A mountainous terrain measurement system and method based on remote sensing technology

Abstract

The application discloses a mountainous area topography measuring system and method based on remote sensing technology, and belongs to the technical field of topography measurement. The system comprises a remote sensor, a remote sensing platform, an information transmission module, a ground receiving platform, an image processing module and a feature extraction module. Feature information of ground object properties and states is extracted, and the extracted feature information is clustered. A topography splicing module is arranged to splice the clustered feature information of ground object properties and states, thereby forming a mountainous area topography measurement map. The application can accurately and objectively measure the mountainous area topography, greatly improve the remote sensing image information processing speed and the efficiency of measuring the mountainous area topography by means of remote sensing technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of topographic surveying, in particular to a mountainous terrain surveying system and method based on remote sensing technology. BACKGROUND

[0002] Remote sensing technology is a detection technology that emerged in the 1960s. According to the theory of electromagnetic waves, various sensing instruments are used to collect, process and finally image the electromagnetic wave information radiated and reflected by distant targets, so as to detect and identify various ground objects. Through remote sensing technology, domestic high-resolution remote sensing images such as Gaofen-1, Gaofen-2 and Resource-3 can be queried.

[0003] A groundwater level observation device for hydrogeological exploration based on remote sensing technology is disclosed in Chinese Patent No. CN114563062A. When the borehole diameter is blocked by soil and stone, the detection rod can move downward while rotating the hole, so as to move the detection rod to the bottom of the borehole and accurately detect the water level inside. When the detection rod moves upward, it can ensure stable upward movement of the detection rod.

[0004] Although the patent solves the problem that the soil at the drilling site is prone to collapse due to changes in the terrain, which affects the water level observation at the bottom of the borehole, it still has the following defects:

[0005] When remote sensing technology is used to measure the terrain of mountainous areas in the prior art, the remote sensing image information processing speed is slow, resulting in low efficiency of measuring the terrain of mountainous areas with the aid of remote sensing technology. SUMMARY

[0006] The present application aims to provide a mountainous terrain surveying system and method based on remote sensing technology, which can accurately and objectively measure the terrain of mountainous areas, greatly improve the remote sensing image information processing speed and the efficiency of measuring the terrain of mountainous areas with the aid of remote sensing technology, to solve the problems raised in the background technology.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A mountainous terrain surveying system based on remote sensing technology, comprising:

[0009] A remote sensor for detecting electromagnetic waves radiated or reflected by ground objects and the environment at a distance, collecting the received electromagnetic waves through a collection system, and then transmitting them to a detection system. After signal conversion, the remote sensing image information is output in an appropriate manner;

[0010] The remote sensing platform is used for placing remote sensors in the air or space, and is a carrier for carrying remote sensors in the remote sensing process.

[0011] The information transmission module is used for connecting the remote sensor and the ground receiving platform, and is a tool for transmitting information between the remote sensor and the ground receiving platform.

[0012] The ground receiving platform is used for receiving remote sensing image information transmitted from the remote sensor.

[0013] The image processing module is used for processing the remote sensing image information received by the ground to obtain characteristic information reflecting the properties and states of ground objects.

[0014] The feature extraction module is used for extracting features of the properties and states of ground objects, and clustering the extracted features of the properties and states of ground objects.

[0015] The terrain splicing module is used for splicing the clustered features of the properties and states of ground objects to form a mountainous terrain survey map.

[0016] Further, the information transmission module includes an information sending unit and an information receiving unit, and the information sending unit and the information receiving unit communicate through a wireless network.

[0017] The information sending unit is used for sending remote sensing image information transmitted from the remote sensor to the ground receiving platform.

[0018] The information receiving unit is used for receiving remote sensing image information transmitted from the remote sensor to the ground receiving platform.

[0019] Further, the image processing module includes a data management unit and a correction processing unit, and the data management unit and the correction processing unit communicate through a wireless network.

[0020] The data management unit is used for performing photographic processing on the remote sensing image information received by the ground receiving platform, and after the photographic processing, performing transformation and digitization processing on the remote sensing image information, so as to convert it into computer-compatible remote sensing image information, and transmitting the remote sensing image information to the correction processing unit.

