Method for removing moisture influence of hyperspectral reflectivity data

Abstract

The invention provides a moisture influence removal method for hyperspectral reflectance data, and relates to the technical field of hyperspectral remote sensing, and the method comprises the steps: carrying out the spectrum denoising preprocessing of the obtained original hyperspectral reflectance data of a ground object in a natural state according to the noise, and obtaining the target hyperspectral reflectance data; calculating a moisture factor related to a short wave infrared band based on the target hyperspectral reflectivity data; and taking a matrix formed by moisture factors as a reference matrix, removing information characteristics related to the reference matrix from a spectral data matrix corresponding to the target hyperspectral reflectivity data to obtain corrected spectral data, and performing correlation calculation on the corrected spectral data and a standard spectral library to obtain a final corrected spectrum after moisture influence is removed. According to the method and the device, moisture removal can be carried out on the hyperspectral reflectivity data on the basis of keeping inherent characteristics of a ground feature spectrum and not depending on external actual measurement auxiliary data, so that the applicability of a subsequent spectrum application model is improved.

Description

Technical Field

[0001] The invention relates to the technical field of hyperspectral remote sensing, and in particular to a method for removing moisture influence from hyperspectral reflectance data. Background Art

[0002] Existing studies have shown that moisture has a nonlinear effect on hyperspectral reflectance data. Taking damp murals as an example, when the moisture content is low, the increase in moisture will reduce the spectral reflectance; when the moisture content is high, as the moisture further increases, the spectral reflectance will increase instead. Moisture not only changes the spectral reflectance value of the mural surface, but also affects the shape of the spectral curve, thereby interfering with the characteristic absorption peaks of mineral pigments, substrate components and other substances in the mural material, affecting the accuracy of information extraction in subsequent spectral application models. Therefore, removing the impact of moisture on hyperspectral reflectance data is an important prerequisite for ensuring that spectral data is more reliable, improving the applicability of subsequent spectral application models, and achieving accurate analysis and effective application.

[0003] Existing studies mostly use spectral transformation methods to remove the influence of environmental factors such as moisture. Among them, the DS and EPO algorithms are the most commonly used, but both require laboratory spectra to complete environmental factor correction. The principle of the DS algorithm is to establish a conversion matrix between the spectrum of the target object without moisture (referred to as dry spectrum) and the spectrum of the target object containing moisture (referred to as wet spectrum) through a standardized method, so that the wet spectrum and the dry spectrum are as similar as possible, thereby achieving the purpose of removing the influence of moisture. The principle of the EPO algorithm is to construct a conversion matrix through the difference matrix between the wet spectrum and the dry spectrum, separate the information related to moisture from the wet spectrum, and finally retain the information unrelated to moisture to achieve the purpose of removing the influence of moisture.

[0004] In the existing technology, moisture removal algorithms such as DS and EPO all rely on measured auxiliary data (laboratory spectra of dry target objects), which increases the complexity of data acquisition and processing. In large-scale image applications, it is difficult to obtain dry spectra, which limits the applicability of the method. In addition, the EPO algorithm will also change the shape of the spectrum, making it impossible to identify objects and extract information about objects by directly matching the spectrum with the standard spectral library, thereby affecting the accuracy of subsequent spectral application models for target recognition and information extraction. Summary of the invention

[0005] The present invention provides a method for removing moisture effects from hyperspectral reflectance data, which can remove moisture from hyperspectral reflectance data while maintaining inherent characteristics of ground object spectra and without relying on external measured auxiliary data, thereby improving the applicability of subsequent spectral application models.

[0006] The present invention provides a method for removing moisture influence of hyperspectral reflectance data, the method comprising the following steps: Obtain original hyperspectral reflectance data of ground objects in their natural state; Performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data; Calculating a moisture factor involving a shortwave infrared band based on the target hyperspectral reflectance data; The matrix composed of the moisture factors is used as a reference matrix, and the information features related to the reference matrix are removed from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data. The corrected spectral data is correlated with the standard spectral library to obtain the final corrected spectrum after removing the influence of moisture.

