Industrial heat source production area identification method and device based on multi-band observation data

Abstract

The present application provides an industrial heat source production area identification method based on multi-band observation data, including: obtaining multi-band observation data of a target area; the multi-band observation data includes multi-band remote sensing satellite data obtained based on high-resolution thermal infrared sensors, multi-spectral sensors and low-light sensors; based on the spectral distribution of the industrial heat source production area, extracting multivariate indicator data reflecting the characteristics of the industrial heat source production area from the multi-band observation data; inputting the multivariate indicator data into an industrial heat source production area identification model to achieve industrial heat source production area identification of the target area; wherein the industrial heat source production area identification model is obtained by training according to a sample set using a random forest machine learning method; the sample set is constructed using high-resolution remote sensing satellite position information and prior industrial heat source production area characteristic indicator information. The present application satisfies the problem of high-precision industrial heat source production area identification better.

Description

Technical Field

[0001] The present application relates to the field of multi-band observation data processing and target discovery and identification, and in particular to a method and device for identifying industrial heat source production areas based on multi-band observation data. Background Art

[0002] In the process of increasingly close economic globalization, a large number of industrial production capacity clusters have emerged in the global industrial industry, which has greatly promoted the development of the global economy. However, in recent years, industrial heat source emissions have also become the main source of regional air pollution and anthropogenic C02 emissions. As a global industrial power, China needs to have independent high-temporal and spatial dynamic monitoring data on the regional distribution and scale characteristics of industrial heat source production in order to fully understand the allocation of industrial resources and monitor the state of the atmospheric environment. Therefore, how to balance development and emission needs and grasp the regional layout and scale characteristics of industrial heat source production in my country and even the world in near real time, objectively and accurately has become an important problem faced by my country in the process of promoting sustainable development.

[0003] The industrial / mining enterprise data obtained through inventory methods such as statistics and field surveys are of poor timeliness, strong subjectivity, and often have large errors, which makes it difficult to meet the needs of real-time discovery and near-real-time, accurate, and objective monitoring of industrial heat source areas in large areas. Recent studies have shown that the use of remote sensing thermal anomaly product data can characterize the production activities and production scale of industrial heat sources, and is the most effective method for monitoring the spatiotemporal layout and changes of industrial heat sources in large areas. At present, most studies use thermal anomaly data to roughly identify industrial heat source areas, but considering the low spatial resolution of the original remote sensing images of the inverted thermal anomaly data and the errors in the inversion process, it is impossible to accurately identify the production areas of industrial heat sources.

[0004] The Sustainable Development Science Satellite 1 (SDGSAT-1) was successfully launched on November 5, 2021. According to the plan, the SDGSAT-1 special application has begun for relevant scientific and technological personnel such as the "Earth Big Data Science Project" pilot project and the International Research Center for Sustainable Development Big Data (CBAS). It is equipped with three payloads: thermal infrared, low-light and multi-spectral imagers. Through the coordinated observation of the three payloads all day long, it has the advantage of high-resolution thermal infrared sensor data. Based on this new data, research in new fields has been carried out, which has broad application prospects. Summary of the invention

[0005] Based on the full use of high spatial resolution thermal infrared remote sensing data, this application focuses on the scientific and technological problem of industrial heat source production area identification accuracy, combines multi-sensor data preprocessing, multi-element industrial heat source production and operation status characterization index extraction, target recognition and other algorithms, and proposes an industrial heat source production area identification method and device based on multi-band observation data.

