the target planning scheme; in some embodiments, the terrain type of each preset planned route may be comprehensively considered, and each preset planned route may be scored according to the terrain type to determine the target planning scheme. This is not limited in the embodiments of the present invention.

[0040] An embodiment of the present invention provides a route planning method. The method obtains multi-source remote sensing information of a target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; inputs the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; maps the multiple preset planned routes to the target remote sensing image to obtain a target remote sensing image containing the preset planned routes; and screens each preset planned route in the target remote sensing image containing the preset planned routes to determine a target planning solution. This technical solution enables screening of planned routes in the target remote sensing image, avoiding on-site exploration and improving the efficiency of route planning.

[0041] Example 2

[0042] Figure 2This is a flowchart diagram of a route planning method provided in the second embodiment of the present invention. Based on the above embodiment, "obtaining multi-source remote sensing information of the target area" and "inputting the panchromatic remote sensing image and multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area" are further refined. For its specific implementation method, please refer to the detailed description of this technical solution. Among them, the technical terms that are the same as or corresponding to the above embodiment are not repeated here. Figure 2 As shown, the method of the embodiment of the present invention specifically includes the following steps:

[0043] S210: Divide the target area into multiple grids, and obtain multiple preset planned routes and panchromatic remote sensing images and multispectral images corresponding to each grid.

[0044] Among them, dividing the target area into multiple grids can realize the regional collection of full-color remote sensing images and multispectral images, so that the full-color remote sensing images and multispectral images corresponding to each grid are obtained, which is conducive to improving the speed of subsequent image processing and thus improving the efficiency of route planning.

[0045] S220 , inputting the panchromatic remote sensing image and the multispectral image corresponding to each grid into the feature extraction module to obtain a splicing feature image corresponding to each grid.

[0046] Among them, the route planning model includes a feature extraction module and a residual learning module. The feature extraction module is used to extract features from the panchromatic remote sensing images and multispectral images corresponding to each grid, and splice the extracted features to obtain the spliced feature maps corresponding to each grid.

[0047] Specifically, the feature extraction module includes a first convolutional neural network and a second convolutional neural network. The first convolutional neural network is used to extract features from the panchromatic remote sensing images and multispectral images corresponding to each grid, and the second convolutional neural network is used to splice the extracted features to obtain a spliced feature map corresponding to each grid.

[0048] For example, the first convolutional neural network can be composed of multiple convolutional layers. It should be noted that in order to prevent feature loss, pooling layers, batch normalization, and ReLU are not used. The panchromatic remote sensing image is input into the first convolutional neural network to obtain the first feature. The multispectral image is input into the first convolutional neural network to obtain the second feature. The first and second features are input into the second convolutional neural network to obtain the spliced feature map corresponding to each grid. The specific calculation process is as follows:

[0049]

[0050]

[0051]

[0052]

[0053] Among them, X p represents a full-color remote sensing image, X m Represents multispectral images, and the features extracted by convolutional neural networks are used Indicates that for the full-color remote sensing image, features are extracted from the first layer. Indicates that for multispectral images, features are extracted from the first layer. Represents the convolution output of the second layer for the full-color remote sensing image, Represents the convolution output of the second layer for multispectral images, W 1 and W 2 denote the weights of the first and second convolutional layers, respectively, and b 1 and b 2 Represent the bias values of the first and second convolution layers respectively, and 3×3 represents the size of the convolution kernel. Then the output of the second layer is and Perform feature splicing to obtain a spliced feature map

[0054] S230: Input the spliced feature map into the residual learning module to perform residual learning to obtain a target fused image corresponding to each grid.

[0055] Among them, the residual learning module can be used to perform residual learning on the spliced feature map, which can realize feature enhancement of the spliced feature map, thereby improving the resolution of the target fused image.

[0056] Based on the above embodiment, the residual learning module includes a convolution unit and a jump connection unit, and the residual learning module outputs a target fused image based on the following formula:

[0057] F b =CA(X b )+F b-1

[0058] Among them, F b represents the target fusion image corresponding to any grid, F b-1 represents the concatenated feature map, X b represents the initial feature, X b By F b-1 It is obtained by performing secondary convolution on the convolution unit, and CA(·) represents the attention mechanism function.

[0059] Among them, the jump connection unit can solve the problem of gradient disappearance in the path planning model, thereby improving the resolution of the target fusion image. The attention mechanism function can enable the route planning model to focus on more useful channels and improve the channel feature learning ability.