[0007] In some embodiments, performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data includes: Removing high-noise band spectral data affected by water vapor from the original hyperspectral reflectance data to obtain spectral data to be processed; Dividing the spectral data to be processed into spectral data of a high noise band, a low noise band, and a medium noise band according to the signal-to-noise ratio; The spectral data of the high noise band, low noise band and medium noise band are denoised respectively to obtain the target high spectral reflectance data; The denoising process includes: For the spectral data to be processed in the high noise band, the wavelet threshold denoising method is used for denoising; For the spectral data to be processed in the low-noise band, the moving average algorithm is used for denoising; The Savitzky-Golay filtering algorithm is used to denoise the spectral data in the medium noise band.

[0008] In some embodiments, the calculating of the moisture factor related to the shortwave infrared band based on the target hyperspectral reflectance data includes: Identify shortwave infrared bands that are sensitive to moisture content; For the target hyperspectral reflectance data, according to the shortwave infrared band, the shortwave infrared water stress index, the normalized shortwave infrared water index and the visible light and shortwave infrared dryness index are calculated as water factors.

[0009] In some embodiments, removing information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data includes: A singular value decomposition algorithm is used to extract characteristic components from a spectral data matrix corresponding to the target hyperspectral reflectance data; Constructing a projection matrix of the spectral data matrix in an orthogonal direction of the reference matrix; The characteristic components are feature mapped in the projection matrix to obtain corrected spectral data.

[0010] In some embodiments, the removing of information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data is achieved by an adaptive iterative optimization strategy, and the adaptive iterative optimization strategy includes: Preset the convergence threshold to remove the effect of moisture; If the current corrected spectrum data does not reach the convergence threshold, the following process is iteratively performed: removing information features associated with the reference matrix from the current corrected spectral data; If the current corrected spectrum data reaches the convergence threshold, the iteration ends.

[0011] In some embodiments, the correlation calculation of the corrected spectrum data with the standard spectrum library to obtain the final corrected spectrum after removing the influence of moisture includes: Calculating the correlation coefficient between the corrected spectral data and the spectra of the standard spectral library; The corrected spectrum data with the largest correlation coefficient between spectra is selected as the final corrected spectrum after removing the influence of moisture.

[0012] The present invention also provides a device for removing moisture influence of hyperspectral reflectance data, the device comprising the following modules: An acquisition module is used to obtain the original hyperspectral reflectance data of the ground object in its natural state; A preprocessing module, used for performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data; A calculation module, used for calculating a moisture factor involving a short-wave infrared band based on the target hyperspectral reflectance data; The removal module is used to use the matrix composed of the moisture factors as a reference matrix, remove the information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data, obtain corrected spectral data, and perform correlation calculation on the corrected spectral data and the standard spectral library to obtain the final corrected spectrum after removing the moisture influence.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for removing moisture influence from the hyperspectral reflectance data as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for removing moisture effects from hyperspectral reflectance data as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above methods for removing moisture effects from hyperspectral reflectance data.

[0016] The method for removing moisture effects from hyperspectral reflectance data provided by the present invention directly utilizes the pre-processed original hyperspectral reflectance data, calculates and analyzes the moisture factor that can accurately characterize the moisture content of the target object based on the short-wave infrared band, without relying on external measured auxiliary data, which greatly simplifies the data acquisition and processing process. The moisture effect can be removed by only using the moisture factor calculated from the hyperspectral reflectance data as a reference, which significantly reduces the dependence on external measured data, improves the applicability of subsequent spectral application models, and can keep the inherent characteristics of the ground object spectrum unchanged, and can provide technical support for the precise application of hyperspectral data in multiple fields such as multi-phase, multi-scene, large-scale quantitative remote sensing, information extraction and target recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a schematic flow chart of the method for removing moisture influence from hyperspectral reflectance data provided by the present invention.

[0019] Figure 2 It is a schematic structural diagram of a device for removing moisture effects on hyperspectral reflectance data provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The method for removing moisture influence from hyperspectral reflectance data of the present invention is described below in conjunction with the accompanying drawings. Figure 1 : is a schematic flow chart of the method for removing moisture influence from hyperspectral reflectance data provided by the present invention, such as Figure 1 As shown, the method includes the following steps 101 to 104, which are described in detail below: Step 101, obtaining original hyperspectral reflectance data of the ground object in a natural state.

[0023] First, the original hyperspectral reflectance data of the objects in their natural state is collected and obtained. During the collection process, a portable object spectrometer or imaging spectrometer is used to perform spectral measurement on the remote sensing objects in their natural state, and then the original hyperspectral reflectance data of the objects in the 400nm-2500nm band is obtained, where the natural state refers to the normal state without interference from other human factors, and the original hyperspectral reflectance data can be either point spectral reflectance data or a hyperspectral reflectance image.