[0006] In a first aspect, the present application provides a method for identifying industrial heat source production areas based on multi-band data, comprising the following steps:

[0007] Acquire multi-band observation data of the target area; the multi-band observation data includes multi-band remote sensing satellite data obtained based on high-resolution thermal infrared sensors, multi-spectral sensors and low-light sensors;

[0008] Based on the spectral distribution of industrial heat source production areas, multivariate indicator data reflecting the characteristics of industrial heat source production areas are extracted from multi-band observation data;

[0009] The multivariate indicator data reflecting the characteristics of the industrial heat source production area is input into the industrial heat source production area identification model to realize the identification of the industrial heat source production area in the target area; wherein, the industrial heat source production area identification model is obtained by training according to the sample set using the random forest machine learning method; the sample set is constructed using high-resolution remote sensing satellite position information and prior industrial heat source production area characteristic indicator information.

[0010] Preferably, the acquired remote sensing satellite data needs to be analyzed and processed before being used as multi-band observation data, including:

[0011] For the thermal infrared sensor data, after radiation, the onboard brightness temperature of the thermal infrared sensor in three bands is inverted;

[0012] For the multi-spectral sensor data, after radiation, the reflectivity of the multi-spectral sensor bands B1-7 is inverted;

[0013] For the low-light sensor data, after irradiation, calculate the light radiation values ​​of the low-light sensor R, G, B, PL, and PH bands.

[0014] Preferably, the multivariate indicator information of the characteristics of the industrial heat source production area includes:

[0015] The reflectivity index, night light radiation value index, and radiation brightness temperature index refer to the reflectivity of the B1-7 band of the multi-spectral sensor, the night light radiation value of the R, G, B, PL, and PH5 bands in the low light sensor data, and the radiation on-board brightness temperature of the T1-3 band of the thermal infrared sensor.

[0016] The thermal radiation index refers to the six short-wave thermal radiation index RI characteristics extracted based on the characteristics of the industrial heat source production area:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023] Among them, T3, T2 and T1 are the radiation brightness temperatures of the thermal infrared sensor with central wavelengths of 11.81μm, 9.32μm and 10.82μm, respectively, and ρ7 and ρ6 are the reflectances of the multi-spectral sensor with central wavelengths of 850nm and 785nm.

[0024] Normalized difference water index NDWI and normalized difference vegetation index NDVI indicator information.

[0025] Preferably, the training step of the industrial heat source production area identification model includes:

[0026] Construct sample sets of industrial heat source production areas and background areas, and comprehensively utilize high-resolution remote sensing satellite data such as Google Earth and multivariate index feature information to build as many sample sets of industrial heat source production areas and background areas with uniform spatial distribution as possible;

[0027] An industrial heat source production area identification model with comprehensive multivariate index characteristics is constructed. The random forest machine learning method is adopted, and the multivariate index characteristics are used as input features. The random forest model for industrial heat source production area identification is trained on the training sample set to form an industrial heat source production area identification model with comprehensive multivariate index characteristics.

[0028] Preferably, the process of identifying the industrial heat source production area includes:

[0029] The multivariate index features of the research area extracted by the industrial heat source index feature extraction layer of the comprehensive multi-band data are used to realize the identification of the industrial heat source production area in the target area by using the industrial heat source production area identification model with comprehensive multivariate index features constructed by the industrial heat source production area identification model training layer;

[0030] Use cloud mask, NDVI, NDWI and other indicators to eliminate false industrial heat source production areas.

[0031] Preferably, after the industrial heat source production area of ​​the target area is identified, high-resolution remote sensing images, Google Earth data, POI data sets, local statistical data sets, and manual field survey data are also needed to conduct qualitative and quantitative analysis and verification of the identification accuracy of the industrial heat source production area in the study area by means of test sample verification, cross-validation, and manual verification of typical areas.

