Aviation surface reflectance image production method and related product

Abstract

The invention discloses an aviation surface reflectance image production method and a related product, and relates to the technical field of optical remote sensing. Comprising the following steps: acquiring an aviation optical image and a ground feature spectrum corresponding to the aviation optical image; and inputting the aviation optical image and the ground feature spectrum into a reflectivity image generation model to obtain an aviation surface reflectivity image. According to the method, after the aviation optical image and the corresponding ground feature spectrum are obtained, complex operation is not needed, the data are directly input into the reflectivity image generation model, and the aviation surface reflectivity image can be obtained.

Description

Technical Field

[0001] The present application relates to the field of optical remote sensing technology, and in particular to a method for producing an aerial surface reflectivity image and related products. Background Art

[0002] Surface reflectivity refers to the ratio of the total solar radiation flux reflected by surface objects in all directions to the total radiation flux reaching the surface of the object; it is an important parameter that characterizes the ground's ability to absorb and reflect solar radiation. Surface reflectivity products are generated based on surface reflectivity images and can provide basic data support for land use classification, long-term land cover change detection, climate change research, environmental monitoring and other fields.

[0003] The surface reflectance images currently widely used are satellite surface reflectance images, including but not limited to: MODIS satellite surface reflectance images with a resolution of 250m, 500m, and 1km, Landsat series satellite surface reflectance images with a resolution of 30m, and Sentinel-2 satellite surface reflectance images with a resolution of 10m, 20m, and 60m.

[0004] The spatial resolution of aerial images is higher than that of satellite images, and they are more advantageous for fine-scale applications. How to make good use of the existing massive aerial optical images to obtain aerial surface reflectance images has become one of the technical problems that need to be solved in the field of optical remote sensing technology. Summary of the invention

[0005] Based on the above problems, the present application provides a method for producing aerial surface reflectance images, which is used to automatically batch produce aerial surface reflectance images based on existing massive aerial optical images.

[0006] The embodiments of the present application disclose the following technical solutions:

[0007] The first aspect of the present application discloses a method for producing an aerial surface reflectance image, comprising:

[0008] Acquire an aerial optical image and a ground object spectrum corresponding to the aerial optical image;

[0009] An aerial surface reflectance image is obtained through a reflectance image generation model according to the aerial optical image and the ground object spectrum.

[0010] In an optional implementation, the reflectance image generation model includes: a data preprocessing module, a hyperspectral data judgment module, a spectral correction module, a fast atmospheric correction module and a secondary correction module;

[0011] The method of generating a reflectivity image model to obtain an aerial surface reflectivity image according to the aerial optical image and the ground object spectrum includes:

[0012] The data preprocessing module performs preprocessing operations on the aerial optical image and the ground object spectrum to obtain preprocessed aerial optical images and preprocessed ground object spectrum; the preprocessing operations include: data screening, format conversion, field of view stitching and image non-uniformity correction;

[0013] By means of the hyperspectral judgment module, it is judged whether the preprocessed aero-optical image is hyperspectral data; if it is determined that the preprocessed aero-optical image is hyperspectral data, the preprocessed aero-optical image is input into the spectral correction module; otherwise, the preprocessed aero-optical image is input into the rapid atmospheric correction module;

[0014] Obtaining a spectrally corrected aero-optical image according to the pre-processed aero-optical image by means of the spectral correction module;

[0015] Obtaining the initial aerial surface reflectance image according to the preprocessed aerial optical image or the spectrally corrected aerial optical image through the rapid atmospheric correction module;

[0016] The secondary correction module is used to obtain a secondary corrected aerial surface reflectivity image based on the initial aerial surface reflectivity image and the measured reflectivity of the ground target.

[0017] In an optional implementation, after the secondary correction module, the reflectivity image generation model further includes an accuracy verification module, and the method further includes:

[0018] The accuracy verification module is used to obtain the correction accuracy based on the secondary corrected aerial surface reflectivity image and the measured reflectivity of the target object.

[0019] In an optional implementation, after the secondary correction model and before the accuracy verification module, the reflectivity image generation model further includes a terrain judgment module and a terrain correction module; the method further includes:

[0020] By means of the terrain judgment module, it is judged whether the terrain in the aerial optical image is a mountainous terrain; if the terrain in the aerial optical image is a mountainous terrain, the secondary corrected aerial surface reflectivity image is input into the terrain correction module; otherwise, the secondary corrected aerial surface reflectivity image is input into the accuracy verification module;

[0021] Obtaining a terrain-corrected aerial surface reflectivity image according to the secondary-corrected aerial surface reflectivity image and the terrain correction factor through the terrain correction module;

[0022] The method of obtaining the correction accuracy through the accuracy verification module according to the secondary corrected aerial surface reflectivity image and the measured reflectivity of the target object includes:

[0023] The correction accuracy is obtained through the accuracy verification module according to the aerial surface reflectivity image after the terrain correction and the measured reflectivity of the target object.