[0021] The correction processing unit is used for performing geometric correction and radiation correction on system errors of the geometric shape and position error of the remote sensing image, and image radiation intensity information error, wherein geometric distortion caused by random attitude error and scanning speed error of the remote sensing platform is eliminated, which is called geometric fine correction; distortion caused by different scattering of light of different spectral segments through the atmosphere is eliminated, which is called atmospheric correction.

[0022] The correction processing unit corrects the remote sensing image information, so that the remote sensing image information after processing can more truly show the real appearance of the original scene.

[0023] Further, the feature extraction module comprises a feature reading unit, an analysis division unit and a category storage unit,

[0024] The feature reading unit is configured to read features of the processed remote sensing image information, wherein the features of the remote sensing image information include feature information of the properties and states of the ground objects.

[0025] The analysis division unit is configured to analyze and divide the read features of the remote sensing image information, so that the same features of the remote sensing image information are classified into one category.

[0026] The category storage unit is configured to store various standard feature categories of the remote sensing image information, for reference when analyzing and dividing the read features of the remote sensing image information.

[0027] Further, when clustering the extracted feature information of the properties and states of the ground objects, the following operations are performed:

[0028] Obtaining the feature information of the properties and states of the ground objects of the remote sensing image information;

[0029] Determining a feature set of the properties and states of the ground objects according to the obtained feature information of the properties and states of the ground objects;

[0030] Analyzing and dividing the feature information of the properties and states of the ground objects in the feature set according to the feature set;

[0031] Reading the feature information of the properties and states of the ground objects in the feature set one by one during the analysis and division;

[0032] Comparing the read single feature information of the properties and states of the ground objects with various standard feature categories of the remote sensing image information stored in the category storage unit;

[0033] If the read feature information is the same as the stored standard feature category, the category of the read feature information is the stored standard feature category;

[0034] If the read feature information is different from the stored standard feature category, the category of the read feature information is not the stored standard feature category;

[0035] If the category of the read feature information is not the stored standard feature category, the read feature information is compared with the next stored standard feature category, and the comparison is repeated until the corresponding standard feature category of the read feature information is found.

[0036] Further, the terrain splicing module comprises a feature combination unit, a scene generation unit and a scene splicing unit,

[0037] The feature combination unit is configured to combine the feature information of the classified ground object properties and states, and determine a combination set of the feature information of the ground object properties and states.

[0038] The scene generation unit is configured to generate corresponding scene images from the combined feature information of the ground object properties and states, and determine a scene image set of the mountainous terrain.

[0039] The scene splicing unit is configured to splice multiple scene images in the scene image set to form a panoramic mountainous terrain measurement map.

[0040] Further, the remote sensing platform comprises an edge server, a self-encoder, a network accelerator and a communication device; the edge server is connected to the self-encoder, the network accelerator and the communication device respectively; wherein,

[0041] The edge server is configured to form an edge remote sensing network through the communication device; wherein,

[0042] The communication device comprises a satellite communication device, a GPRS communication device and an optical fiber communication port, and any communication network can be selected;

[0043] The edge server is configured with a virtual switching device, which is configured to group the remote sensing image information received by the edge server according to the ground object features, and generate corresponding flow entries; wherein,

[0044] The ground object features include geology, geomorphology, soil, vegetation, hydrology and artificial structures.

[0045] The self-encoder is configured to encode the high-latitude original data corresponding to each flow entry into low-latitude hidden variables.

[0046] The noise enhancement network and the signal enhancement network built-in the network accelerator are used to denoise and enhance the low-latitude hidden variables, and the enhanced remote sensing image is transmitted to the cloud platform of the terrain measurement through the communication device.

[0047] Further, the denoising and enhancement processing comprises the following steps:

[0048] Step 1: Obtain the low-latitude hidden variables corresponding to each flow entry, and determine the denoising loss function based on the self-supervised learning of the noise enhancement network:

[0049]

[0050] Wherein, S represents the denoising loss function of the low-latitude hidden variable; M represents the total number of low-latitude hidden variables; i is a positive integer, i∈M; c i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q

[0051] Step 2: Transform and enhance the low-latitude hidden variable through the denoising loss function:

[0052]

[0053] Wherein, f(c) represents the image scale parameter of the enhanced low-latitude hidden variable; d(c i ) represents the scale enhancement function of the i th low-latitude hidden variable transformed and enhanced from the starting scale; μ(c i ) represents the mapping distribution function of the i th low-latitude hidden variable; γ(c i ) represents the reconstruction enhancement function of the i th low-latitude hidden variable wavelet transformed and scale enhanced;

[0054] Step 3: Gradient transform the transformed and enhanced image scale parameter through the following formula to generate the enhanced remote sensing image:

[0055]

[0056] Wherein, T represents the gradient transformed remote sensing image; B represents the minimum value of the gradient modulus during gradient transformation; A represents the maximum value of the gradient modulus during gradient transformation.