[0032] In a second aspect, the present application provides an industrial heat source production area identification device based on multi-band observation data, characterized in that the device comprises:

[0033] Multi-band data preprocessing module, which acquires thermal infrared, low light and multi-spectral sensor data from remote sensing satellites and performs analytical processing;

[0034] The industrial heat source index feature extraction module integrates multi-band data to extract the operating status index information of the industrial heat source production area from the data obtained through analytical processing;

[0035] A module for building a model for identifying industrial heat source production areas based on comprehensive multivariate index characteristics, building a sample set of industrial heat source production areas and background areas with uniform spatial distribution, training a random forest model for industrial heat source production area identification, and building a model for identifying industrial heat source production areas based on comprehensive multivariate index characteristics;

[0036] An industrial heat source production area identification module with comprehensive multivariate index characteristics combines the industrial heat source production area index information and the industrial heat source production area identification model to realize the industrial heat source production area identification in the studied area;

[0037] The industrial heat source production area identification result verification module performs qualitative and quantitative analysis and verification on the accuracy of the industrial heat source production area identification.

[0038] In a third aspect, the present application also provides an industrial heat source production area identification device based on multi-band data, comprising at least one processor, wherein the processor is used to execute a program stored in a memory, and when the program is executed, the device performs the various steps in the first aspect and various possible implementations.

[0039] In a fourth aspect, the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the various steps in the first aspect and various possible implementations.

[0040] This application comprehensively analyzes the spectral characteristics of industrial heat source production areas based on multi-band data, and extracts the following multivariate industrial heat source index characteristics: reflectivity index, night light radiation value index, radiation brightness temperature index, thermal radiation index, NDWI and NDVI indicator information. Among them, in view of the lack of short-wave infrared data in multi-band data, 6 new thermal infrared radiation indicators are proposed on the premise of fully investigating the distribution characteristics of industrial heat source production areas in the thermal infrared sensor spectrum. On the one hand, the multivariate industrial heat source index characteristics effectively improve the accuracy and comprehensiveness of the description of the information characteristics of industrial heat source production areas; on the other hand, the multivariate industrial heat source characteristics effectively improve the accuracy and reliability of the identification of industrial heat source production areas in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more simply illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 This is a flow chart of a method for identifying industrial heat source production areas based on multi-band observation data in an embodiment of the present application;

[0043] Figure 2 This is a structural framework diagram of an industrial heat source production area identification device based on multi-band observation data in an embodiment of the present application;

[0044] Figure 3 This is another structural framework diagram of an industrial heat source production area identification device based on multi-band observation data in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The embodiments in the present application and all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0046] The SDGSAT-1 satellite is designed for a sun-synchronous orbit and is equipped with three payloads: high-resolution wide-band thermal infrared, low-light and multi-spectral imagers. The orbit altitude is 505km, the inclination is 97.5°, the spatial resolution is 30m and 10m respectively, the width is 300km, and global coverage can be achieved in 11 days. It is designed with survey observation modes such as "thermal infrared + multi-spectral", "thermal infrared + low-light" and single-payload observation, which can realize all-day, multi-payload collaborative observation. At the same time, it has on-board and field calibration modes such as lunar calibration, blackbody variable temperature calibration, LED light calibration, and one-line flight calibration to ensure the needs of accurate quantitative detection.

[0047] This application takes the SDGSAT-1 satellite as an example to further describe the technical solution in detail.

[0048] Figure 1 A flow chart of the method for identifying industrial heat source production areas based on multi-band observation data in an embodiment of the present application, the steps are as follows:

[0049] Step S110: Acquire multi-band observation data of the target area; the multi-band observation data includes multi-band remote sensing satellite data obtained based on high-resolution thermal infrared sensors, multi-spectral sensors and low-light sensors.

[0050] The four-level product data, including the data of multi-spectral (MII), thermal infrared (TIS) and GIU sensors, were downloaded from the SDGSAT-1 data open system of CBAS (http: / / 124.16.184.48:6008 / home). After the data acquisition was completed, the different data were cut and analyzed according to the administrative boundaries of Shanghai. The original data only included geometric correction and simple radiation correction, and did not include the processing of the data in this paper.