[0024] In an optional implementation, after the secondary correction model, the reflectivity image generation model further includes a terrain judgment module and a terrain correction module; the method further includes:

[0025] By means of the terrain judgment module, it is judged whether the terrain in the aero-optical image is a mountainous terrain; if the terrain in the aero-optical image is a mountainous terrain, the aero-surface reflectivity image after the secondary correction is input into the terrain correction module; otherwise, the aero-surface reflectivity image after the secondary correction is output;

[0026] The terrain correction module obtains the terrain-corrected aerial surface reflectivity image based on the secondary-corrected aerial surface reflectivity image and the terrain correction factor.

[0027] In an optional implementation, obtaining a secondary-corrected aerial surface reflectivity image according to the initial aerial surface reflectivity image and the measured reflectivity of the ground target by the secondary correction module includes:

[0028] Fitting the initial aerial surface reflectivity image and the measured reflectivity of the target to obtain a correction coefficient;

[0029] Based on the correction coefficient and the initial aerial surface reflectivity image, the secondary corrected aerial surface reflectivity image is obtained.

[0030] The second aspect of the present application discloses an aerial surface reflectivity image production device, the device comprising:

[0031] A parameter acquisition module, used to acquire an aerial optical image and a spectrum of a ground object corresponding to the aerial optical image;

[0032] The surface reflectance image production module is used to generate a model through a reflectance image, and obtain an aerial surface reflectance image according to the aerial optical image and the ground object spectrum.

[0033] A third aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method introduced in any implementation manner of the first aspect.

[0034] A fourth aspect of the present application provides an electronic device, including:

[0035] a memory having a computer program stored thereon;

[0036] A processor is used to execute the computer program in the memory to implement the steps of the method introduced in any implementation manner of the first aspect.

[0037] Compared with the prior art, this application has the following beneficial effects:

[0038] The present application discloses a method for producing an aerial surface reflectance image, comprising: obtaining an aerial optical image and a surface spectrum corresponding to the aerial optical image; inputting the aerial optical image and the surface spectrum into a reflectance image generation model to obtain an aerial surface reflectance image. Through the method of the present application, after obtaining the aerial optical image and the corresponding surface spectrum, the above data is directly input into the reflectance image generation model without complicated operations to directly obtain the aerial surface reflectance image. That is, the solution in the present application can realize automatic batch production of surface reflectance images of aerial optical cameras. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 A flowchart of a method for producing an aerial surface reflectance image provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram of the structure of a reflectivity image generation model provided in an embodiment of the present application;

[0042] Figure 3 A flowchart of a rapid atmospheric correction provided for this application;

[0043] Figure 4 A schematic diagram of the structure of an aerial surface reflectivity image production device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] Surface reflectivity refers to the ratio of the total solar radiation flux reflected by the surface objects in all directions to the total radiation flux reaching the surface of the object; it is an important parameter that characterizes the ground's ability to absorb and reflect solar radiation. Surface reflectivity images can provide basic data support for land use classification, long-term land cover change detection, climate change research, environmental monitoring and other fields.

[0045] The surface reflectance images currently widely used are satellite surface reflectance images, including but not limited to: MODIS satellite surface reflectance images with a resolution of 250m, 500m, and 1km, Landsat series satellite surface reflectance images with a resolution of 30m, and Sentinel-2 satellite surface reflectance images with a resolution of 10m, 20m, and 60m.

[0046] As the earliest and most widely used aerial remote sensing camera, aerial optical cameras are usually used for surveying and mapping and building three-dimensional models. In the early days, only their geometric characteristics were focused on. In recent years, with the rapid development of quantitative remote sensing, people have gradually paid attention to the radiation performance of such cameras, hoping to tap their potential for quantitative applications.

[0047] The number of optical images taken by aerial optical cameras is increasing, and related applications are becoming more and more extensive. As the basic data source for quantitative remote sensing applications, users have an increasingly urgent demand for their reflectance images. How to make good use of the existing massive aerial optical images to obtain aerial surface reflectance images has become one of the technical problems that need to be solved in the field of optical remote sensing technology.

[0048] Based on this, the present application discloses a method for batch production of commercial aerial surface reflectance images, including: obtaining aerial optical images and ground object spectra corresponding to the aerial optical images; obtaining aerial surface reflectance images according to the aerial optical images and the ground object spectra through a reflectance image generation model; the aerial surface reflectance images include the reflectance of the ground area corresponding to each pixel point in the aerial optical image. The method disclosed in the present application can realize automatic batch production of aerial optical camera ground reflectance images.