[0057] According to another aspect of the present application, a mountain terrain measurement method of a mountain terrain measurement system based on remote sensing technology is provided, comprising the following steps:

[0058] S10: Detecting the electromagnetic waves radiated or reflected by the ground objects and environment at a distance by using the remote sensor mounted on the remote sensing platform, and outputting the remote sensing image information in a proper way;

[0059] S20: Transmitting the remote sensing image information detected by the remote sensor to the ground receiving platform in a wireless network transmission manner by using the information transmission module;

[0060] S30: Processing the remote sensing image information received by the ground to obtain the characteristic information reflecting the properties and states of the ground objects by using the image processing module;

[0061] S40: The feature extraction module is used for feature extraction of the feature information of the ground object properties and states, and the extracted feature information of the ground object properties and states is clustered;

[0062] S50: The terrain splicing module is used for terrain splicing of the clustered feature information of the ground object properties and states, so as to form a mountain terrain measurement map.

[0063] Compared with the prior art, the present application has the following beneficial effects:

[0064] 1. The mountain terrain measurement system and method based on remote sensing technology of the present application can accurately and objectively measure the mountain terrain by using the remote sensor carried on the remote sensing platform to remotely detect the electromagnetic waves radiated or reflected by the ground object and the environment, outputting the remote sensing image information in a proper manner, transmitting the remote sensing image information detected by the remote sensor to the ground receiving platform in a wireless network transmission manner, processing the remote sensing image information received by the ground to obtain the feature information reflecting the properties and states of the ground object, extracting the feature information of the properties and states of the ground object, clustering the extracted feature information of the properties and states of the ground object, splicing the clustered feature information of the properties and states of the ground object, and forming a mountain terrain measurement map.

[0065] 2. When the feature information of the properties and states of the ground object extracted by the present application is clustered, the feature information of the properties and states of the ground object obtained from the remote sensing image information is obtained, the feature set of the properties and states of the ground object is determined according to the obtained feature information of the properties and states of the ground object, the feature information of the properties and states of the ground object in the feature set is analyzed and divided according to the feature set, the feature information of the properties and states of the ground object in the feature set is read one by one during the analysis and division, the read single feature information of the properties and states of the ground object is compared with various standard feature categories of the remote sensing image information stored in the category storage unit, the same remote sensing image information features are classified into one category, and the remote sensing image information processing speed and the efficiency of measuring the mountain terrain by means of remote sensing technology can be greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 The module diagram of the mountain terrain measurement system based on remote sensing technology of the present application;

[0067] Figure 2 The architecture diagram of the information transmission module of the present application;

[0068] Figure 3 The architecture diagram of the image processing module of the present application;

[0069] Figure 4 The architecture diagram of the feature extraction module of the present application;

[0070] Figure 5 A flowchart for clustering the feature information of the extracted ground object properties and states of the present application;

[0071] Figure 6 An architecture diagram of the terrain splicing module of the present application. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0073] In order to solve the technical problem that the existing remote sensing technology for measuring the mountain terrain has a slow remote sensing image information processing speed, resulting in low efficiency of measuring the mountain terrain by means of the remote sensing technology, please refer to Figures 1-6 The technical solutions are provided in the embodiments as follows:

[0074] A mountain terrain measurement system based on remote sensing technology, comprising:

[0075] A remote sensor is used for remotely detecting electromagnetic waves radiated or reflected by ground objects and environment, collecting the received electromagnetic waves by means of a collection system, and then transmitting the electromagnetic waves to a detection system, and outputting remote sensing image information in a proper way after signal conversion.

[0076] It should be noted that the remote sensor is divided according to the frequency or wave band selected when designed. Commonly used remote sensors include ultraviolet remote sensors, visible light remote sensors, and infrared remote sensors, etc.