[0060] In the embodiment of the present invention, the convolution unit may be a network unit comprising two convolution layers. Specifically, F b-1 Input to the first convolutional layer of the convolution unit to obtain the extracted features X of the first convolutional layer b-1 , the calculation formula is as follows:

[0061]

[0062] Among them, δ(·) represents the ReLU activation function, represents the weight of the first convolutional layer, Represents the bias of the first convolutional layer, 3×3 represents the convolution kernel; then the initial feature X is obtained through the second convolutional layer b , the calculation formula is as follows:

[0063]

[0064] in, are the weights of the second convolutional layer, Represents the bias of the second convolutional layer, X b As input to the attention mechanism function.

[0065] For example, the convolution unit can use 2 layers of 3×3×256 convolution kernels, the jump connection unit can use 1 layer of 1×1×256 convolution kernels, the number of attention mechanism channels can be represented by C=(1,2,...,c), and the initial features can be represented by X=[X1,X2,...X c ], it can be understood that X is X b Any feature in , the descriptor Z of the Cth attention mechanism channel c The expression is:

[0066]

[0067] Among them, f GP (·) is the global average pooling function, H, W are the sizes of the feature map, x c (i,j) is the c-th layer feature x c The value at (i, j);

[0068] After the attention mechanism channel descriptor z passes through the downsampling layer and the upsampling layer in sequence, the channel statistic w is obtained. The channel statistic w contains the weight coefficient w of each attention mechanism channel c ,The specific expression of channel statistic w is:

[0069] w=S(W U δ(W D z));

[0070] Where S(·) represents the sigmoid activation function, δ(·) represents the ReLU activation function, and W D is the weight set of the dimensionality reduction convolution layer, W U is the weight set of the dimension-raising convolutional layer;

[0071] The role of the dimensionality reduction convolution layer is to reduce the dimensionality of the attention mechanism channel. For example, set the dimensionality reduction ratio to r, activate the reduced data with the ReLU activation function, and then increase the number of channels by r times through the dimensionality increase convolution layer to obtain the weight coefficient w of each channel. c , and obtain the final channel statistics w, which is used to rescale the initial features X.

[0072] The weight coefficient w c Multiply it by the initial feature X to obtain the enhanced feature Enhanced Features The specific expression is:

[0073]

[0074] Among them, w c and x c are the weight coefficients and initial features of the c-th layer channel, respectively. The enhanced features are magnified by the deconvolution layer, and then reconstructed and magnified by a convolution layer to obtain the target fused image corresponding to each grid of the grid.

[0075] S240 , stitching the target fusion images corresponding to the grids to obtain a target remote sensing image of the target area.

[0076] In an embodiment of the present invention, image stitching can specifically be performed by extracting feature points from the target fusion images corresponding to each grid, matching the feature points, and performing image registration on the successfully matched target fusion images to achieve image edge stitching and obtain a target remote sensing image of the target area.

[0077] S250, mapping the plurality of preset planned routes to the target remote sensing map respectively to obtain a target remote sensing map containing the preset planned routes, and screening each preset planned route in the target remote sensing map containing the preset planned routes to determine a target planning scheme.

[0078] An embodiment of the present invention provides a route planning method, which obtains multi-source remote sensing information of a target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; inputs the panchromatic remote sensing image and the multispectral image corresponding to each grid into the feature extraction module to obtain a spliced feature map corresponding to each grid; inputs the spliced feature map into a residual learning module for image fusion to obtain a target fused image corresponding to each grid; and splices the target fused images corresponding to each grid to obtain a target fused image with high spatial resolution and high spectral resolution, thereby achieving accurate recognition of the ground.

[0079] Example 3

[0080] Figure 3 The flowchart of a route planning method provided in the third embodiment of the present invention is a flowchart of the present embodiment of the present invention and the various optional schemes in the above embodiments can be combined. In the embodiment of the present invention, "mapping the plurality of preset planned routes to the target remote sensing map respectively to obtain the target remote sensing map containing the preset planned routes, and screening each preset planned route in the target remote sensing map containing the preset planned routes to determine the target planning scheme" is further refined. For its specific implementation method, please refer to the detailed description of this technical solution. Among them, the technical terms that are the same as or corresponding to the above embodiments will not be repeated here.

[0081] like Figure 3 As shown, the method of the embodiment of the present invention specifically includes the following steps:

[0082] S310: Acquire multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image.

[0083] S320: Input the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area.

[0084] S330: Mapping multiple load points in the preset planned route to a target remote sensing map to obtain a target remote sensing map including the preset planned route.