[0049] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution 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 only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] Figure 1 A flowchart of a method for producing an aerial surface reflectivity image provided in an embodiment of the present application. Figure 1 As shown, the method for producing an aerial surface reflectance image disclosed in the present application comprises:

[0051] S101, acquiring an aero-optical image and a ground object spectrum corresponding to the aero-optical image.

[0052] Aerial optical images refer to ground images taken by optical cameras mounted on aircraft, helicopters or drones. These images mainly record visible light and can capture detailed information such as the natural color, texture, and shape of the ground. Aerial optical images have the characteristics of high spatial resolution, large coverage and high precision, and are widely used in many fields.

[0053] The ground object spectrum corresponding to the aerial optical image refers to the ground object spectrum obtained by or approximately synchronously. The ground object spectrum refers to the intensity distribution of the reflected, transmitted or emitted light of the ground object at different wavelengths. In the acquisition of aerial optical images, the ground object spectrum data can be used to analyze the composition, properties and interactions of the ground objects. In this application, the internationally used FieldSpec-4ASD spectrometer and the matching whiteboard are used to measure the ground object spectrum.

[0054] It should be noted that obtaining the spectrum of ground objects corresponding to the aerial optical image is well known to those skilled in the art, and the specific process of obtaining the spectrum of ground objects corresponding to the aerial optical image is not limited in the present application.

[0055] S102, obtaining an aerial surface reflectance image according to the aerial optical image and the ground feature spectrum by using a reflectance image generation model.

[0056] The reflectivity image generation model in the present application is a model that is constructed in advance and can generate an aerial surface reflectivity image. The aerial surface reflectivity image in the present application is a reflectivity image corresponding to the aerial surface image.

[0057] Figure 2 A schematic diagram of the structure of a reflectivity image generation model provided in an embodiment of the present application. Figure 2 The reflectance image generation model 200 includes: a data preprocessing module 201, a hyperspectral data judgment module 202, a spectral correction module 203, a fast atmospheric correction module 204, a secondary correction module 205, a terrain judgment module 206, a terrain correction module 207 and an accuracy verification module 208.

[0058] It should be emphasized that the data preprocessing module 201, the hyperspectral data judgment module 202, the spectral correction module 203, the rapid atmospheric correction module 204, the secondary correction module 205 and the accuracy verification module 208 in the reflectance image generation model 200 in the present application are mandatory modules, while the terrain judgment module 206 and the terrain correction module 207 are optional modules.

[0059] In an optional implementation, the reflectance image generation model 200 includes: a data preprocessing module 201, a hyperspectral data judgment module 202, a spectral correction module 203, a fast atmospheric correction module 204, a secondary correction module 205, a terrain judgment module 206, a terrain correction module 207 and an accuracy verification module 208.

[0060] In another optional implementation, the reflectance image generation model includes: data preprocessing module 201, hyperspectral data judgment module 202, spectral correction module 203, fast atmospheric correction module 204, secondary correction module 205 and accuracy verification module 208 all required modules.

[0061] It is understandable that those skilled in the art can decide whether to enable the terrain determination module 206 and the terrain correction module 207 in the reflectivity image generation model 200 according to business needs.

[0062] For ease of understanding, in this application Figure 2 As an example, the reflectivity image generation model in Figure 2 After the reflectivity image generation model is constructed, the processing of input data by each module in the model is as follows:

[0063] In the first step, a preprocessing operation is performed on the aero-optical image and the ground object spectrum through the data preprocessing module 201 to obtain a preprocessed aero-optical image and a preprocessed ground object spectrum.

[0064] The data preprocessing module 201 can be used to perform preprocessing operations such as data checking operation, auxiliary data separation operation, format conversion operation and field of view stitching operation on the aero-optical image to obtain the preprocessed aero-optical image.

[0065] It can be understood that the data check operation ensures that the received image data is complete and undamaged, and preliminarily evaluates the image quality; the auxiliary data separation operation separates the auxiliary information (such as GPS trajectory, sensor calibration parameters, etc.) that may be contained in the aerial image data from the aerial image data for separate processing or storage. The format conversion operation converts the aerial image data into a standard format suitable for subsequent processing or analysis to ensure compatibility and efficiency. The field of view stitching operation stitches multiple aerial images with partially overlapping areas into a complete, larger-scale image.

[0066] Through the above preprocessing steps, aerial optical images can be optimized into a form that is more suitable for subsequent processing and analysis, thereby improving data utilization efficiency and accuracy.

[0067] This application uses the FieldSpec-4 ASD spectrometer and a matching whiteboard to measure the spectrum of ground objects. The file type of the raw data of the ground object spectrum measured by the spectrometer is .asd; the spectrometer directly measures the DN (Digital Number) value of the ground object and the whiteboard. Ground object spectrum processing means that the data preprocessing module converts the .asd type file into a text file, and obtains the measured reflectivity of the target ground object through "ground object reflectivity = (ground object average DN value / whiteboard average DN value) × whiteboard reflectivity". Among them, the DN value is usually used to describe the pixel value that has not been calibrated to a unit with actual physical meaning.