[0077] The ultraviolet remote sensor uses a near ultraviolet wave band, and the wavelength is selected in the range of 0.3-0.4 micrometers. Commonly used ultraviolet remote sensors include ultraviolet cameras and ultraviolet scanners. The multispectral camera of the near ultraviolet wave band also belongs to this category.

[0078] The visible light remote sensor receives visible light reflected by ground objects, and the wavelength is selected in the range of 0.38-0.76 micrometers. This kind of remote sensor includes various conventional cameras, as well as multispectral cameras, multispectral scanners, and charge-coupled device (CCD) scanners of the visible light wave band, etc. In addition, it also includes laser altimeters and laser scanners of the visible light wave band, etc.

[0079] The infrared remote sensor receives electromagnetic waves of the infrared wave band radiated or reflected by ground objects and environment. The used wave band is about in the range of 0.7-14 micrometers, in which the wavelength of 0.7-2.5 micrometers is called a reflected infrared wave band.

[0080] Remote sensing platform, for placing remote sensor in the air or space, is a carrier tool of remote sensor in remote sensing process;

[0081] It should be noted that the remote sensing platform is a carrier tool of remote sensor in remote sensing process, which is like a tripod for placing a camera when taking photos on the ground, and is a device for placing remote sensor in the air or space. The main remote sensing platforms include high-altitude balloon, airplane, rocket, artificial satellite, manned spacecraft, etc.

[0082] Information transmission module, for connecting remote sensor and ground receiving platform, is a tool for transmitting information between remote sensor and ground receiving platform;

[0083] Ground receiving platform, for receiving remote sensing image information transmitted from remote sensor;

[0084] Image processing module, for processing remote sensing image information received on the ground to obtain characteristic information reflecting the properties and states of ground objects;

[0085] Feature extraction module, for extracting features of the properties and states of ground objects, and clustering the extracted features of the properties and states of ground objects;

[0086] Terrain splicing module, for splicing the clustered features of the properties and states of ground objects to form a mountain terrain measurement map.

[0087] Specifically, the remote sensor carried on the remote sensing platform detects electromagnetic waves radiated or reflected by ground objects and environment at a long distance, and outputs remote sensing image information in a proper way. The information transmission module transmits the remote sensing image information detected by the remote sensor to the ground receiving platform in a wireless network transmission manner. The image processing module processes the remote sensing image information received on the ground to obtain characteristic information reflecting the properties and states of ground objects. The feature extraction module extracts features of the properties and states of ground objects, and clusters the extracted features of the properties and states of ground objects. The terrain splicing module splices the clustered features of the properties and states of ground objects to form a mountain terrain measurement map, which can accurately and objectively measure the mountain terrain.

[0088] The information transmission module includes an information sending unit and an information receiving unit, and the information sending unit and the information receiving unit communicate through a wireless network;

[0089] The information sending unit is used for sending remote sensing image information transmitted from the remote sensor to the ground receiving platform;

[0090] The information receiving unit is used for receiving remote sensing image information transmitted from the remote sensor to the ground receiving platform.

[0091] The image processing module comprises a data management unit and a correction processing unit, and the data management unit and the correction processing unit communicate through a wireless network;

[0092] The data management unit is used for photographic processing of remote sensing image information received by a ground receiving platform, and after the photographic processing, the remote sensing image information is transformed and digitally processed to be converted into computer-compatible remote sensing image information, and the remote sensing image information is transmitted to the correction processing unit;

[0093] The correction processing unit is used for geometric correction and radiation correction of system errors of the remote sensing image information, such as geometric shape and position errors of the remote sensing image, and image radiation intensity information errors, wherein geometric distortion caused by random attitude errors and scanning speed errors of the remote sensing platform is eliminated, which is called geometric fine correction; distortion caused by different scattering of light of different spectral bands when passing through the atmosphere is eliminated, which is called atmospheric correction;

[0094] After the correction processing unit performs correction processing on the remote sensing image information, the processed remote sensing image information can more truly show the real appearance of the original scene.

