A stitching hyperspectral camera non-uniform correction method and device and storage medium

Abstract

The application discloses a splicing hyperspectral camera non-uniform correction method and device and a storage medium, and the method comprises the following steps: separately performing non-uniform correction on each hyperspectral camera to obtain correction coefficients of each hyperspectral camera; and performing non-uniform correction between different cameras of the splicing hyperspectral camera by using an overlapping field of view, wherein when a spectral difference between a ground target in an overlapping area between different cameras and a ground target in a non-overlapping area is greater than a predetermined error value, a filtering algorithm is used to correct the correction coefficients between different cameras until the spectral difference is not greater than the predetermined error value, so that a non-uniformly corrected hyperspectral image is obtained. The application has a good non-uniform correction effect, can maximize the preservation of spectral characteristics, does not need to use a large-diameter lighting device, is less disturbed by external conditions during work, and does not need to perform statistical rule analysis on a large amount of data like a scene-based correction method.

Description

Technical Field

[0001] This invention relates to the field of stitched hyperspectral imaging technology, and in particular to a method, apparatus and storage medium for non-uniform correction of a stitched hyperspectral camera. Background Technology

[0002] Hyperspectral imaging technology can acquire spectral information while simultaneously obtaining images of the target's spatial dimension, and is widely used in precision agriculture, Earth observation, and environmental monitoring. Due to limitations in technology and manufacturing processes, a single hyperspectral camera cannot currently meet the application requirements of both a large field of view and high resolution. Therefore, stitched hyperspectral camera technology has emerged. Stitched hyperspectral camera technology combines the fields of view of multiple hyperspectral cameras to expand the effective field of view.

[0003] Non-uniformity in hyperspectral cameras refers to the inconsistency in the output of individual pixels of the detector under the same uniform radiation source. This non-uniformity mainly stems from factors such as detector pixel response, optical systems, and electronic systems. For mosaic hyperspectral cameras, in addition to the non-uniformity of individual cameras, there is also non-uniformity between cameras. Non-uniformity correction is crucial for improving the quality of hyperspectral data.

[0004] Currently, non-uniformity correction is mainly divided into calibration-based correction methods and scene-based correction methods. Calibration-based correction methods mainly include laboratory calibration and ground-based field calibration, which are performed by obtaining relative radiometric calibration coefficients. Scene-based correction methods mainly include histogram matching and moment matching, which, under the assumption that the signals acquired by each pixel of the imaging system's detector have the same distribution, use statistical laws to perform non-uniformity correction. However, for stitched large-field-of-view hyperspectral cameras, calibration-based correction methods require illumination equipment with sufficient aperture and are easily affected by external conditions during operation; the statistical premise of scene-based correction methods only holds true when a sufficient amount of data is acquired in the pushbroom direction, which also has certain limitations.

[0005] In view of this, the present invention is hereby proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a method, apparatus, and storage medium for non-uniformity correction of a stitched hyperspectral camera, thereby solving the aforementioned technical problems existing in the prior art. This invention has excellent non-uniformity correction performance, preserves spectral characteristics to the greatest extent, and eliminates the need for large-aperture illumination equipment. It is also less affected by external conditions during operation and does not require the large amount of data for statistical analysis as is necessary for scene-based correction methods.

[0007] The objective of this invention is achieved through the following technical solution:

[0008] A method for non-uniformity correction of a stitched hyperspectral camera, characterized by comprising the following steps:

[0009] Step S1: Perform non-uniformity correction on each hyperspectral camera individually to obtain the correction coefficient for each hyperspectral camera individually;

[0010] Step S2: Use overlapping fields of view to perform non-uniform correction between different cameras of the stitched hyperspectral camera. When the spectral difference between the ground target in the overlapping area and the ground target in the non-overlapping area between different cameras is greater than a predetermined error value, use a filtering algorithm to correct the correction coefficients between different cameras until the spectral difference is no greater than the predetermined error value, thereby obtaining a hyperspectral image after non-uniform correction.

[0011] In step S1, non-uniformity correction is performed on each hyperspectral camera individually using a radiometric calibration correction method. The gain coefficient and bias coefficient of each pixel of the individual hyperspectral camera detector are corrected, and the gain coefficient and bias coefficient of each pixel are used as the correction coefficients for each individual hyperspectral camera.

[0012] The radiation calibration method includes at least one of the following: one-point calibration method, two-point calibration method, and two-point multi-segment calibration method.

[0013] The method of performing non-uniformity correction on each hyperspectral camera individually using radiometric calibration includes: using an integrating sphere, adjusting it to different brightness levels, taking pictures of the integrating sphere with each individual hyperspectral camera, and then calibrating the acquired data to correct the gain coefficient and bias coefficient of each pixel of the individual hyperspectral camera detector, and using the gain coefficient and bias coefficient of each pixel as the correction coefficient for each individual hyperspectral camera.