[0068] In the second step, the hyperspectral judgment module 202 is used to determine whether the preprocessed aero-optical image is hyperspectral data; if it is determined that the preprocessed aero-optical image is hyperspectral data, the preprocessed aero-optical image is input into the spectral correction module; otherwise, the preprocessed aero-optical image is input into the rapid atmospheric correction module.

[0069] Hyperspectral data in this application refers to image data or image data taken by a hyperspectral camera. Generally speaking, the spectral resolution of hyperspectral data is very high, usually reaching the nanometer level (such as within 10nm), which means that it can capture more subtle spectral differences.

[0070] In the third step, the spectrum correction module 203 is used to obtain a spectrum-corrected aero-optical image based on the pre-processed aero-optical image.

[0071] In an optional implementation, the specific process of obtaining the spectrally corrected aero-optical image according to the pre-processed aero-optical image by the spectral correction module 203 is as follows:

[0072] The 760nm oxygen channel is used to detect spectral drift. Assuming that the selected spectral range is 740nm to 780nm, the preset range of the drift Δλ is -5 to +5nm, and the drift step is 0.1nm, the radiance spectrum with different spectral drift is used as the test spectrum, and the radiance spectrum calculated based on the original spectral parameters is used as the reference spectrum. The spectral angle (SA) between the test spectrum and the reference spectrum is calculated. The Δλ corresponding to the minimum spectral angle is the spectral drift of the load. The formula for the spectral angle is shown in formula (1):

[0073]

[0074] In formula (1), x i represents the central wavelength of the i-th channel in the reference spectrum, y i Represents the central wavelength of the i-th channel in the test spectrum, where i is an integer greater than or equal to 1 and less than or equal to n.

[0075] Assuming the full width at half maximum corresponding to the i-th channel, that is, fwhm(i) remains unchanged, the spectral response function SRF(i,Δλ) corresponding to the channel with different drift amounts Δλ is obtained by Gaussian function simulation as shown in formula (2):

[0076]

[0077] Wherein, SRF(i,Δλ) in formula (2) represents the channel spectral response function corresponding to the i-th channel obtained after spectral drift; i is the number of channels; Δλ is the drift amount; λ is the central wavelength at the oxygen absorption channel; λc(i) is the original central wavelength of the i-th channel.

[0078] The drift Δλ of 760nm is taken as the drift of all channels, that is, it is assumed that each channel in all channels has drifted by Δλ; the drift Δλ is added to the original central wavelength to obtain a new central wavelength, and then a rapid atmospheric correction is performed on it.

[0079] In the fourth step, the rapid atmospheric correction module 204 is used to obtain an initial aerial surface reflectance image based on the preprocessed aerial optical image or the spectrally corrected aerial optical image.

[0080] It can be understood that if the aerial optical image input into the reflectance image generation model is hyperspectral data, the data is successively processed through the data preprocessing module 201, the hyperspectral data judgment module 202, and the spectral correction module 203 to obtain a spectrally corrected aerial optical image, which is then input into the rapid atmospheric correction module 204 to obtain an initial aerial surface reflectance image.

[0081] If the aerial optical image input into the reflectance image generation model is not hyperspectral data, the data is successively processed by the data preprocessing module 201 and the hyperspectral data judgment module 202, and is directly input into the fast atmospheric correction module 204 without being processed by the spectral correction module 203 to generate an initial aerial surface reflectance image.

[0082] Conventional atmospheric correction is a process of inverting the surface reflectivity based on the actual measured atmospheric parameters such as aerosol optical thickness, water vapor content, imaging angle, spectral response function, and flight altitude, according to radiation transfer models such as the 6S radiation transfer model or the MODTRAN radiation transfer model.

[0083] The purpose of atmospheric correction is to eliminate the interference of atmospheric scattering, absorption and reflection on the reflected signal of ground objects in order to obtain the true spectral reflectance or emissivity of the ground surface.

[0084] The process of atmospheric correction of radiance image based on radiation transfer model is shown in formula (3), formula (4) and formula (5). The expressions of formula (3), formula (4) and formula (5) are as follows:

[0085]

[0086] L λ =ρ * ×cos(θ s )×E0 / (d 2 ×π) (4)

[0087]

[0088] Among them, ρ in formula (3) * is the apparent reflectivity under optical load in complex environment, namely TOA (Top of Atmosphere) reflectivity, θ s and θ v are the solar zenith angle and the observation zenith angle, is the relative azimuth between the observation azimuth and the solar azimuth; ρ a is the path radiation; ρ t is the surface reflectivity; S is the hemispherical albedo; T(θ s ) and T(θ v ) are the total atmospheric transmittance in the sun direction (downward) and observation direction (upward), respectively.