[0095] The feature extraction module comprises a feature reading unit, an analysis and division unit, and a category storage unit,

[0096] The feature reading unit is used for feature reading of the processed remote sensing image information, wherein the feature of the remote sensing image information includes feature information of the property and state of the ground object;

[0097] The analysis and division unit is used for analysis and division of the read feature of the remote sensing image information, so that the same feature of the remote sensing image information is classified into a category;

[0098] The category storage unit is used for storing various standard feature categories of the remote sensing image information, which is used for reference when the read feature of the remote sensing image information is analyzed and divided.

[0099] When the extracted feature information of the property and state of the ground object is clustered, the following operations are performed:

[0100] The feature information of the property and state of the ground object of the remote sensing image information is obtained;

[0101] According to the obtained feature information of the property and state of the ground object, a feature set of the property and state of the ground object is determined;

[0102] According to the feature set, the feature information of the property and state of the ground object in the feature set is analyzed and divided;

[0103] When the analysis and division are performed, the feature information of the property and state of the ground object in the feature set is read one by one;

[0104] The read single feature information of the ground object property and state is compared with various standard feature categories of the remote sensing image information stored in the category storage unit;

[0105] If the read feature information is the same as the stored standard feature category, the category of the read feature information is the stored standard feature category;

[0106] If the read feature information is different from the stored standard feature category, the category of the read feature information is not the stored standard feature category;

[0107] If the category of the read feature information is not the stored standard feature category, the read feature information is compared with the next stored standard feature category, and the cycle is repeated until the corresponding standard feature category of the read feature information is found.

[0108] The terrain splicing module includes a feature combination unit, a scene generation unit and a scene splicing unit,

[0109] The feature combination unit is used to associate and combine the feature information of the ground object property and state after classification, and determine the associated combination set of the feature information of the ground object property and state;

[0110] The scene generation unit is used to generate corresponding scene images from the associated and combined feature information of the ground object property and state, and determine the scene image set of the mountainous terrain;

[0111] The scene splicing unit is used to splice multiple scene images in the scene image set to form a panoramic mountainous terrain measurement map.

[0112] Further, the remote sensing platform comprises an edge server, a self-encoder, a network accelerator and a communication device; the edge server is connected with the self-encoder, the network accelerator and the communication device respectively; wherein,

[0113] The edge server is used to form an edge remote sensing network through the communication device; wherein,

[0114] The communication device comprises a satellite communication device, a GPRS communication device and an optical fiber communication port, and any communication network is selected;

[0115] The edge server is configured with a virtual switching device, which is used to group the remote sensing image information received by the edge server according to the ground object features, and generate corresponding flow entries; wherein,

[0116] The ground object features include geology, geomorphology, soil, vegetation, hydrology and artificial structures;

[0117] The self-encoder is used to encode the high-latitude original data corresponding to each flow entry into low-latitude hidden variables;

[0118] And through the built-in noise enhancement network and signal enhancement network in the network accelerator, the low-latitude hidden variables are denoised and enhanced, and the enhanced remote sensing image is transmitted to the cloud platform of topographic survey through the communication device.

[0119] The principle of the above technical solution is that the present application is a scheme for uploading and obtaining remote sensing images through a remote sensing platform, therefore, whether the remote sensing platform can perform network transmission and whether it can more accurately upload remote sensing images to the cloud are very important. In the prior art, after obtaining the remote sensing image, it is directly transmitted, but the directly obtained remote sensing image may have noise due to the characteristics of the remote sensor, that is, when the remote sensor performs remote measurement of electromagnetic waves, the generated remote sensing image will have certain noise, which is the fluctuation noise of electromagnetic waves due to weather conditions or environmental changes. For example, when remote sensing detection is performed, if there is a strong wind, the reflection of electromagnetic waves will be lost to a certain extent, therefore, certain denoising and enhancement are needed to transmit more accurate remote sensing images. The present application designs a remote sensing platform, and the edge server of the remote sensing platform is used to enhance the collected remote sensing image, so as to be transmitted to the cloud platform. When denoising and enhancing, the noise is removed, and the speed of transmitting the remote sensing image is also enhanced. In this process, the present application designs various communication devices, which can all transmit remote sensing images. The most important thing in the transmission process is the grouping processing of the remote sensing image. In the present application, because there is an edge server, the remote sensing image can be divided into different flow entries according to the features of the ground objects for variable coding processing. In the prior art, the image is directly enhanced, and the direct processing of the image can speed up, but does not have the security of the remote sensing image and the rigor of the image logic. The present application is provided with a self-encoder, after coding, all data can be highly quantized, from high-latitude data to low-latitude data to reduce the processing difficulty. Then, the noise enhancement network and the signal enhancement network are used for denoising and enhancing, the purpose of noise enhancement is to determine the loss function caused by noise, and then the signal is enhanced through the loss function, so as to be more quickly transmitted to the cloud platform.