[0051] In one embodiment, according to the characteristics of different sensor data and the requirements for industrial heat source production area identification, different analysis processes need to be performed on the data of different sensors:

[0052] For the SDGSAT-1 thermal infrared sensor data, after radiation, the 3-band temperature of the SDGSAT-1 thermal infrared sensor is inverted;

[0053] Specifically, according to the radiation calibration formula of L = DN*GAIN+BIAS, the latest version of the radiation calibration parameters is used to calibrate the data. Among them, DN is the effective recorded value of the image, GAIN is the calibration gain, BIAS is the calibration offset, and L is the calibrated emissivity. Since the SDGSAT-1 thermal infrared sensor is composed of 4 CCDs, the calibration parameters of each band (T1-3) and each piece are different. For details, you can check the corresponding "*.calib.xml" file.

[0054] After the radiation calibration is completed, the radiation brightness temperature of the three thermal infrared bands is calculated respectively. The conversion formula of the radiation brightness temperature is shown below:

[0055]

[0056] Where T is the brightness temperature (K), h is Planck's constant (h = 6.62606896 × 10 -34 J·S), k is the Boltzmann constant (k = 1.38 × 10 -23 J / S), c is the speed of light (c = 2.99792458 × 10 8 m / s), μ is the central wavelength (m), and the unit of L is W / (m 2 ·sr·um). The central wavelengths of the three thermal infrared bands of SDGSAT-1 are 11.81μm, 9.32μm and 10.82μm respectively.

[0057] For the SDGSAT-1 multi-spectral sensor data, after radiometry, the reflectivity of the SDGSAT-1 multi-spectral sensor bands B1-7 was inverted;

[0058] Specifically, for the SDGSAT-1 multi-spectral sensor data, the data is calibrated according to the radiation calibration formula of L = DN * GAIN + BIAS, using the latest version of the radiation calibration parameters. The calibration parameters for each band of B1-7 are different, and you can view the corresponding "*.calib.xml" file for details. And according to the radiation transfer model, calculate the reflectivity information of each band. And use the thermal infrared data as a reference and align it with 30m resolution data.

[0059] For the SDGSAT-1 low-light sensor data, after radiation, the light radiation values ​​of the R, G, B, PL, and PH bands of the SDGSAT-1 low-light sensor are calculated.

[0060] Specifically, for the SDGSAT-1 low-light sensor data, according to the radiation calibration formula of L = DN*GAIN+BIAS, the latest version of the radiation calibration parameters is used to calibrate the data. The calibration parameters for each band of R, G, B, PL, and PH are different. For details, you can view the corresponding "*.calib.xml" file. And with the thermal infrared data as a reference, it is aligned to 30m resolution data.

[0061] This application applies SDGSAT-1 multi-sensor data to industrial heat source production identification for the first time. At the same time, SDGSAT-1 high-resolution thermal infrared sensor data better meets the requirements of industrial heat source production area identification accuracy.

[0062] Step S120: Based on the spectral distribution of the industrial heat source production area, extract multivariate indicator data reflecting the characteristics of the industrial heat source production area from the multi-band observation data.

[0063] After comprehensively analyzing the spectral distribution characteristics of industrial heat source production areas, the features extracted in this application mainly include: reflectivity index, night light radiation value index, radiation brightness temperature index, thermal radiation index, NDWI (Normalized Difference Water Index, NDWI) and NDVI (Normalized Difference Vegetation Index, NDVI) and other indicator information.

[0064] In one embodiment, the industrial heat source production area index information includes:

[0065] The reflectivity index, night light radiation value index, and radiation brightness temperature index refer to the reflectivity of the B1-7 band of the multi-spectral sensor, the night light radiation value of the R, G, B, PL, and PH5 bands in the low light sensor data, and the radiation brightness temperature of the T1-3 band of the thermal infrared sensor.