[0085] Among them, when the route planning method is a distribution network planning scenario, the target point in the preset planned route is a load point. In an embodiment of the present invention, multiple load points in the preset planned route can be mapped to a target remote sensing map to obtain a target remote sensing map containing the load points in the preset planned route. The mapping method can specifically be to manually draw multiple load points in the preset planned route to the target remote sensing map, or to automatically map the load points to the target remote sensing map by matching key information with the target remote sensing map. For example, the load point can include the key information factory B. If the target remote sensing map also includes factory B, factory B will be marked in the target remote sensing map.

[0086] S340. In the target remote sensing map including the preset planned routes, score the coordinates of each load point of any preset planned route to obtain a score corresponding to each load point in any preset planned route.

[0087] Among them, the score of the load point coordinates can be used to evaluate the rationality of the load point selection. The higher the score, the more reasonable the load point selection.

[0088] On the basis of the above embodiment, in the target remote sensing map containing the preset planned route, the coordinates of each load point of any preset planned route are scored to obtain the score corresponding to each load point in any preset planned route, including: in the target remote sensing map containing the preset planned route, identifying the terrain category corresponding to the coordinates of each load point; and determining the score corresponding to each load point based on the terrain category corresponding to the coordinates of each load point.

[0089] The terrain category refers to the type of landform corresponding to the coordinates of each load point. This includes, but is not limited to, terrain type, vegetation type, water body, first-class highway, expressway, railway, and housing. Different terrain categories can be assigned different scores, and the score for each load point is calculated based on the corresponding terrain category score.

[0090] For example, the score corresponding to each load point can be determined based on the terrain type at the load point's coordinate location and the surrounding terrain type. For example, if the load point's coordinate location is urban and the surrounding terrain type is residential, the urban base score can be set to R, the distance between the load point and the surrounding houses is K, and the corresponding score for the load point is R*K. Similarly, if the load point's coordinate location is outdoor and the surrounding terrain type is other terrain, the outdoor base score can be set to A, the distance between the load point and other surrounding terrain is B, and the corresponding score for the load point is A*B.

[0091] S350: Screen the preset planned routes based on the scores corresponding to the load points in the preset planned routes to determine a target planning solution.

[0092] The scores corresponding to the load points in the preset planned routes can be added or multiplied to obtain the scores of the preset planned routes. Then, a target planned route is selected based on the planned route scores, and the target planned route is used as the target planning scheme. The score selected can be the maximum or minimum value among the planned route scores, which can be determined based on the planned route score calculation method. In the embodiment of the present invention, the preset planned routes are screened based on the scores corresponding to the load points, which is simple and efficient, thereby improving the efficiency of route planning.

[0093] On the basis of the above embodiment, the preset planned routes are screened based on the scores corresponding to the load points in each preset planned route to determine the target planning scheme, including: adding the scores corresponding to the load points in any preset planned route to obtain the planned route score; and determining the preset planned route with the largest planned route score among the preset planned routes as the target planning scheme.

[0094] In an embodiment of the present invention, the scores corresponding to the load points in any preset planned route are added together to obtain a planned route score, and the preset planned route with the largest planned route score among the preset planned routes is determined as the target planning scheme, that is, the planned route with the lowest economic cost is selected as the target planning scheme, thereby improving the feasibility of the target planning scheme.

[0095] An embodiment of the present invention provides a route planning method. By identifying the terrain category corresponding to the coordinates of each load point in a preset planned route and determining the score corresponding to each load point based on the terrain category corresponding to the coordinates of each load point, it is possible to screen the planned route in the target remote sensing map, avoid on-site exploration, and improve the efficiency of route planning.

[0096] Example 4

[0097] Figure 4 This is a schematic diagram of the structure of a route planning device provided in a fourth embodiment of the present invention. The route planning device provided in this embodiment can be implemented using software and / or hardware and can be configured in a terminal and / or server to implement the route planning method provided in the embodiments of the present invention. The device specifically includes an information acquisition module 410, a target image generation module 420, and a target solution determination module 430.

[0098] Among them, the information acquisition module 410 is used to obtain multi-source remote sensing information of the target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; the target image generation module 420 is used to input the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area, wherein the resolution of the target remote sensing image is higher than the resolution of the panchromatic remote sensing image; the target scheme determination module 430 is used to map the multiple preset planned routes to the target remote sensing image respectively to obtain a target remote sensing image containing the preset planned routes, and screen each preset planned route in the target remote sensing image containing the preset planned routes to determine the target planning scheme.