[0120] Further, the denoising and enhancing processing includes the following steps:

[0121] Step 1: Obtain the low-latitude hidden variable corresponding to each flow entry, determine the denoising loss function based on the self-supervised learning of the noise enhancement network:

[0122]

[0123] Wherein, S represents the denoising loss function of the low-latitude hidden variable; M represents the total number of low-latitude hidden variables; i is a positive integer, i∈M; c i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i represents the i th low-latitude hidden variable; x represents the training parameter of the noise enhancement network; q i ; x) represents the predicted noise distribution of the i th low-latitude hidden variable; z represents the denoising enhancement network;

[0124] Step 2: Transform and enhance the low-latitude hidden variable through the denoising loss function:

[0125]

[0126] Wherein, f(c) represents the image scale parameter of the enhanced low-latitude hidden variable; d(c i ) represents the scale enhancement function of the i th low-latitude hidden variable from the starting scale; μ(c i ) represents the mapping distribution function of the i th low-latitude hidden variable; γ(c i ) represents the reconstruction enhancement function of the wavelet transform and scale enhancement of the i th low-latitude hidden variable.

[0127] Step 3: Gradient transform the image scale parameter after transformation and enhancement through the following formula to generate the enhanced remote sensing image:

[0128]

[0129] Wherein, T represents the gradient transformed remote sensing image; B represents the minimum value of the gradient modulus during gradient transformation; A represents the maximum value of the gradient modulus during gradient transformation.

[0130] In the process of denoising enhancement, the present application has undergone three steps, determining the loss function caused by noise, transforming and enhancing the hidden variable through the loss function, and finally gradient transforming the hidden variable after transformation and enhancement in the form of image scale parameter to form the enhanced remote sensing image.

[0131] In step 1: The present application predicts the noise distribution of each low-dimensional hidden variable, determines the distribution of noise, subtracts the dynamic variable value that needs to be denoised, and thus obtains the denoising loss function.

[0132] In step 2: After the de-noising loss function is determined, the hidden variable is enhanced, in which process, the actual value of scale enhancement under the loss function is calculated first, then the transformed value of wavelet transform and scale enhancement under the mapping distribution is added, that is, the enhanced value + the transformed value, to obtain the final enhancement result. In step 3, the enhanced result is transformed through gradient to generate the final remote sensing image, in which process, the maximum and minimum values of gradient transformation are determined to realize multi-level transformation to obtain the final enhanced remote sensing image.

[0133] In order to better show the mountain terrain measurement process of the mountain terrain measurement system based on remote sensing technology, the embodiment proposes a mountain terrain measurement method of the mountain terrain measurement system based on remote sensing technology, which comprises the following steps:

[0134] S10: The remote sensor mounted on the remote sensing platform remotely detects the electromagnetic waves radiated or reflected by the ground objects and the environment, and outputs the remote sensing image information in a proper manner;

[0135] S20: The information transmission module transmits the remote sensing image information detected by the remote sensor to the ground receiving platform in a wireless network transmission manner;

[0136] S30: The image processing module processes the remote sensing image information received by the ground to obtain characteristic information reflecting the properties and states of the ground objects;

[0137] S40: The characteristic extraction module extracts the characteristic information of the properties and states of the ground objects, and clusters the extracted characteristic information of the properties and states of the ground objects;

[0138] S50: The terrain splicing module splices the clustered characteristic information of the properties and states of the ground objects to form a mountain terrain measurement map.