[0024] Step 102: Perform spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data.

[0025] The original hyperspectral reflectance data obtained generally needs to be preprocessed, first removing some bands that are severely affected by noise, such as bands that are severely affected by water vapor. In addition, in order to retain the spectral information to the maximum extent, a denoising process is required. In the embodiment of the present invention, different denoising algorithms can be used to perform spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level of the spectral data to obtain the target hyperspectral reflectance data. The denoising algorithm can be a Savitzky-Golay filtering algorithm, a wavelet threshold denoising method, a moving average algorithm, etc.

[0026] Step 103: Calculate the moisture factor related to the shortwave infrared band based on the target hyperspectral reflectance data.

[0027] For the pre-processed target hyperspectral reflectance data, the embodiment of the present invention directly calculates the moisture factor to characterize the moisture content as accurately as possible. Shortwave infrared bands with higher sensitivity to moisture content are preferentially selected, and multiple moisture factors are calculated based on the selected shortwave infrared bands, including but not limited to: Shortwave Infrared Water Stress Index (SIWSI), Normalized Shortwave Infrared Water Index (NSWI) and Visible and Shortwave Infrared Drought Index (VSDI). When calculating the above moisture factors, all band combinations within the wavelength range involved are traversed and calculated one by one. In addition, the wavelength range used to calculate the moisture factor can be dynamically adjusted for different targets and application scenarios, effectively improving the practicality and stability of the moisture factor in different scenarios.

[0028] Step 104, using the matrix composed of moisture factors as a reference matrix, removing information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data, and performing correlation calculation between the corrected spectral data and the standard spectral library to obtain the final corrected spectrum after removing the influence of moisture.

[0029] After calculating the moisture factor of the spectral data, the matrix composed of the moisture factor is used as the reference matrix. Then, the information features related to the reference matrix are removed from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain the corrected spectral data, thereby eliminating the moisture features in the target hyperspectral reflectance data and achieving the purpose of removing the influence of moisture.

[0030] The key point of removal is to eliminate the interference information that is highly correlated with the moisture factor. The removal process can be to first extract the main features of the spectral data, and then perform orthogonal calculations on the features and the reference matrix to eliminate the feature information related to the reference matrix in the features, and obtain corrected spectral data. This can more accurately reflect the inherent spectral characteristics of the target object and provide high-quality data for subsequent hyperspectral information extraction and quantitative remote sensing applications.

[0031] Finally, the correlation between the corrected spectrum data and the standard spectrum library is calculated. The standard spectrum library stores a large amount of normal spectrum data that is not affected by external factors and does not contain any moisture content. By calculating the correlation coefficient between each set of corrected spectra and the standard spectrum library spectrum, the spectrum with the strongest correlation is selected as the final corrected spectrum after removing the influence of moisture.

[0032] The embodiment of the present invention directly uses the pre-processed original hyperspectral reflectance data to calculate and analyze the moisture factor that can accurately characterize the moisture content of the target object based on the short-wave infrared band, without relying on external measured auxiliary data, which greatly simplifies the data acquisition and processing process. Only by using the moisture factor calculated from the hyperspectral reflectance data as a reference, the moisture effect can be removed, which significantly reduces the dependence on external measured data, improves the applicability of subsequent spectral application models, and can keep the inherent characteristics of the ground object spectrum unchanged, and can provide technical support for the precise application of hyperspectral data in multiple fields such as multi-phase, multi-scene, large-scale quantitative remote sensing, information extraction and target recognition.

[0033] In some embodiments, spectral denoising preprocessing is performed on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data, which can be achieved through the following process.

[0034] First, it is necessary to remove the high-noise band spectral data affected by water vapor in the original hyperspectral reflectance data to obtain the spectral data to be processed. For example, the high-noise bands of the ground objects in the range of 1400nm to 1900nm or nearby are seriously affected by water vapor, and the spectral information cannot be accurately retained, so they can be directly removed.

[0035] Next, the spectral data to be processed is divided into spectral data of high noise band, low noise band and medium noise band according to the signal-to-noise ratio. Here, different noise levels can be divided according to the signal-to-noise ratio, and the spectral data to be processed can be divided into spectral data of high noise band, low noise band and medium noise band according to this noise level.