[0066] The thermal radiation index refers to the characteristics of the industrial heat source production area. On the basis of the normalized combustion index, in order to increase the difference between the industrial heat source production area and the background area, and in view of the lack of short-wave infrared data in SDGSAT-1, this application proposes 6 new thermal infrared radiation index (RI) characteristics on the premise of fully investigating the distribution characteristics of the industrial heat source production area in the thermal infrared sensor spectrum. The thermal infrared index proposed in this application includes:

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073] Among them, T3, T2 and T1 are the radiation brightness temperatures of the thermal infrared sensor with central wavelengths of 11.81μm, 9.32μm and 10.82μm, respectively, and ρ7 and ρ6 are the reflectances of the multi-spectral sensor with central wavelengths of 850nm and 785nm.

[0074] Normalized difference water index NDWI and normalized difference vegetation index NDVI indicator information.

[0075] Specifically, the calculation method is as follows:

[0076]

[0077]

[0078] Among them, ρ4 and ρ5 are the reflectances of the multi-spectral sensor with central wavelengths of 560nm and 660nm.

[0079] Step S130: Input the multivariate indicator data reflecting the characteristics of the industrial heat source production area into the industrial heat source production area identification model to realize the identification of the industrial heat source production area in the target area; wherein, the industrial heat source production area identification model is trained according to the sample set using the random forest machine learning method; the sample set is constructed using high-resolution remote sensing satellite position information and prior industrial heat source production area characteristic indicator information.

[0080] In one embodiment, the steps include:

[0081] Construct sample sets of industrial heat source production areas and background areas, and comprehensively utilize high-resolution remote sensing satellite data such as Google Earth and multivariate index feature information to build as many sample sets of industrial heat source production areas and background areas with uniform spatial distribution as possible;

[0082] An industrial heat source production area identification model with comprehensive multivariate index characteristics is constructed. The random forest machine learning method is adopted. The prior multivariate index characteristics are used as input features. The random forest model for industrial heat source production area identification is trained on the training sample set to form an industrial heat source production area identification model with comprehensive multivariate index characteristics.

[0083] In one embodiment, the industrial heat source production area identification model of the comprehensive multivariate index feature constructed by the industrial heat source production area identification model training layer based on the comprehensive multivariate index feature is used to identify the industrial heat source production area in the study area based on the multivariate index feature of the industrial heat source index feature extraction layer of the comprehensive SDGSAT-1 satellite multi-band data, so as to realize the identification of the industrial heat source production area in the study area;

[0084] Use cloud mask, NDVI, NDWI and other indicators to eliminate false industrial heat source production areas.

[0085] In one embodiment, the accuracy of the industrial heat source production area identification is qualitatively and quantitatively analyzed and verified.

[0086] The industrial heat source production areas identified in the study area are assisted by high-resolution remote sensing images, Google Earth data, POI datasets, local statistical datasets, and manual field survey data. The identification accuracy of the industrial heat source production areas in the study area is qualitatively and quantitatively analyzed and verified by using test sample verification, cross-validation, and manual verification in typical areas.

[0087] Figure 2 The structural framework diagram of the industrial heat source production area identification device based on multi-band observation data in the embodiment of the present application includes:

[0088] The multi-band data preprocessing module acquires the thermal infrared, low-light and multi-spectral sensor data of the SDGSAT-1 satellite and performs analytical processing;

[0089] The industrial heat source index feature extraction module integrates multi-band data to extract the operating status index information of the industrial heat source production area from the data obtained through analytical processing;

[0090] The module for building an industrial heat source production area identification model with comprehensive multivariate index characteristics comprehensively utilizes high-resolution remote sensing satellite data such as Google Earth and prior multivariate index characteristic information to build as many sample sets of industrial heat source production areas and background areas with uniform spatial distribution as possible, trains the random forest model for industrial heat source production area identification, and builds an industrial heat source production area identification model with comprehensive multivariate index characteristics;

[0091] An industrial heat source production area identification module with comprehensive multivariate index characteristics combines the industrial heat source production area index information and the industrial heat source production area identification model to realize the industrial heat source production area identification in the studied area;

[0092] The industrial heat source production area identification result verification module performs qualitative and quantitative analysis and verification on the accuracy of the industrial heat source production area identification.