[0089] Where, formula (4) represents ρ * and the apparent radiance L λ The linear relationship between them. In formula (4), ρ * is the apparent reflectivity under optical load in complex environment, E0 is the solar irradiance at the top of the atmosphere; d is the distance between the sun and the earth, in AU; L λ is the apparent radiance; θ s is the solar zenith angle.

[0090] Wherein, formula (5) represents the calculated surface reflectivity ρ t In formula (5), T = T(θ s )×T(θ v ), the meanings of the remaining letters refer to formula (3) and formula (4), which will not be repeated here.

[0091] It is understandable that in order to obtain high-quality aerial optical images, the camera parameters need to be adjusted frequently during flight due to light changes, differences in ground objects, etc. Each time the camera parameters are adjusted, the corresponding radiation performance of the camera will change, and the ground synchronization calibration operation needs to be re-performed. The ground synchronization calibration consumes manpower and material resources and is very costly. Due to the huge number of aerial optical images, the atmospheric correction method in the traditional scheme is used to correct the obtained radiance image, that is, the atmospheric correction method in formula (3), formula (4) and formula (5) is used. It takes a long time and has low work efficiency, and is not suitable for batch production of aerial surface reflectance images.

[0092] In order to improve work efficiency and reduce work time, the present application adopts a fast atmospheric correction model to perform atmospheric correction on the preprocessed aero-optical image or the spectrally corrected aero-optical image.

[0093] Figure 3 A flowchart of a fast atmospheric correction provided in this application. The full name of the fast atmospheric correction is: Quick Atmospheric Correction; the English abbreviation is: QUAC. Figure 3 As shown in FIG. 1 , the process of performing atmospheric correction on the preprocessed aero-optical image or the spectrally corrected aero-optical image includes:

[0094] First, the benchmark spectra of multispectral and hyperspectral data are collected, and then the collected benchmark spectra are normalized (i.e. Figure 3 The reference spectrum in is normalized); the normalized reference spectrum data is then subjected to vegetation filtering and endmember collection, and the collected data is then input into a bright line spectrum filter for endmember averaging, and the result of the endmember averaging and the value of the reference average endmember are synchronously input into the correction model for determination, and the reflectance data is calculated, that is, the initial aerial surface reflectance image is obtained.

[0095] The QUAC method usually does not require the user to input detailed atmospheric parameters (such as atmospheric composition, aerosol type, etc.). It directly uses the default parameters in the fast atmospheric correction module in the reflectance image generation model to process the preprocessed aero-optical image or the spectrally corrected aero-optical image, which makes the atmospheric correction process more efficient and simplified.

[0096] In the fifth step, a secondary correction module 205 is used to obtain a secondary corrected aerial surface reflectivity image based on the initial aerial surface reflectivity image and the measured reflectivity of the target.

[0097] The measured reflectivity of the target refers to the measured reflectivity of the target. The target usually refers to a standard object or area with known reflective properties, which is used to calibrate and verify the data obtained by the remote sensing sensor. The target can be artificially set or can be some natural objects with stable reflective properties. The type of target is not limited in this application, and those skilled in the art can select a suitable object as the target in this application according to actual needs.

[0098] After obtaining the measured reflectivity of the target, the initial aerial surface reflectivity image obtained in the fifth step and the measured reflectivity of the target are fitted to obtain the correction coefficient; the correction coefficient is applied to the initial aerial surface reflectivity image to obtain a secondary corrected (absolutely corrected) surface reflectivity image.

[0099] The sixth step is to use a terrain determination module 206 to determine whether the ground captured by the aerial optical image is a flat ground or a mountainous ground. If the ground captured by the aerial optical image is a flat ground, the data of the surface reflectivity image after absolute correction is input into the accuracy verification module 208; if the ground captured by the aerial optical image is a mountainous ground, the data of the surface reflectivity image after absolute correction is input into the terrain correction module 207.

[0100] The seventh step, the terrain correction module 207, is used to obtain a terrain-corrected aerial surface reflectivity image based on the absolutely corrected aerial surface reflectivity image (also called the secondary-corrected aerial surface reflectivity image) and the terrain correction factor.

[0101] The terrain correction module in this application uses the C terrain correction model to perform terrain correction on the secondary corrected aerial surface reflectivity image. Specifically, assuming that the solar zenith angle is S Z , the solar azimuth is S A , the slope and aspect of the inclined slope are θ and μ respectively, and the sun incidence angle on the slope is α, then the terrain correction factor H is:

[0102]

[0103] cosα=cos S Z cosθ+sinS Z sinθcos(S A -μ) (9)

[0104] Where t is the sky factor, t0 is the sky factor when cosα is equal to 0, and λ is the regression coefficient, which is obtained by regression analysis of the relationship between the surface reflectance before terrain correction and cosα.