[0036] Finally, the spectral data of the high noise band, low noise band, and medium noise band are denoised respectively to obtain the target hyperspectral reflectance data. The denoising process can be: for the spectral data to be processed in the high noise band, the wavelet threshold denoising method is used for denoising, for the spectral data to be processed in the low noise band, the moving average algorithm is used for denoising, and for the spectral data to be processed in the medium noise band, the Savitzky-Golay filtering algorithm is used for denoising. Finally, the spectral data after denoising of each part of the band are combined to obtain the target hyperspectral reflectance data.

[0037] The embodiment of the present invention performs spectral denoising preprocessing on the original hyperspectral reflectance data, and further adopts different denoising algorithms to perform denoising according to different noise bands, so as to retain more spectral information to the maximum extent, effectively remove the influence of external noise factors on the spectral data, and improve the quality of the spectral data.

[0038] In some embodiments, the moisture factor is calculated based on the target hyperspectral reflectance data, specifically, the short-wave infrared band that is sensitive to the moisture content is first determined, and the range of this short-wave infrared band is generally 1000nm-2500nm. Within this band, the moisture content has a higher sensitivity, so this short-wave infrared band is selected to calculate the moisture factor.

[0039] Then, for the target hyperspectral reflectance data, the shortwave infrared water stress index, the normalized shortwave infrared water index, and the visible light and shortwave infrared dryness index are calculated according to the shortwave infrared band as water factors. In actual implementation, the calculated water factors include but are not limited to the shortwave infrared water stress index, the normalized shortwave infrared water index, and the visible light and shortwave infrared dryness index, which should be determined according to the specific scenario.

[0040] For the shortwave infrared water stress index (SIWSI), the calculation formula is as follows: (1) In the above formula (1), Represents the reflectivity of the short-wave infrared band (such as 1000nm-2500nm) in the target hyperspectral reflectivity data. Represents the reflectance of the near-infrared band (such as 700nm-1000nm) in the target hyperspectral reflectance data.

[0041] For the Normalized Shortwave Infrared Water Index (NSWI), the calculation formula is as follows: (2) In the above formula (2), Indicates the reflectance of the wavelength band of 1550nm-1750nm in the target hyperspectral reflectance data, Indicates the reflectance of the wavelength band of 2080nm-2350nm in the target hyperspectral reflectance data.

[0042] For the visible and shortwave infrared dryness index (VSDI), the calculation formula is as follows: (3) In the above formula (3), Represents the reflectivity of the short-wave infrared band (such as 1000nm-2500nm) in the target hyperspectral reflectivity data. Indicates the reflectance of the red light band (such as 627nm-700nm) in the target hyperspectral reflectance data. Indicates the reflectance of the blue light band (such as 446nm-524nm) in the target hyperspectral reflectance data.

[0043] When calculating the above moisture factor, all band combinations within the wavelength range involved are traversed and calculated one by one. In addition, the wavelength range used to calculate the moisture factor can be dynamically adjusted according to different targets and application scenarios, effectively improving the practicality and stability of the moisture factor in different scenarios.

[0044] After calculating the moisture factor, it is necessary to construct the corresponding reference matrix Y to characterize the degree of moisture influence. When constructing the reference matrix Y, one of the shortwave infrared moisture stress index, the normalized shortwave infrared moisture index, and the visible light and shortwave infrared dryness index can be selected as the moisture factor to directly construct the corresponding reference matrix Y, or the three indexes can be constructed into corresponding matrices respectively, and then the three corresponding matrix weights are set to weight the matrices. The three matrix weights can be the same or different, depending on the specific situation. Finally, the three weighted matrices are merged to obtain the final reference matrix Y. The construction method can be linear mapping to visualize the index into a matrix.

[0045] The embodiment of the present invention simultaneously calculates three moisture factors, namely the shortwave infrared moisture stress index, the normalized shortwave infrared moisture index, and the visible light and shortwave infrared dryness index, and performs traversal calculations on all band combinations within the wavelength range involved one by one. In addition, the wavelength range for calculating the moisture factor can be dynamically adjusted according to different target objects and application scenarios, so as to more accurately characterize the moisture content in the target hyperspectral reflectance data and facilitate subsequent moisture removal.

[0046] In some embodiments, after calculating the moisture factor as a reference matrix, it is necessary to remove information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data. This process can be achieved by following steps 201 to 203, which are described in detail below.

