Dispersion type hyperspectral camera color distortion measuring device and measuring method

Abstract

The application provides a dispersion type hyperspectral camera color distortion measuring device and a measuring method, the device comprises a halogen lamp integrating sphere, a linear target, a collimator, a to-be-measured hyperspectral camera, a rotary table, a supporting leveling platform, a rotary table control system and a data acquisition and analysis unit; the to-be-measured hyperspectral camera comprises a telescope, a hyperspectral camera slit, a hyperspectral camera spectrometer assembly and a detector; the application reduces the difficulty of point-to-line alignment caused by simultaneous adjustment of pitch and azimuth, and simplifies the alignment requirement of optical path measurement; through selection of the target, the measurement difficulty is reduced, and the measurement precision is improved; the application can be widely applied to color distortion measurement of a dispersion type high-precision spectral imaging system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of optical imaging and spectral detection, and particularly relates to a dispersion type hyperspectral camera color distortion measuring device and a measuring method. BACKGROUND

[0002] The color distortion of a grating spectrometer is mainly caused by aberrations caused by off-axis of the optical system and defects of the grating itself, and the essence is the nonlinear change of the magnification of different wavelengths in the spatial dimension, which leads to the physical scale of the target region corresponding to the same spatial dimension pixel shifting with the wavelength. This nonlinearity may cause the spectrum received by a single pixel to contain information of adjacent spatial positions, causing spectral aliasing in adjacent spatial dimensions, which seriously affects the fidelity of high-resolution imaging spectrum. By measuring the color distortion, the wavelength calibration error can be corrected, the data reliability is improved, and it is crucial in application scenarios that require high-precision spectral measurement.

[0003] Patent CN112284537A describes a method for detecting spectral curvature and trapezoidal distortion of a push-broom hyperspectral imager. The method makes a spectral curvature and trapezoidal distortion detection target, collects dark background data and target imaging data, performs non-uniformity correction and radiation calibration, averages the data in the time dimension, and finally calculates the spectral curvature and trapezoidal distortion through a same point search algorithm, with a precision of sub-pixel level. The patent mainly introduces the measurement process of spectral curvature and trapezoidal distortion, and does not introduce the specific target design form, nor does it consider the difficulty brought by the two-dimensional adjustment of pitch and azimuth in optical path calibration. XUAN ZHANG et al. of North University in "Xuan Zhang, et al. "A design method for direct vision coaxial linear dispersion spectrometers." Optics Express 30.21 (2022): 38266-38283." proposed a spectrometer design method based on a prism-prism-grating (PPG) dispersion module, which is used to correct spectral nonlinear distortion. The method analyzes the dispersion characteristics of the prism and the grating, selects appropriate optical materials, and uses a loop traversal algorithm to optimize the system structure parameters. The spectral linearity index of the system is established through the dispersion equation of the PPG module, and the structure parameters of the PPG module are determined in combination with the design index to meet the linear dispersion requirement. However, this method has strong dependence on optical materials, and the selection of materials that do not meet the requirements may affect the measurement accuracy; at the same time, it has high requirements for environmental conditions and light source uniformity, and if the conditions are unstable, it may cause errors, making it difficult to cope with instrument aging or environmental changes in long-term operation. These limitations may limit its universality and reliability in practical applications.

[0004] In view of this, the present application provides a dispersion type hyperspectral camera color distortion measurement device and measurement method. SUMMARY

[0005] The present application aims to provide a dispersion type hyperspectral camera color distortion measurement device and measurement method, which reduces the difficulty of point-to-line alignment caused by simultaneous adjustment of pitch and azimuth, and simplifies the alignment requirements of optical path measurement. By selecting the target, the measurement difficulty is reduced and the measurement accuracy is improved. It can be widely applied to color distortion measurement of dispersion type high-precision spectral imaging system.

[0006] To achieve the above-mentioned purpose, the technical solutions of the present application are as follows:

[0007] In the first aspect, the present application provides a dispersion type hyperspectral camera color distortion measurement device, which comprises a halogen lamp integrating sphere, a one-dimensional target, a collimator, a to-be-measured hyperspectral camera, a rotary table, a support leveling platform, a rotary table control system and a data acquisition and analysis unit.

[0008] The to-be-measured hyperspectral camera comprises a telescope objective, a hyperspectral camera slit, a hyperspectral camera spectrometer assembly and a detector; the hyperspectral camera slit is arranged orthogically to the opening length direction of the one-dimensional target, and the overlapping area of the hyperspectral camera slit and the one-dimensional target opening is one pixel size.

[0009] The rotation axis of the rotary table is perpendicular to the opening length direction of the one-dimensional target, the to-be-measured hyperspectral camera is connected with the rotary table and placed on the support leveling platform through the rotary table, so as to ensure that the optical axis of the to-be-measured hyperspectral camera is always in the same plane as the optical axes of the halogen lamp integrating sphere, the one-dimensional target and the collimator during rotation around the rotation axis; the optical axes of the halogen lamp integrating sphere, the one-dimensional target, the collimator and the to-be-measured hyperspectral camera in the initial state are all on the same straight line, and the one-dimensional target is placed at the focal plane of the collimator, so as to ensure that the wide spectrum light emitted by the halogen lamp integrating sphere is sequentially incident to the to-be-measured hyperspectral camera after passing through the one-dimensional target and the collimator to form a parallel light beam, and then sequentially passes through the telescope objective, the hyperspectral camera slit and the hyperspectral camera spectrometer assembly, and finally forms discrete point images on the detector according to wavelength separation.

[0010] The rotary table control system and the data acquisition and analysis unit are used to control the whole mechanism after the to-be-measured hyperspectral camera is connected with the rotary table, to rotate the to-be-measured hyperspectral camera relative to the rotation axis within a preset angle range while performing image acquisition, and to calculate the difference of different wavelength centroid coordinates according to the acquired images to obtain the color distortion.