[0099] An embodiment of the present invention provides a route planning device that obtains multi-source remote sensing information of a target area and multiple preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; inputs the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; maps the multiple preset planned routes to the target remote sensing image to obtain a target remote sensing image containing the preset planned routes; and screens each preset planned route in the target remote sensing image containing the preset planned routes to determine a target planning solution. This technical solution enables screening of planned routes in the target remote sensing image, avoiding on-site exploration and improving the efficiency of route planning.

[0100] Based on any optional technical solution in the embodiments of the present invention, optionally, the information acquisition module 410 may also be used to:

[0101] The target area is divided into multiple grids, and the panchromatic remote sensing images and multispectral images corresponding to each grid are obtained.

[0102] Based on any optional technical solution in the embodiment of the present invention, optionally, the target image generation module 420 further includes:

[0103] A feature stitching subunit is used to stitch the panchromatic remote sensing image and the multispectral image corresponding to each grid into the feature extraction module to obtain a stitching feature image corresponding to each grid;

[0104] A residual learning subunit, configured to input the spliced feature map into the residual learning module for residual learning to obtain a target fused image corresponding to each grid;

[0105] The target image generation subunit is used to splice the target fusion images corresponding to each grid to obtain the target remote sensing image of the target area.

[0106] Based on any optional technical solution in the embodiment of the present invention, optionally, the residual learning module includes a convolution unit and a jump connection unit,

[0107] The residual learning module outputs the target fused image based on the following formula:

[0108] F b =CA(X b )+F b-1

[0109] Among them, F b represents the target fusion image corresponding to any grid, F b-1 represents the concatenated feature map, X b represents the initial feature, X b By F b-1It is obtained by performing secondary convolution on the convolution unit, and CA(·) represents the attention mechanism function.

[0110] Based on any optional technical solution in the embodiment of the present invention, optionally, any preset planned route includes multiple load points, and the target solution determination module 430 further includes:

[0111] A line mapping unit, configured to map a plurality of load points in the preset planned line to a target remote sensing map, to obtain a target remote sensing map including the preset planned line;

[0112] A load scoring unit is configured to score the coordinates of each load point of any preset planned route in the target remote sensing map containing the preset planned route, and obtain a score corresponding to each load point in the preset planned route;

[0113] The scheme determination unit is used to screen the preset planned routes based on the scores corresponding to the load points in the preset planned routes to determine the target planning scheme.

[0114] Based on any optional technical solution in the embodiments of the present invention, optionally, the load scoring unit may be specifically configured to:

[0115] In the target remote sensing map containing the preset planned route, identifying the terrain category corresponding to the coordinates of each load point;

[0116] The score corresponding to each load point is determined based on the terrain category corresponding to the coordinates of each load point.

[0117] Based on any optional technical solution in the embodiments of the present invention, optionally, the solution determination unit may be specifically configured to:

[0118] Add up the scores corresponding to each load point in any preset planned route to obtain the planned route score;

[0119] The preset planned route with the largest planned route score among the preset planned routes is determined as the target planning scheme.

[0120] The route planning device provided in the embodiment of the present invention can execute the route planning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0121] Example 5

[0122] Figure 5 This is a structural diagram of an electronic device provided in Example 5 of the present invention. Figure 5 A block diagram of an exemplary electronic device 12 suitable for implementing embodiments of the present invention is shown. Figure 5The electronic device 12 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present invention.

[0123] like Figure 5 As shown, electronic device 12 is implemented as a general-purpose computing device. Components of electronic device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, and a bus 18 that connects various system components (including system memory 28 and processing unit 16).

[0124] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0125] The electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 12, including volatile and non-volatile media, removable and non-removable media.

[0126] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 5 Not shown, often called a "hard drive"). Although Figure 5 Not shown, a magnetic disk drive for reading and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0127] A program / utility 36 having a set (at least one) of program modules 26 may be stored, for example, in system memory 28. Such program modules 26 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 26 generally perform the functions and / or methods of the embodiments described herein.

[0128] The electronic device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, a pointing device, a display 24, etc.), one or more devices that enable a user to interact with the electronic device 12, and / or any device that enables the electronic device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the electronic device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the electronic device 12 via the bus 18. Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0129] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing a route planning method provided in an embodiment of the present invention.

[0130] Example 6

[0131] Embodiment 6 of the present invention further provides a storage medium containing computer-executable instructions, wherein the computer-executable instructions, when executed by a computer processor, are used to perform a route planning method, the method comprising:

[0132] Acquiring multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image;

[0133] Inputting the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area;

[0134] The plurality of preset planned routes are respectively mapped to the target remote sensing map to obtain a target remote sensing map containing the preset planned routes, and each preset planned route is screened in the target remote sensing map containing the preset planned routes to determine a target planning scheme.