[0047] Step 201: extract characteristic components from the spectral data matrix corresponding to the target hyperspectral reflectance data using a singular value decomposition algorithm.

[0048] Here, the singular value decomposition algorithm (SVD) is used to extract the characteristic component from the spectral data matrix X corresponding to the target hyperspectral reflectance data. This characteristic component can characterize the main features of the target hyperspectral reflectance data. Of course, this main feature is affected by moisture, and the removal of the information features related to the reference matrix Y is calculated based on this main feature.

[0049] Step 202: construct a projection matrix of the spectral data matrix in a direction orthogonal to the reference matrix.

[0050] Based on the spectral data matrix X corresponding to the target hyperspectral reflectance data, a projection matrix in the orthogonal direction of the reference matrix Y is constructed, denoted as PROJ, which is expressed as the following formula: (4) In the above formula (4), I is the unit matrix and Y is the reference matrix.

[0051] Step 203: feature mapping the characteristic components in the projection matrix to obtain corrected spectral data.

[0052] Finally, the characteristic components extracted in step 201 are feature mapped in the projection matrix PROJ, thereby calculating the corresponding mapping features. The feature mapping process can be to directly perform a matrix multiplication operation on the matrix corresponding to the characteristic component and the projection matrix, and the obtained mapping feature is the result of removing the information features related to the reference matrix Y from the characteristic component, and the corresponding corrected spectral data can be restored based on this result.

[0053] The embodiment of the present invention can remove the influence of moisture in spectral data by only using the moisture factor calculated from the hyperspectral reflectance data as a reference, significantly reducing the dependence on external measured data, simplifying the spectral processing process, and improving the applicability of subsequent spectral application models. At the same time, it can keep the inherent characteristics of the ground object spectrum unchanged, and can provide technical support for the precise application of hyperspectral data in multiple fields such as multi-phase, multi-scene, large-scale quantitative remote sensing, information extraction and target recognition.

[0054] In some embodiments, the information features related to the reference matrix are removed from the spectral data matrix corresponding to the target hyperspectral reflectance data by an adaptive iterative optimization strategy. Because in actual environments, the corrected spectral data obtained only by the removal steps of steps 201 to 203 above often cannot completely and effectively remove the moisture features therein, so the embodiment of the present invention introduces an adaptive iterative optimization strategy to remove the moisture effect in the target hyperspectral reflectance data.

[0055] In the adaptive iterative optimization strategy, the convergence threshold for removing the influence of moisture needs to be preset first. This convergence threshold represents the effective standard to be achieved when removing the influence of moisture from the target hyperspectral reflectance data. When this convergence threshold is reached, it means that the target hyperspectral reflectance has effectively removed the moisture feature. The convergence threshold can be determined from normal spectral data that is not affected by moisture.

[0056] According to the adaptive iterative optimization strategy, after the correction spectrum data is calculated using the above steps 201 to 203, it is necessary to determine whether the current correction spectrum data meets the convergence threshold for removing the influence of moisture. If the current correction spectrum data does not reach the convergence threshold, the following process is iteratively performed: remove the information features related to the reference matrix from the current correction spectrum data. That is, continue to perform the moisture removal process of the above steps 201 to 203 based on the correction spectrum data after the current moisture. In each iteration, the correlation between the spectrum data matrix corresponding to the current correction spectrum data and the reference matrix Y can be gradually weakened. Through this continuous iterative process, the moisture interference information is gradually reduced until after a certain iteration process is completed, the current correction spectrum data obtained reaches the preset convergence threshold.

[0057] If the current corrected spectral data reaches the convergence threshold, the iteration ends. By iteratively executing the moisture removal steps from step 201 to step 203, the corrected spectral data obtained will gradually approach and meet the convergence threshold. When the convergence threshold is reached, it means that the effective removal of moisture has been completed, and the iteration can be stopped.

[0058] The embodiment of the present invention adopts an adaptive iterative optimization strategy and presets a convergence threshold for removing moisture effects to ensure that the target hyperspectral reflectance data can meet the moisture removal standard, thereby further improving the effectiveness of removing moisture effects from the corrected spectral data.