[0093] This application proposes a new method for discovering industrial heat source production areas, namely, a method for discovering industrial heat source production areas based on comprehensive SDGSAT-1 multi-band data. Its purpose is to make full use of the advantages of SDGSAT-1 high-resolution thermal infrared sensor data, superimpose multi-spectral and low-light band data, and combine multi-sensor data preprocessing, multi-industrial heat source production and operation status characterization index extraction, target recognition and other algorithms to identify industrial heat source production areas. As a new method for discovering industrial heat source production areas based on comprehensive SDGSAT-1 multi-band data, this method better meets the problem of high-precision industrial heat source production area identification, and provides favorable auxiliary data for environmental monitoring, heavy industry economic benefit evaluation, and related auxiliary industry configuration. It provides an independent scientific basis for my country to actively respond to the sustainable development process (SDG indicators 7.3 / 9.2 / 12.2 / 12.a) of "adjusting structure and de-capacity" industrial model upgrading, carbon tax domestic and foreign transactions, and atmospheric environment improvement.

[0094] Figure 3 Another structural framework diagram 300 of the industrial heat source production area identification device based on multi-band observation data in an embodiment of the present application includes:

[0095] The device 300 may be a chip system. In the embodiment of the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices. The device 300 includes at least one processor 310 for implementing the method provided in the embodiment of the present application. The device 300 may also include a communication interface 320. In the embodiment of the present application, the communication interface 320 may be a transceiver, a circuit, a bus, a module, or other types of communication interfaces for communicating with other devices through a transmission medium.

[0096] The processor 310 may execute Figure 2 The functions performed by the SDGSAT-1 satellite multi-band data preprocessing module, the industrial heat source index feature extraction module based on the SDGSAT-1 satellite multi-band data, the industrial heat source production area identification model construction module based on the multi-index feature, the industrial heat source production area identification model construction module based on the multi-index feature, and the industrial heat source production area identification result verification module in the device shown; the communication interface 320 can be used to perform Figure 2 The device shown performs functions of acquiring thermal infrared, low light and multi-spectral sensor data from the SDGSAT-1 satellite, as well as obtaining information such as high-resolution remote sensing images from Google Earth.

[0097] When the device 300 is used to execute the above method, the communication interface 320 is used to obtain the thermal infrared, low light and multi-spectral sensor data of the SDGSAT-1 satellite, and obtain information such as Google Earth high-resolution remote sensing images; the processor 310 is used to parse and process the thermal infrared, low light and multi-spectral sensor data of the SDGSAT-1 satellite; extract the operating status indicator information of the industrial heat source production area from the obtained data; from the information such as Google Earth high-resolution remote sensing images, build a sample set of industrial heat source production areas and background areas with uniform spatial distribution, train a random forest model for industrial heat source production area identification, and construct an industrial heat source production area identification model with comprehensive multivariate index characteristics; combine the industrial heat source production area indicator information and the industrial heat source production area identification model to realize the identification of the industrial heat source production area in the studied area.

[0098] The communication interface 320 is also used to perform other steps or operations in the above method embodiment except obtaining the thermal infrared, low light and multi-spectral sensor data of the SDGSAT-1 satellite and obtaining information such as Google Earth high-resolution remote sensing images. The processor 310 can also be used to perform other steps or operations in addition to the processing process in the above method embodiment, which will not be described one by one here.

[0099] The device 300 may also include at least one memory 330 for storing program instructions and / or data. The memory 330 is coupled to the processor 310. The coupling in the embodiment of the present application is an indirect coupling or communication connection between devices, units or modules, which may be electrical, mechanical or other forms, and is used for information exchange between devices, units or modules. The processor 310 may operate in conjunction with the memory 330. The processor 310 may execute program instructions stored in the memory 330. In one possible implementation, at least one of the at least one memory may be integrated with the processor. In another possible implementation, the memory 330 is located outside the device 300.