[0135] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0136] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0137] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0138] The computer program code for performing the operations of the embodiments of the present invention can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0139] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A route planning method, characterized in that: include: Acquiring multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; Inputting the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; Mapping the plurality of preset planned routes to the target remote sensing map respectively to obtain a target remote sensing map containing the preset planned routes, and screening each preset planned route in the target remote sensing map containing the preset planned routes to determine a target planning scheme; Wherein, any of the preset planned routes includes multiple load points, and mapping the multiple preset planned routes to the target remote sensing map to obtain a target remote sensing map including the preset planned routes, and screening each preset planned route in the target remote sensing map including the preset planned routes to determine a target planning scheme, including: Mapping multiple load points in the preset planned route to a target remote sensing map to obtain a target remote sensing map containing the preset planned route; In the target remote sensing map containing the preset planned routes, scoring the coordinates of each load point of any preset planned route to obtain a score corresponding to each load point in the any preset planned route; The preset planning routes are screened based on the scores corresponding to the load points in the preset planning routes to determine the target planning scheme.

2. The method according to claim 1, characterized in that The acquiring of multi-source remote sensing information of the target area includes: The target area is divided into multiple grids, and the panchromatic remote sensing images and multispectral images corresponding to each grid are obtained.

3. The method according to claim 2, characterized in that The route planning model includes a feature extraction module and a residual learning module. The panchromatic remote sensing image and the multispectral image are input into the pre-trained route planning model to obtain a target remote sensing image of the target area, including: Inputting the panchromatic remote sensing image and the multispectral image corresponding to each grid into the feature extraction module to obtain the splicing feature map corresponding to each grid; Inputting the spliced feature map into the residual learning module for residual learning to obtain the target fusion image corresponding to each grid; The target fusion images corresponding to the grids are spliced together to obtain a target remote sensing image of the target area.

4. The method according to claim 3, characterized in that The residual learning module includes a convolution unit and a jump connection unit. The residual learning module outputs a target fused image based on the following formula: F b =CA(X b )+F b-1 Among them, F b represents the target fusion image corresponding to any grid, F b-1 represents the concatenated feature map, X b represents the initial feature, X b By F b-1 It is obtained by performing secondary convolution on the convolution unit, and CA(·) represents the attention mechanism function.

5. The method according to claim 1, wherein In the target remote sensing map including the preset planned route, scoring the coordinates of each load point of any preset planned route to obtain the score corresponding to each load point in the preset planned route includes: In the target remote sensing map containing the preset planned route, identifying the terrain category corresponding to the coordinates of each load point; The score corresponding to each load point is determined based on the terrain category corresponding to the coordinates of each load point.

6. The method according to claim 1, characterized in that The screening of the preset planned routes based on the scores corresponding to the load points in the preset planned routes to determine the target planning scheme includes: Add up the scores corresponding to each load point in any preset planned route to obtain the planned route score; The preset planned route with the largest planned route score among the preset planned routes is determined as the target planning scheme.

7. A route planning device, characterized in that: include: An information acquisition module is used to acquire multi-source remote sensing information of a target area and a plurality of preset planned routes, wherein the multi-source remote sensing information includes a panchromatic remote sensing image and a multispectral image; a target image generation module, configured to input the panchromatic remote sensing image and the multispectral image into a pre-trained route planning model to obtain a target remote sensing image of the target area; a target solution determination module, configured to map the plurality of preset planned routes to the target remote sensing map, obtain a target remote sensing map containing the preset planned routes, and screen each preset planned route in the target remote sensing map containing the preset planned routes to determine a target planning solution; Wherein, any of the preset planned routes includes multiple load points, and the target solution determination module includes a route mapping unit, a load scoring unit, and a solution determination unit; The line mapping unit is used to map multiple load points in the preset planned line to a target remote sensing map to obtain a target remote sensing map containing the preset planned line; The load scoring unit is configured to score the coordinates of each load point on any preset planned route in the target remote sensing map containing the preset planned route, and obtain a score corresponding to each load point on the preset planned route; The scheme determination unit is used to screen the preset planned routes based on the scores corresponding to the load points in the preset planned routes to determine the target planning scheme.

8. An electronic device, characterized in that: The electronic device comprises: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the route planning method according to any one of claims 1 to 6.

9. A storage medium containing computer-executable instructions, characterized in that: When the computer executable instructions are executed by a computer processor, they are used to perform the route planning method according to any one of claims 1 to 6.

Images