[0059] In some embodiments, after the moisture effect is removed, multiple correction spectrum features are obtained, so it is necessary to perform correlation calculation between the correction spectrum data and the standard spectrum library to obtain the final correction spectrum after the moisture effect is removed. In the embodiment of the present invention, the correction spectrum data is further selected through the standard spectrum library, and the standard spectrum library stores standard spectrum data without moisture effect. These spectrum data can be used as templates to further optimize the correction spectrum data, and the spectrum with the strongest correlation is selected as the final correction spectrum after the moisture effect is removed.

[0060] The specific process is as follows: first, the correlation coefficient between the corrected spectral data and the standard spectral library is calculated. The larger the correlation coefficient between the spectra, the closer the corrected spectral data is to the template. The correlation coefficient between the spectra can be calculated as the similarity of the spectral features. For example, the spectral features of the corrected spectral data and the template are extracted respectively, and then the similarity of the spectral features is calculated. The similarity can be the Euclidean distance or the cosine distance, etc., which is not limited here. The greater the similarity, the greater the correlation coefficient between the spectra.

[0061] The multiple corrected spectral features are obtained and the correlation coefficients between the spectra are calculated one by one with the standard spectral library, and finally the corrected spectral data with the largest correlation coefficient between the spectra are selected as the final corrected spectrum after removing the influence of moisture. In other words, the corrected spectral data with the highest spectral feature similarity is selected as the final corrected spectrum after removing the influence of moisture.

[0062] In the embodiment of the present invention, after removing moisture, the corrected spectral data is further optimized through a standard spectral library, which can further ensure that the moisture influence of the spectral data is accurately eliminated, effectively improve the quality of the spectral data, and provide technical support for the precise application of hyperspectral data in multiple fields such as multi-phase, multi-scene, large-scale quantitative remote sensing, information extraction and target recognition.

[0063] The device for removing moisture influence on hyperspectral reflectance data provided by the present invention is described below. The device for removing moisture influence on hyperspectral reflectance data described below and the method for removing moisture influence on hyperspectral reflectance data described above can be referred to each other.

[0064] like Figure 2 As shown, the device for removing moisture influence of hyperspectral reflectance data provided by the present invention specifically includes: an acquisition module 201, a preprocessing module 202, a calculation module 203, and a removal module 204. Among them, the acquisition module 201 is used to acquire the original hyperspectral reflectance data in the natural state of the ground object; the preprocessing module 202 is used to perform spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise size to obtain the target hyperspectral reflectance data; the calculation module 203 is used to calculate the moisture factor involving the short-wave infrared band based on the target hyperspectral reflectance data; the removal module 204 is used to use the matrix composed of the moisture factors as a reference matrix, remove the information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data, obtain the corrected spectral data, and perform correlation calculation on the corrected spectral data with the standard spectral library to obtain the final corrected spectrum after removing the moisture influence.

[0065] It should be noted that the beneficial effects of the device for removing moisture influence of hyperspectral reflectance data here and the method for removing moisture influence of hyperspectral reflectance data mentioned above can correspond to each other, so the beneficial effects of the device for removing moisture influence of hyperspectral reflectance data are not repeated here.

[0066] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the method for removing the moisture effect of the hyperspectral reflectance data, the method comprising: obtaining the original hyperspectral reflectance data in the natural state of the ground object; performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain the target hyperspectral reflectance data; calculating the moisture factor involving the short-wave infrared band based on the target hyperspectral reflectance data; taking the matrix composed of the moisture factors as the reference matrix, removing the information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain the corrected spectral data, and performing correlation calculation on the corrected spectral data and the standard spectral library to obtain the final corrected spectrum after removing the moisture effect.

[0067] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0068] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for removing moisture effects on hyperspectral reflectance data provided by the above methods, the method including: obtaining original hyperspectral reflectance data in a natural state of the object; performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise size to obtain target hyperspectral reflectance data; calculating moisture factors involving the shortwave infrared band based on the target hyperspectral reflectance data; using the matrix composed of the moisture factors as a reference matrix, removing information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data, and performing correlation calculation on the corrected spectral data with a standard spectral library to obtain the final corrected spectrum after removing the moisture effect.

[0069] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the method for removing moisture effects on hyperspectral reflectance data provided by the above-mentioned methods, the method comprising: obtaining original hyperspectral reflectance data of the ground object in a natural state; performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data; calculating moisture factors involving the shortwave infrared band based on the target hyperspectral reflectance data; using the matrix composed of the moisture factors as a reference matrix, removing information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data, and performing correlation calculation on the corrected spectral data with a standard spectral library to obtain a final corrected spectrum after removing the moisture effects.