[0139] In summary, the mountainous terrain measurement system and method based on remote sensing technology utilizes remote sensors on a remote sensing platform to remotely detect electromagnetic waves radiated or reflected by ground objects and the environment, and outputs remote sensing image information in an appropriate manner. The remote sensing image information detected by the remote sensors is transmitted to a ground receiving platform in a wireless network transmission manner using an information transmission module. The remote sensing image information received on the ground is processed using an image processing module to obtain characteristic information reflecting the properties and states of ground objects. The characteristic information of the properties and states of ground objects is extracted using a feature extraction module, and the extracted characteristic information of the properties and states of ground objects is clustered. The clustered characteristic information of the properties and states of ground objects is terrain spliced using a terrain splicing module to form a mountainous terrain measurement map. The mountainous terrain can be accurately and objectively measured. When the extracted characteristic information of the properties and states of ground objects is clustered, the characteristic information of the properties and states of ground objects obtained from the remote sensing image information is determined based on the obtained characteristic information of the properties and states of ground objects. A feature set of the properties and states of ground objects is determined based on the characteristic information. The characteristic information of the properties and states of ground objects in the feature set is analyzed and divided based on the feature set. When the characteristic information of the properties and states of ground objects in the feature set is read one by one, the read single characteristic information of the properties and states of ground objects is compared with various standard feature categories of remote sensing image information stored in a category storage unit. The same remote sensing image information features are classified into a category, which can greatly improve the remote sensing image information processing speed and the efficiency of measuring mountainous terrain with the aid of remote sensing technology.

[0140] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can make equivalent replacements or changes to the technical solution and inventive concept of the present application within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A mountainous terrain surveying system based on remote sensing technology, characterized by, It comprises: a remote sensor for remotely detecting electromagnetic waves radiated or reflected by ground objects and the environment, collecting the received electromagnetic waves through a collection system, and then transmitting them to a detection system, which outputs remote sensing image information after signal conversion; a remote sensing platform for placing the remote sensor in the air or space; an information transmission module for connecting the remote sensor and the ground receiving platform; a ground receiving platform for receiving remote sensing image information transmitted from the remote sensor; an image processing module for processing the remote sensing image information received by the ground to obtain characteristic information reflecting the properties and states of ground objects; a feature extraction module for extracting features of the properties and states of ground objects, and clustering the extracted features of the properties and states of ground objects; a terrain splicing module for splicing the clustered features of the properties and states of ground objects to form a mountainous terrain survey map; The remote sensing platform comprises an edge server, a self-encoder, a network accelerator and a communication device; the edge server is connected with the self-encoder, the network accelerator and the communication device; wherein, The edge server is used to form an edge remote sensing network through the communication device; wherein, The communication device is composed of a satellite communicator, a GPRS communication device and an optical fiber communication port, and any communication network is selected; A virtual switching device is configured in the edge server, which is used to group the remote sensing image information received by the edge server according to ground object features, and generate corresponding flow entries; wherein, The ground object features include geology, topography, soil, vegetation, hydrology and artificial structures; The self-encoder is used to encode high-latitude original data corresponding to each flow entry into low-latitude hidden variables; And the low-latitude hidden variables are denoised and enhanced through the noise enhancement network and the signal enhancement network built in the network accelerator, and the enhanced remote sensing image is transmitted to the cloud platform of terrain measurement through the communication device; The denoising and enhancement processing includes the following steps: Step 1: Obtain the low-latitude hidden variables corresponding to each flow entry, determine the denoising loss function based on the self-supervised learning of the noise enhancement network: wherein, denotes a denoising loss function for the low-latitude hidden variables; denotes the total number of low-latitude hidden variables; is a positive integer, ; denotes the low-latitude hidden variable; denotes a training parameter of the noise augmentation network; denotes the low-latitude hidden variable after denoising; denotes the low-latitude hidden variable; denotes a denoising augmentation network; Step 2: Transform and enhance the low-latitude hidden variables through the denoising loss function: wherein, represents the image scale parameter of the enhanced hidden variable at low latitude; represents the scale enhancement function of the enhanced hidden variable at low latitude; represents the scale enhancement function of the enhanced hidden variable at low latitude; represents the mapping distribution function of the enhanced hidden variable at low latitude; represents the mapping distribution function of the enhanced hidden variable at low latitude; represents the reconstruction enhancement function of the enhanced hidden variable at low latitude; represents the reconstruction enhancement function of the enhanced hidden variable at low latitude; Step 3: Perform gradient transformation on the image scale parameters after transformation and enhancement through the following formula to generate the enhanced remote sensing image: wherein denotes the gradient transformed remote sensing image; denotes the minimum value of the gradient modulus when the gradient is transformed; A denotes the maximum value of the gradient modulus when the gradient is transformed.