[0100] The specific connection medium between the communication interface 320, the processor 310 and the memory 330 is not limited in the embodiment of the present application. Figure 3 In the embodiment, the memory 330, the processor 310 and the communication interface 320 are connected via a bus 340. Figure 3 The connections between the other components are shown in bold lines, which are only for illustration and are not intended to be limiting. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0101] Exemplarily, the processor 310 may be one or more central processing units (CPUs). When the processor 310 is a CPU, the CPU may be a single-core CPU or a multi-core CPU. The processor 410 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0102] Exemplarily, the memory 330 may include, but is not limited to, non-volatile memories such as a hard disk drive (HDD) or a solid-state drive (SSD), random access memory (RAM), erasable programmable read-only memory (EPROM), read-only memory (ROM) or portable read-only memory (CD-ROM), etc. The memory is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present application can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0103] The present application provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0104] Acquire thermal infrared, low light and multi-spectral sensor data from SDGSAT-1 satellite and perform analysis and processing;

[0105] Extracting the information of the operating status indicators of the industrial heat source production area from the data obtained by analytical processing;

[0106] Build a sample set of industrial heat source production areas and background areas with uniform spatial distribution, train a random forest model for industrial heat source production area identification, and construct an industrial heat source production area identification model with comprehensive multivariate index characteristics;

[0107] Combining the industrial heat source production area index information and the industrial heat source production area identification model, the industrial heat source production area in the studied area can be identified;

[0108] The accuracy of industrial heat source production area identification was qualitatively and quantitatively analyzed and verified.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0112] This application proposes an ultrasonic CT imaging based on wave equations, specifically, an ultrasonic CT multi-parameter image reconstruction system and method based on FWI. This method can fully utilize the information of all waveforms (reflected waves, transmitted waves, diffracted waves, refracted waves, etc.) to reconstruct high-resolution images of multiple acoustic parameters (sound velocity, acoustic impedance, attenuation coefficient, etc.) by utilizing the numerical solution of the multi-parameter viscoacoustic wave equation. The numerical solution of the multi-parameter wave equation can be developed on the CPU and deployed on the GPU for actual detection tasks. An innovative multi-scale deconvolution objective function is constructed and a specific use strategy of the corresponding objective function in actual reconstruction is proposed. The robustness of the model reconstruction is improved by focusing on comparing the dynamic characteristics of the simulation data and the observation data.

[0113] It should be noted that the method provided herein is not inherently related to any specific computer, virtual device or other equipment. Various general devices can also be used together with the teaching based on this. According to the above description, it is obvious to construct the structure required for this type of device. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​can be utilized to implement the present invention described herein, and the description of the call of specific languages ​​and system function modules above is only for the purpose of disclosing the best mode of the invention.