[0105] The terrain correction factor H can be obtained by formula (6) to formula (9); the terrain correction factor H is then applied to each pixel of the secondary corrected surface reflectance image to obtain the terrain corrected aerial surface reflectance image.

[0106] In the eighth step, the accuracy verification module 208 is used to obtain the correction accuracy based on the aerial surface reflectivity image after terrain correction and the measured reflectivity of the target object; or the correction accuracy is obtained based on the surface reflectivity image after absolute correction and the measured reflectivity of the target object.

[0107] In the accuracy verification module 208, the reflectivity obtained by inversion and the measured reflectivity of the target object are substituted into formula (10) to obtain the correction accuracy.

[0108] The inverted reflectivity refers to the reflectivity of each pixel in the optical image obtained after being processed by the data preprocessing module 201, the hyperspectral data judgment module 202, the spectral correction module 203, the fast atmospheric correction module 204 and the secondary correction module 205, as well as the optional terrain judgment module 206 and the terrain correction module 207, which is the aerial surface reflectivity image after terrain correction.

[0109] The measured reflectivity refers to the actual reflectivity measured from typical ground objects, such as vegetation, cement or the ground.

[0110] Wherein, the expression of formula (10) is as follows:

[0111]

[0112] In formula (10), σ represents the value of the correction accuracy; ρ 反演 is the reflectivity obtained by inversion; ρ 实测 is the measured reflectivity of the target object.

[0113] It is understandable that in actual use, those skilled in the art may Figure 2 The reflectivity image generation model in the software image is developed to develop a corresponding software image, and the terrain judgment module 206 and the terrain correction module 207 in the software image are set as optional modules to obtain the Figure 2 The reflectivity image generation model.

[0114] If the terrain determination module 206 and the terrain correction module 207 are not enabled in the reflectivity image generation model 200 , the secondary-corrected aerial surface reflectivity image output by the secondary correction module 205 is directly input to the accuracy verification module 208 .

[0115] If the terrain judgment module 206 and the terrain correction module 207 are enabled in the reflectivity image production model 200, the secondary corrected aerial surface reflectivity image output by the secondary correction module 205 is input into the terrain judgment module 206; if it is determined that the terrain in the aerial optical image is mountainous terrain, the secondary corrected aerial surface reflectivity image is input into the terrain correction module 207 to obtain a terrain-corrected aerial surface reflectivity image, and then the terrain-corrected aerial surface reflectivity image is input into the accuracy verification module 208; if it is determined that the terrain in the aerial optical image is not mountainous terrain, the secondary corrected aerial surface reflectivity image is directly input into the accuracy verification module 208.

[0116] Through the introduction of the above embodiments, it can be seen that the technical solution in the present application can be summarized as follows: after obtaining the aerial image taken by the aerial camera and the ground parameters measured synchronously when taking the aerial image, the above two types of data can be input into Figure 2 The corresponding software image is input into the reflectivity image generation model, and the optional modules are selected through the interactive interface of the software image. After all the working modules in the reflectivity image generation model are determined, the input data is automatically processed to obtain the aero-optical reflectivity image.

[0117] The satellite surface reflectivity image generation system in the traditional solution is not universal for the image data collected by the aerial optical camera, and does not consider the impact of terrain undulation on the reflectivity results, and lacks the accuracy verification link. The method of this application is based on the existing principles and algorithms, combined with the characteristics of the aerial optical camera, and comprehensively considers the sensor radiation calibration, atmospheric scattering and absorption, terrain undulation and accuracy verification, so as to achieve the high-precision commercial batch generation of aerial optical camera reflectivity images.

[0118] In summary, the advantage of the present invention is that it solves the problem of automatic batch generation of surface reflectance images of aerial optical cameras.

[0119] Specifically, the advantages of this application include the following four aspects:

[0120] First, this application integrates the work of multiple stages such as data preprocessing, spectral correction, and atmospheric correction into a reflectance image generation model, realizing the construction of an automatic batch generation model for the commercialization of surface reflectance images of aerial optical cameras.

[0121] Second, this application takes into account the impact of terrain undulations on reflectivity calculation results (i.e., reflectivity inversion). Through terrain correction, the impact of terrain undulations on the sensor receiving signal is weakened or eliminated, the true reflectivity of the ground under horizontal surface conditions is restored, and the inversion accuracy is improved.

[0122] Third, in this application, the absolute radiation difference is eliminated through the fast atmospheric correction module, and the relative coefficient of the initial reflectivity image and the measured reflectivity of the target is calculated through the secondary correction module. The initial reflectivity image is corrected again using the relative coefficient to obtain a surface reflectivity image with higher accuracy.

[0123] Fourth, the accuracy verification of the reflectivity inversion results is considered in this application to ensure the authenticity of the reflectivity image.