2. The mountainous terrain measuring system based on remote sensing technology according to claim 1, characterized in that, The information transmission module comprises an information sending unit and an information receiving unit, which communicate through a wireless network; The information sending unit is used to send remote sensing image information transmitted from the remote sensor to the ground receiving platform; The information receiving unit is used to receive remote sensing image information transmitted from the remote sensor to the ground receiving platform.

3. The mountainous terrain measuring system based on remote sensing technology according to claim 1, characterized in that, The image processing module comprises a data management unit and a correction processing unit, which communicate through a wireless network; The data management unit is used to perform photographic processing on the remote sensing image information received by the ground receiving platform, and after photographic processing, to transform and digitize the remote sensing image information so that it is converted into computer-compatible remote sensing image information, and to transmit the remote sensing image information to the correction processing unit; The correction processing unit is used for geometric correction and radiation correction on geometric shape and position error, image radiation intensity information error of the remote sensing image of the remote sensing image information. After the correction processing unit performs correction processing on the remote sensing image information, the processed remote sensing image information can more truly reflect the real appearance of the original scene.

4. The mountainous terrain measuring system based on remote sensing technology according to claim 1, wherein, The feature extraction module comprises a feature reading unit, an analysis and division unit and a category storage unit, The feature reading unit is used for reading features of the processed remote sensing image information, wherein the features of the remote sensing image information comprise feature information of the properties and states of the ground objects; The analysis and division unit is used for analyzing and dividing the read features of the remote sensing image information, so that the features of the remote sensing image information of the same kind are classified into one category; The category storage unit is used for storing various standard feature categories of the remote sensing image information, which are used for reference when the read features of the remote sensing image information are analyzed and divided.

5. The mountainous terrain measuring system based on remote sensing technology according to claim 4, characterized in that, When the extracted feature information of the properties and states of the ground objects is clustered, the following operations are performed: Obtain the feature information of the properties and states of the ground objects of the remote sensing image information; Determine a feature set of the properties and states of the ground objects according to the obtained feature information of the properties and states of the ground objects; Analyze and divide the feature information of the properties and states of the ground objects in the feature set according to the feature set; Read the feature information of the properties and states of the ground objects in the feature set one by one during the analysis and division; Compare the read single feature information of the properties and states of the ground objects with various standard feature categories of the remote sensing image information stored in the category storage unit; If the read feature information is the same as the stored standard feature category, the category of the read feature information is the stored standard feature category; If the read feature information is different from the stored standard feature category, the category of the read feature information is not the stored standard feature category; If the category of the read feature information is not the stored standard feature category, compare the read feature information with the next standard feature category stored, and repeat the comparison until the corresponding standard feature category of the read feature information is found.

6. The mountainous terrain measuring system based on remote sensing technology according to claim 1, wherein, The terrain splicing module comprises a feature combination unit, a scene generation unit and a scene splicing unit, The feature combination unit is used for associatively combining the classified feature information of the properties and states of the ground objects, and determining an associated combination set of the feature information of the properties and states of the ground objects; The scene generation unit is used for generating corresponding scene images from the associated combination of the feature information of the properties and states of the ground objects, and determining a scene image set of the mountainous terrain; The scene splicing unit is used for splicing multiple scene images in the scene image set to form a panoramic mountainous terrain measurement map.

7. A mountainous topographic survey method using a remote sensing technology-based mountainous topographic survey system according to any one of claims 1 to 6, characterized by, The method comprises the following steps: S10: remotely detect electromagnetic waves radiated or reflected by ground objects and environment by using a remote sensor mounted on a remote sensing platform, and output remote sensing image information in a proper manner; S20: transmit the remote sensing image information detected by the remote sensor to a ground receiving platform in a wireless network transmission manner by using an information transmission module; S30: process the remote sensing image information received by the ground to obtain feature information reflecting the properties and states of the ground objects by using an image processing module; S40: The feature extraction module is used to extract the feature information of the object property and state, and the extracted feature information of the object property and state is clustered; S50: The terrain splicing module is used to splice the clustered feature information of the object property and state, and a mountain terrain measurement map is formed.

Images