[0114] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be implemented without these specific details. In some examples, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0115] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. A method for identifying industrial heat source production areas based on multi-band observation data, characterized in that: The method comprises: Acquire multi-band observation data of the target area; the multi-band observation data includes multi-band remote sensing satellite data obtained based on high-resolution thermal infrared sensors, multi-spectral sensors and low-light sensors; the obtained remote sensing satellite data needs to be analyzed and processed before being used as multi-band observation data, including: for thermal infrared sensor data, after radiation, invert the on-board brightness temperature of the thermal infrared sensor 3 bands; for multi-spectral sensor data, after radiation, invert the reflectivity of multi-spectral sensor bands B1-7; for low-light sensor data, after radiation, calculate the light radiation values ​​of the low-light sensor R, G, B, PL, PH bands; Based on the spectral distribution of the industrial heat source production area, multivariate index data reflecting the characteristics of the industrial heat source production area are extracted from the multi-band observation data; the multivariate index data reflecting the characteristics of the industrial heat source production area include: reflectivity index, night light radiation value index, and radiation brightness temperature index, which respectively refer to the reflectivity of the B1-7 band of the multi-spectral sensor, the night light radiation value of the R, G, B, PL, PH5 bands in the low light sensor data, and the radiation on-board brightness temperature of the T1-3 band of the thermal infrared sensor; the thermal radiation index refers to the 6 short-wave thermal radiation index RI characteristics extracted according to the characteristics of the industrial heat source production area: Among them, T3, T2 and T1 are the radiation brightness temperatures of the thermal infrared sensor with central wavelengths of 11.81μm, 9.32μm and 10.82μm, respectively, and ρ7 and ρ6 are the reflectances of the multi-spectral sensor with central wavelengths of 850nm and 785nm; Normalized Water Index NDWI and Normalized Vegetation Index NDVI indicator information; The multivariate index data reflecting the characteristics of the industrial heat source production area is input into the industrial heat source production area identification model to realize the identification of the industrial heat source production area in the target area; wherein, the industrial heat source production area identification model is obtained by training according to the sample set using the random forest machine learning method; the sample set is constructed using high-resolution remote sensing satellite position information and prior industrial heat source production area characteristic index information; the training steps of the industrial heat source production area identification model include: constructing a sample set of industrial heat source production areas and background areas, comprehensively utilizing Google Earth high-resolution remote sensing satellite data and multivariate index feature information, and constructing as many sample sets of industrial heat source production areas and background areas with uniform spatial distribution as possible; constructing an industrial heat source production area identification model with comprehensive multivariate index features, adopting the random forest machine learning method, taking the multivariate index features as input features, and training the random forest model for industrial heat source production area identification on the training sample set to form an industrial heat source production area identification model with comprehensive multivariate index features.

2. The industrial heat source production area identification method according to claim 1, characterized in that: The process of industrial heat source production area identification includes: The multivariate index features of the research area extracted by the industrial heat source index feature extraction layer of the comprehensive multi-band data are used to realize the identification of the industrial heat source production area in the target area by using the industrial heat source production area identification model with comprehensive multivariate index features constructed by the industrial heat source production area identification model training layer; Cloud mask, NDVI and NDWI indicators are used to eliminate false industrial heat source production areas.

3. The industrial heat source production area identification method according to claim 1, characterized in that: After the industrial heat source production area of ​​the target area is identified, it is also necessary to use high-resolution remote sensing images, Google Earth data, POI data sets, local statistical data sets, and manual field survey data to conduct qualitative and quantitative analysis and verification of the identification accuracy of the industrial heat source production area in the study area by means of test sample verification, cross-validation, and manual verification of typical areas.

4. An industrial heat source production area identification device based on multi-band observation data, used to execute the method according to claim 1, characterized in that: The device comprises: Multi-band data preprocessing module, which acquires thermal infrared, low light and multi-spectral sensor data from remote sensing satellites and performs analytical processing; The industrial heat source index feature extraction module integrates multi-band data to extract the operating status index information of the industrial heat source production area from the data obtained through analytical processing; A module for building a model for identifying industrial heat source production areas based on comprehensive multivariate index characteristics, building a sample set of industrial heat source production areas and background areas with uniform spatial distribution, training a random forest model for industrial heat source production area identification, and building a model for identifying industrial heat source production areas based on comprehensive multivariate index characteristics; An industrial heat source production area identification module with comprehensive multivariate index characteristics combines the industrial heat source production area index information and the industrial heat source production area identification model to realize the industrial heat source production area identification in the studied area; The industrial heat source production area identification result verification module performs qualitative and quantitative analysis and verification on the accuracy of the industrial heat source production area identification.

5. An industrial heat source production area identification device based on multi-band observation data, characterized in that: The device comprises at least one processor, wherein the processor is used to execute a program stored in a memory, and when the program is executed, the device executes the method according to any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images