[0124] Based on the aerial surface reflectivity image generation method disclosed in the aforementioned embodiment, the present application also discloses an aerial surface reflectivity image generation device. Figure 4 A schematic diagram of the structure of an aerial surface reflectivity image production device provided in an embodiment of the present application. Figure 4 As shown, the aerial surface reflectivity image generating device 400 disclosed in the present application includes:

[0125] The parameter acquisition module 401 is used to acquire the aerial optical image and the ground object spectrum corresponding to the aerial optical image;

[0126] The surface reflectance image generation module 402 is used to obtain an aerial surface reflectance image according to the aerial optical image and the surface feature spectrum through a reflectance image generation model.

[0127] In an optional implementation, the reflectance image generation model includes: a data preprocessing module, a hyperspectral data judgment module, a spectral correction module, a fast atmospheric correction module and a secondary correction module; the surface reflectance image production module 402 includes:

[0128] A preprocessing data acquisition unit, used to perform preprocessing operations on the aerial optical image and the ground object spectrum through the data preprocessing module to obtain the preprocessed aerial optical image and the preprocessed ground object spectrum; the preprocessing operations include: data screening, format conversion, field of view stitching and image non-uniformity correction;

[0129] A hyperspectral judgment unit, used to judge whether the preprocessed aero-optical image is hyperspectral data through the hyperspectral judgment module; if it is determined that the preprocessed aero-optical image is hyperspectral data, the preprocessed aero-optical image is input into the spectrum correction module; otherwise, the preprocessed aero-optical image is input into the rapid atmospheric correction module;

[0130] a spectrally corrected aero-optical image acquisition unit, configured to obtain a spectrally corrected aero-optical image according to the pre-processed aero-optical image through the spectral correction module;

[0131] An initial surface reflectance image generating unit, configured to obtain the initial aerial surface reflectance image according to the preprocessed aerial optical image or the spectrally corrected aerial optical image;

[0132] The secondary corrected aerial surface reflectivity image generating unit is used to obtain the secondary corrected aerial surface reflectivity image according to the initial aerial surface reflectivity image and the measured reflectivity of the ground target through the secondary correction module.

[0133] In an optional implementation, after the secondary correction module, the reflectivity image generation model further includes an accuracy verification module, and the reflectivity image production module 402 includes:

[0134] The first correction accuracy acquisition unit is used to obtain the correction accuracy according to the secondary corrected aerial surface reflectivity image and the measured reflectivity of the target object through the accuracy verification module.

[0135] In an optional implementation, after the secondary correction model and before the accuracy verification module, the reflectivity image generation model further includes a terrain judgment module and a terrain correction module, and the surface reflectivity image production module 402 further includes:

[0136] a first terrain determination unit, configured to determine whether the terrain in the aero-optical image is a mountainous terrain by means of the terrain determination module; if the terrain in the aero-optical image is a mountainous terrain, inputting the secondary corrected aero-surface reflectivity image into the terrain correction module; otherwise, inputting the secondary corrected aero-surface reflectivity image into the accuracy verification module;

[0137] A first terrain correction unit, configured to obtain a terrain-corrected aerial surface reflectivity image according to the secondary-corrected aerial surface reflectivity image and a terrain correction factor through the terrain correction module;

[0138] The second correction accuracy acquisition unit is used to obtain the correction accuracy according to the aerial surface reflectivity image after the terrain correction and the measured reflectivity of the target object through the accuracy verification module.

[0139] In an optional implementation, the surface reflectance image generation model further includes an accuracy verification module after the secondary correction module, and the surface reflectance image production module 402 further includes:

[0140] a second terrain determination unit, configured to determine whether the terrain in the aero-optical image is a mountainous terrain by using the terrain determination module; if the terrain in the aero-optical image is a mountainous terrain, input the secondary corrected aero-surface reflectivity image to the terrain correction module; otherwise, output the secondary corrected aero-surface reflectivity image;

[0141] The second terrain correction unit is used to obtain the terrain-corrected aerial surface reflectivity image according to the secondary-corrected aerial surface reflectivity image and the terrain correction factor through the terrain correction module.

[0142] In an optional implementation, the surface reflectance image generating unit includes:

[0143] A correction coefficient generating subunit is used to fit the initial aerial surface reflectivity image and the measured reflectivity of the target to obtain a correction coefficient;

[0144] The aerial surface reflectivity image generation subunit is used to obtain the secondary corrected aerial surface reflectivity image based on the correction coefficient of the initial aerial surface reflectivity image.

[0145] Based on the aerial surface reflectivity image production method and device provided in the aforementioned embodiments, the present application also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements some or all of the steps in the aerial surface reflectivity image production method mentioned above.