[0070] 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.

[0071] 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, 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.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 invention.

Claims

1. A method for removing moisture effects from hyperspectral reflectance data, characterized in that: The method comprises: Obtain original hyperspectral reflectance data of ground objects in their natural state; Performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data; Calculating a moisture factor involving a shortwave infrared band based on the target hyperspectral reflectance data; The matrix composed of the moisture factors is used as a reference matrix, and the information features related to the reference matrix are removed from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data. The corrected spectral data is correlated with the standard spectral library to obtain the final corrected spectrum after removing the influence of moisture.

2. The method for removing moisture influence from hyperspectral reflectance data according to claim 1, characterized in that: The step of performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data includes: Removing high-noise band spectral data affected by water vapor from the original hyperspectral reflectance data to obtain spectral data to be processed; Dividing the spectral data to be processed into spectral data of a high noise band, a low noise band, and a medium noise band according to the signal-to-noise ratio; The spectral data of the high noise band, low noise band and medium noise band are denoised respectively to obtain the target high spectral reflectance data; The denoising process includes: For the spectral data to be processed in the high noise band, the wavelet threshold denoising method is used for denoising; For the spectral data to be processed in the low-noise band, the moving average algorithm is used for denoising; The Savitzky-Golay filtering algorithm is used to denoise the spectral data in the medium noise band.

3. The method for removing moisture effects from hyperspectral reflectance data according to claim 1, characterized in that: The calculation of the moisture factor involving the short-wave infrared band based on the target hyperspectral reflectance data includes: Identify shortwave infrared bands that are sensitive to moisture content; For the target hyperspectral reflectance data, according to the shortwave infrared band, the shortwave infrared water stress index, the normalized shortwave infrared water index and the visible light and shortwave infrared dryness index are calculated as water factors.

4. The method for removing moisture effects from hyperspectral reflectance data according to claim 1, characterized in that: The step of removing information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data to obtain corrected spectral data includes: A singular value decomposition algorithm is used to extract characteristic components from a spectral data matrix corresponding to the target hyperspectral reflectance data; Constructing a projection matrix of the spectral data matrix in an orthogonal direction of the reference matrix; The characteristic components are feature mapped in the projection matrix to obtain corrected spectral data.

5. The method for removing moisture influence from hyperspectral reflectance data according to claim 4, characterized in that: The removing of information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data is achieved by an adaptive iterative optimization strategy, which includes: Preset the convergence threshold to remove the effect of moisture; If the current corrected spectrum data does not reach the convergence threshold, the following process is iteratively performed: removing information features associated with the reference matrix from the current corrected spectral data; If the current corrected spectrum data reaches the convergence threshold, the iteration ends.

6. The method for removing moisture effects from hyperspectral reflectance data according to claim 1, characterized in that: The step of performing correlation calculation on the corrected spectrum data and the standard spectrum library to obtain a final corrected spectrum after removing the influence of moisture includes: Calculating the correlation coefficient between the corrected spectral data and the spectra of the standard spectral library; The corrected spectrum data with the largest correlation coefficient between spectra is selected as the final corrected spectrum after removing the influence of moisture.

7. A device for removing moisture effects from hyperspectral reflectance data, characterized in that: The device comprises: An acquisition module is used to obtain the original hyperspectral reflectance data of the ground object in its natural state; A preprocessing module, used for performing spectral denoising preprocessing on the original hyperspectral reflectance data according to the noise level to obtain target hyperspectral reflectance data; A calculation module, used for calculating a moisture factor involving a short-wave infrared band based on the target hyperspectral reflectance data; The removal module is used to use the matrix composed of the moisture factors as a reference matrix, remove the information features related to the reference matrix from the spectral data matrix corresponding to the target hyperspectral reflectance data, obtain corrected spectral data, and perform correlation calculation on the corrected spectral data and the standard spectral library to obtain the final corrected spectrum after removing the moisture influence.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method for removing moisture effects from the hyperspectral reflectance data according to any one of claims 1 to 6 is implemented.

9. 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 for removing moisture effects from hyperspectral reflectance data according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for removing moisture effects from hyperspectral reflectance data according to any one of claims 1 to 6 is implemented.

Images