[0014] The method of using overlapping fields of view to perform non-uniformity correction between different cameras of a stitched hyperspectral camera includes: taking one of the hyperspectral cameras as a reference, combining the overlapping fields of view between different cameras, and using the characteristic that the response of the overlapping fields of view should be consistent to perform non-uniformity correction.

[0015] The filtering algorithms include wavelet decomposition, Fourier filtering, or smoothing filtering.

[0016] A non-uniformity correction device for a stitched hyperspectral camera, the device comprising a separate correction module and a joint correction module;

[0017] The individual correction module is used to perform non-uniformity correction on each hyperspectral camera individually, thereby obtaining the correction coefficient for each hyperspectral camera individually.

[0018] The joint correction module is used to perform non-uniform correction between different cameras of the stitched hyperspectral camera using overlapping fields of view. When the spectral difference between the ground target in the overlapping area and the ground target in the non-overlapping area between different cameras is greater than a predetermined error value, the correction coefficient between different cameras is corrected by a filtering algorithm until the spectral difference is no greater than the predetermined error value, thereby obtaining a hyperspectral image after non-uniform correction.

[0019] A computer-readable storage medium for storing a computer program that, when executed by a processor, implements the above-described method.

[0020] Compared with existing technologies, this invention creatively first performs non-uniformity correction on each hyperspectral camera individually, thereby obtaining the correction coefficients for each individual hyperspectral camera. Then, it uses overlapping fields of view to perform non-uniformity correction between different cameras in a stitched hyperspectral camera system. When the spectral difference between ground targets in the overlapping area and the non-overlapping area of ​​different cameras exceeds a predetermined error value, a certain amount of interference is introduced. A filtering algorithm is then used to correct the correction coefficients between different cameras until the spectral difference does not exceed the predetermined error value, thus obtaining a non-uniformly corrected hyperspectral image. This method has good non-uniformity correction effect, preserves spectral characteristics to the greatest extent, does not require large-aperture lighting equipment, is less affected by external conditions during operation, and does not require a large amount of data for statistical analysis like scene-based correction methods. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic flowchart of the non-uniformity correction method for a stitched hyperspectral camera provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them, and do not constitute a limitation on the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0024] First, the following explanations are provided for the terms that may be used in this article:

[0025] The terms “including,” “comprising,” “containing,” “having,” or other similar semantic descriptions should be interpreted as non-exclusive inclusion. For example, “including a technical feature element (such as raw material, component, ingredient, carrier, dosage form, material, size, part, component, mechanism, device, step, process, method, reaction conditions, processing conditions, parameter, algorithm, signal, data, product or article of manufacture, etc.)” should be interpreted as including not only the expressly listed technical feature element, but also other technical feature elements that are not expressly listed and are well-known in the art.

[0026] The following is a detailed description of the non-uniformity correction method, apparatus, and storage medium for a stitched hyperspectral camera provided by this invention. Contents not described in detail in the embodiments of this invention are prior art known to those skilled in the art. Where specific conditions are not specified in the embodiments of this invention, they should be performed according to conventional conditions in the art or conditions recommended by the manufacturer. Reagents or instruments used in the embodiments of this invention, unless otherwise specified by the manufacturer, are all commercially available conventional products.

[0027] like Figure 1 As shown, this invention provides a method for non-uniformity correction of a stitched hyperspectral camera, used to correct the non-uniformity of a stitched hyperspectral camera, which specifically includes the following steps:

[0028] Step S1: Perform non-uniformity correction on each hyperspectral camera individually to obtain the correction coefficient for each individual hyperspectral camera.

[0029] Step S2: Non-uniformity correction is performed between different cameras using an overlapping field of view on the stitched hyperspectral camera. When the spectral difference between ground targets in the overlapping area and those in the non-overlapping area exceeds a predetermined error value, a filtering algorithm is used to correct the correction coefficients between the different cameras until the spectral difference is no greater than the predetermined error value, thus obtaining a non-uniformly corrected hyperspectral image. The predetermined error value can be set to 10%. The predetermined error value can be adjusted according to the complexity of the actual ground target types. The more complex the ground target types, the smaller the predetermined error value can be; the simpler the ground target types, the larger the predetermined error value can be. In some specific embodiments of the present invention, in step S1, non-uniformity correction can be performed on each hyperspectral camera individually using a radiometric calibration correction method. This corrects the gain coefficient and bias coefficient of each pixel of the individual hyperspectral camera detector, and uses the gain coefficient and bias coefficient of each pixel as the correction coefficient for each individual hyperspectral camera. The radiometric calibration correction method includes at least one of a one-point correction method, a two-point correction method, and a two-point multi-segment correction method. In practical applications, the non-uniformity correction of each hyperspectral camera individually using the radiometric calibration correction method includes: using an integrating sphere, adjusting it to different brightness levels, taking pictures of the integrating sphere with a single hyperspectral camera, and then calibrating the acquired data to correct the gain coefficient and bias coefficient of each pixel of the single hyperspectral camera detector, and using the gain coefficient and bias coefficient of each pixel as the correction coefficient for each individual hyperspectral camera.