[0146] Based on the method and device for producing an aerial surface reflectivity image provided in the above-mentioned embodiment, the present application further provides an electronic device, including:

[0147] a memory having a computer program stored thereon;

[0148] The processor is used to execute the computer program in the memory to implement part or all of the steps in the method for producing aerial surface reflectance image provided in the above embodiment.

[0149] Based on the aerial surface reflectance image production method and device provided in the aforementioned embodiments, the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements some or all of the steps in the aerial surface reflectance image production method mentioned above.

[0150] It should be noted that each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is merely schematic, in which the units described as separate components may or may not be physically separated, and the components indicated 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 can 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 work.

[0151] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for producing an aerial surface reflectance image, characterized in that: The method comprises: Acquire an aerial optical image and a ground object spectrum corresponding to the aerial optical image; An aerial surface reflectance image is obtained through a reflectance image generation model according to the aerial optical image and the ground object spectrum.

2. The method according to claim 1, characterized in that The reflectance image generation model includes: a data preprocessing module, a hyperspectral data judgment module, a spectral correction module, a fast atmospheric correction module and a secondary correction module; The method of generating a reflectivity image model to obtain an aerial surface reflectivity image according to the aerial optical image and the ground object spectrum includes: The data preprocessing module performs preprocessing operations on the aerial optical image and the ground object spectrum to obtain preprocessed aerial optical images and preprocessed ground object spectrum; the preprocessing operations include: data screening, format conversion, field of view stitching and image non-uniformity correction; By means of the hyperspectral judgment module, it is judged whether the preprocessed aero-optical image is hyperspectral data; if it is determined that the preprocessed aero-optical image is hyperspectral data, the preprocessed aero-optical image is input into the spectral correction module; otherwise, the preprocessed aero-optical image is input into the rapid atmospheric correction module; Obtaining a spectrally corrected aero-optical image according to the pre-processed aero-optical image by means of the spectral correction module; Obtaining the initial aerial surface reflectance image according to the preprocessed aerial optical image or the spectrally corrected aerial optical image through the rapid atmospheric correction module; The secondary correction module is used to obtain a secondary corrected aerial surface reflectivity image based on the initial aerial surface reflectivity image and the measured reflectivity of the ground target.

3. The method according to claim 2, characterized in that After the secondary correction module, the reflectivity image generation model further includes an accuracy verification module, and the method further includes: The accuracy verification module is used to obtain the correction accuracy based on the secondary corrected aerial surface reflectivity image and the measured reflectivity of the target object.

4. The method according to claim 3, characterized in that After the secondary correction model and before the accuracy verification module, the reflectivity image generation model also includes a terrain judgment module and a terrain correction module; the method also includes: By means of the terrain judgment module, it is judged whether the terrain in the aerial optical image is a mountainous terrain; if the terrain in the aerial optical image is a mountainous terrain, the secondary corrected aerial surface reflectivity image is input into the terrain correction module; otherwise, the secondary corrected aerial surface reflectivity image is input into the accuracy verification module; Obtaining a terrain-corrected aerial surface reflectivity image according to the secondary-corrected aerial surface reflectivity image and the terrain correction factor through the terrain correction module; The method of obtaining the correction accuracy through the accuracy verification module according to the secondary corrected aerial surface reflectivity image and the measured reflectivity of the target object includes: The correction accuracy is obtained through the accuracy verification module according to the aerial surface reflectivity image after the terrain correction and the measured reflectivity of the target object.

5. The method according to claim 2, characterized in that: After the secondary correction model, the reflectivity image generation model further includes a terrain judgment module and a terrain correction module; the method further includes: By means of the terrain judgment module, it is judged whether the terrain in the aero-optical image is a mountainous terrain; if the terrain in the aero-optical image is a mountainous terrain, the aero-surface reflectivity image after the secondary correction is input into the terrain correction module; otherwise, the aero-surface reflectivity image after the secondary correction is output; The terrain correction module obtains the terrain-corrected aerial surface reflectivity image based on the secondary-corrected aerial surface reflectivity image and the terrain correction factor.

6. The method according to claim 2, characterized in that The method of obtaining a secondary-corrected aerial surface reflectivity image according to the initial aerial surface reflectivity image and the measured reflectivity of the ground target by the secondary correction module includes: Fitting the initial aerial surface reflectivity image and the measured reflectivity of the target to obtain a correction coefficient; Based on the correction coefficient and the initial aerial surface reflectivity image, the secondary corrected aerial surface reflectivity image is obtained.

7. An aerial surface reflectivity image production device, characterized in that: The device comprises: A parameter acquisition module, used to acquire an aerial optical image and a spectrum of a ground object corresponding to the aerial optical image; The surface reflectance image production module is used to generate a model through a reflectance image, and obtain an aerial surface reflectance image according to the aerial optical image and the ground object spectrum.

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

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images