[0030] In some specific embodiments of the present invention, the non-uniformity correction between different cameras of a stitched hyperspectral camera using overlapping fields of view includes: taking one of the hyperspectral cameras as a reference, combining the overlapping fields of view between different cameras, and using the characteristic that the response of the overlapping fields of view should be consistent to perform non-uniformity correction. In practical applications, when the spectral difference between ground targets in the overlapping area and ground targets in the non-overlapping area between different cameras is greater than a predetermined error value, a certain amount of interference will be introduced. In order to eliminate errors and interference, filtering algorithms can be used to correct the correction coefficients between different cameras until the spectral difference is no greater than the predetermined error value; the filtering algorithm here can be a variety of filtering algorithms such as wavelet decomposition Fourier filtering or smoothing filtering in the prior art.

[0031] This invention also provides a non-uniformity correction device for a stitched hyperspectral camera. The device includes an individual correction module and a joint correction module. The individual correction module is used to perform non-uniformity correction on each hyperspectral camera individually, thereby obtaining the correction coefficients for each individual hyperspectral camera. The joint correction module is used to perform non-uniformity correction between different cameras of the stitched hyperspectral camera using an overlapping field of view. When the spectral difference between ground targets in the overlapping area and ground targets in the non-overlapping area between different cameras is greater than a predetermined error value, a filtering algorithm is used to correct the correction coefficients between different cameras until the spectral difference is no greater than the predetermined error value, thereby obtaining a non-uniformly corrected hyperspectral image.

[0032] Compared with existing technologies, this invention creatively first performs non-uniformity correction on each hyperspectral camera individually, thereby obtaining the correction coefficients for each individual hyperspectral camera. Then, it uses overlapping fields of view to perform non-uniformity correction between different cameras in a stitched hyperspectral camera system. When the spectral difference between ground targets in the overlapping area and the non-overlapping area of ​​different cameras exceeds a predetermined error value, a certain amount of interference is introduced. A filtering algorithm is then used to correct the correction coefficients between different cameras until the spectral difference does not exceed the predetermined error value, thus obtaining a non-uniformly corrected hyperspectral image. This method has good non-uniformity correction effect, preserves spectral characteristics to the greatest extent, does not require large-aperture lighting equipment, is less affected by external conditions during operation, and does not require a large amount of data for statistical analysis like scene-based correction methods.

[0033] In summary, the embodiments of the present invention have excellent non-uniformity correction effect, preserve spectral characteristics to the greatest extent, and do not require the use of large-aperture lighting equipment. They are less affected by external conditions during operation and do not require a large amount of data for statistical analysis as in scene-based correction methods.

[0034] To more clearly demonstrate the technical solution and its effects provided by the present invention, the following detailed description of the non-uniformity correction method, apparatus and storage medium for the stitched hyperspectral camera provided by the present invention is given with reference to specific embodiments.

[0035] Example 1

[0036] like Figure 1 As shown, a non-uniformity correction method for a stitched hyperspectral camera is used to correct the non-uniformity of the stitched hyperspectral camera, and may specifically include the following steps:

[0037] Step A1: Perform non-uniformity correction on each hyperspectral camera individually using the radiometric calibration correction method, correct the gain coefficient and bias coefficient of each pixel of the individual hyperspectral camera detector, and use the gain coefficient and bias coefficient of each pixel as the correction coefficient of each individual hyperspectral camera.

[0038] Step A2: Perform non-uniformity correction between different cameras using overlapping fields of view in a stitched hyperspectral camera system. Taking one hyperspectral camera as a reference, and combining the overlapping fields of view of different cameras, the non-uniformity correction is performed based on the characteristic that the response of the overlapping fields of view should be consistent. When the spectral difference between ground targets in the overlapping area and the non-overlapping area of ​​different cameras exceeds a predetermined error value, wavelet decomposition algorithm is used to correct the correction coefficients between the different cameras until the spectral difference is no greater than the predetermined error value, thus obtaining the non-uniformly corrected hyperspectral image. Here, the predetermined error value can be set to 10%. The predetermined error value can be adjusted according to the complexity of the actual ground target types. The more complex the ground target types, the smaller the predetermined error value can be; the simpler the ground target types, the larger the predetermined error value can be.

[0039] Example 2

[0040] A non-uniformity correction device for a stitched hyperspectral camera, the device comprising a separate correction module and a combined correction module.

[0041] The separate calibration module is used to implement step S1 of embodiment 1 above.

[0042] The joint correction module is used to implement step S2 of embodiment 1 above.

[0043] Example 3

[0044] A computer-readable storage medium for storing a computer program that, when executed by a processor, implements the method of Embodiment 1 described above.

[0045] In summary, the embodiments of the present invention have excellent non-uniformity correction effect, preserve spectral characteristics to the greatest extent, and do not require the use of large-aperture lighting equipment. They are less affected by external conditions during operation and do not require a large amount of data for statistical analysis as in scene-based correction methods.

[0046] The above embodiment 1 is merely an exemplary illustration of the technical solution of the present invention and should not be construed as a limitation of the present invention. Any obvious partial modifications to the present invention should be considered as alternatives to the present invention. Such alternatives include the selection of camera radiometric calibration correction methods and correction parameter filtering methods, etc. These modifications and variations do not depart from the essential scope of the present invention.

[0047] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims. The information disclosed in the background section is intended only to enhance the understanding of the overall background technology of the present invention and should not be construed as an admission or implication in any way that such information constitutes prior art known to those skilled in the art.

Claims

1. A method for non-uniformity correction of a stitched hyperspectral camera, characterized in that, Includes the following steps: Step S1: Perform non-uniformity correction on each hyperspectral camera individually to obtain the correction coefficient for each hyperspectral camera individually; In step S1, non-uniformity correction is performed on each hyperspectral camera individually using a radiometric calibration correction method. The gain coefficient and bias coefficient of each pixel of the detector of a single hyperspectral camera are corrected, and the gain coefficient and bias coefficient of each pixel are used as the correction coefficient of each hyperspectral camera individually. Step S2: Use overlapping fields of view to perform non-uniform correction between different cameras of the stitched hyperspectral camera. When the spectral difference between the ground target in the overlapping area and the ground target in the non-overlapping area between different cameras is greater than a predetermined error value, use a filtering algorithm to correct the correction coefficient between different cameras until the spectral difference is no greater than the predetermined error value, thereby obtaining a hyperspectral image after non-uniform correction. The method of using overlapping fields of view to perform non-uniformity correction between different cameras of a stitched hyperspectral camera includes: taking one of the hyperspectral cameras as a reference, combining the overlapping fields of view between different cameras, and using the characteristic that the response of the overlapping fields of view should be consistent to perform non-uniformity correction.

2. The non-uniformity correction method for a stitched hyperspectral camera according to claim 1, characterized in that, The radiation calibration method includes at least one of the following: one-point calibration method, two-point calibration method, and two-point multi-segment calibration method.

3. The non-uniformity correction method for a stitched hyperspectral camera according to claim 1, characterized in that, The method of performing non-uniformity correction on each hyperspectral camera individually using radiometric calibration includes: using an integrating sphere, adjusting it to different brightness levels, taking pictures of the integrating sphere with each individual hyperspectral camera, and then calibrating the acquired data to correct the gain coefficient and bias coefficient of each pixel of the individual hyperspectral camera detector, and using the gain coefficient and bias coefficient of each pixel as the correction coefficient for each individual hyperspectral camera.

4. The non-uniformity correction method for a stitched hyperspectral camera according to any one of claims 1 to 3, characterized in that, The filtering algorithms include wavelet decomposition, Fourier filtering, or smoothing filtering.

5. A non-uniformity correction device for a stitched hyperspectral camera, characterized in that, The non-uniformity correction method for a stitched hyperspectral camera according to any one of claims 1 to 4 is adopted; the non-uniformity correction device for the stitched hyperspectral camera includes a separate correction module and a joint correction module; The individual correction module is used to perform non-uniformity correction on each hyperspectral camera individually, thereby obtaining the correction coefficient for each hyperspectral camera individually. The joint correction module is used to perform non-uniform correction between different cameras of the stitched hyperspectral camera using overlapping fields of view. When the spectral difference between the ground target in the overlapping area and the ground target in the non-overlapping area between different cameras is greater than a predetermined error value, the correction coefficient between different cameras is corrected by a filtering algorithm until the spectral difference is no greater than the predetermined error value, thereby obtaining a hyperspectral image after non-uniform correction.

6. A computer-readable storage medium for storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 4.

Images