[0011] Preferably, the width of the opening of the one-dimensional target is 1-10mm. d' The calculation of the width of the opening of the one-dimensional target is specifically as follows:

[0012]

[0013] In the formula, f 1 is the focal length of the collimator; f 2 is the focal length of the hyperspectral camera to be measured; d is the slit width of the hyperspectral camera slit.

[0014] Preferably, the support leveling platform is a three-point support leveling platform.

[0015] Preferably, the gas float optical platform is further included, and the halogen lamp integrating sphere, the linear target, and the collimator are fixedly installed on the gas float optical platform, so as to isolate and eliminate external vibration on the halogen lamp integrating sphere, the linear target, and the collimator through the gas float optical platform, and maintain the relative position stability among the three.

[0016] In a second aspect, the present application provides a method for measuring color distortion of a dispersive hyperspectral camera, and the method is implemented by using any of the above-mentioned color distortion measuring devices of the dispersive hyperspectral camera, and specifically includes the following steps:

[0017] S1, turn on the halogen lamp integrating sphere, and after the emitted wide-spectrum light forms a linear bright light through the linear target, the linear bright light forms a parallel light beam through the collimator and is incident to the hyperspectral camera to be measured;

[0018] S2, control the rotation of the turntable to drive the hyperspectral camera to be measured to rotate and scan to cover the full field of view, synchronously trigger the hyperspectral camera to collect n frames of spectral-space two-dimensional images under the same field of view at each position, after completion, shield the entrance pupil of the hyperspectral camera to be measured, continuously collect m frames of images, and pre-process all original image data;

[0019] S3, take the average value of the m frames of images collected after the pre-processing of shielding the entrance pupil of the hyperspectral camera to be measured as the average noise brightness value of the image plane;

[0020] S4, after the n frames of images collected under the same field of view after the pre-processing are subjected to time domain median filtering to eliminate random noise, the spatial dimension centroid coordinates of each wavelength point image are extracted in combination with the average noise brightness value of the image plane, color distortion value calculation is performed, and a full-band color distortion distribution curve is generated.

[0021] Preferably, the pre-processing specifically includes band-pass filtering and pixel response non-uniformity correction on the original image data.

[0022] Preferably, the linear target opening length direction is parallel to the ground, the hyperspectral camera slit is perpendicular to the ground, and the collected images have a horizontal axis representing the spectral dimension and a vertical axis representing the spatial dimension.

[0023] Preferably, the spatial dimension centroid coordinates of each wavelength point image are extracted in combination with the average noise brightness value of the image plane, and the calculation is as follows:

[0024]

[0025] In the formula, y wavelenth is the centroid coordinate of the image formed after scanning the point image along the slit direction at a certain wavelength, y i is the average spatial dimension coordinate of the nth acquisition of the mth field of view, i DN i is the average spatial dimension pixel brightness value of the nth acquisition of the mth field of view, i DN noise is the average noise brightness value of the image plane.

[0026] Preferably, the color distortion value is calculated as follows:

[0027]

[0028] In the formula, y smile is the color distortion between wavelength λ 1 and wavelength λ 2, and are the spatial dimension centroid coordinates of wavelength λ 1 and wavelength λ 2, respectively.

[0029] In a third aspect, the present application provides a computer readable storage medium storing a computer program, which, when executed by a processor, implements any of the above color dispersion type hyperspectral camera color distortion measurement methods.

[0030] Compared with the prior art, the present application has the following beneficial effects:

[0031] Firstly, the present application realizes accurate measurement through high-precision linear target and uniform illumination, avoiding the dependence on complex reference targets and reducing the requirements for target design and manufacturing. Secondly, the present application simplifies the data processing process through turntable scanning and centroid calculation. At the same time, the device design is flexible and can be dynamically adjusted to adapt to environmental changes and instrument aging problems, improving the stability and reliability of the measurement. The present application can be used for color distortion measurement of hyperspectral cameras, especially in scenes requiring high-precision spectral imaging, which has important value. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 is the structure diagram of the device of the present application;

[0033] Figure 2 is the structure diagram of the hyperspectral camera to be measured of the present application; ​​

[0034] Figure 3 For the method flowchart of the present application;

[0035] Figure 4 For the image coordinate axis schematic diagram of the present application.

[0036] In the figure:

[0037] 1-halogen lamp integrating sphere; 2-linear target; 3-parallel light pipe; 4-high spectral camera to be measured; 5-rotary table; 6-supporting and leveling platform; 7-rotary table control system and data acquisition and analysis unit; 8-telescopic objective; 9-high spectral camera slit; 10-high spectral camera light splitting assembly; 11-detector. DETAILED DESCRIPTION

[0038] The technical solutions of the present application will be specifically described below in combination with the accompanying Figures 1-4 The technical solutions of the present application will be specifically described below in combination with the accompanying

[0039] The present application provides a color distortion measurement device for a dispersive hyperspectral camera, which is suitable for color distortion measurement of a hyperspectral camera with planar, concave and convex grating imaging. Figure 1 The device comprises a halogen lamp integrating sphere 1, a linear target 2, a parallel light pipe 3, a hyperspectral camera to be measured 4, a rotary table 5, a supporting and leveling platform 6, and a rotary table control system and data acquisition and analysis unit 7.

[0040] As shown in Figure 2 The hyperspectral camera to be measured 4 comprises a telescopic objective 8, a high spectral camera slit 9, a high spectral camera light splitting assembly 10, and a detector 11; the high spectral camera slit 9 is arranged orthogonally to the opening length direction of the linear target 2, and the overlapping area of the high spectral camera slit 9 and the opening of the linear target 2 is one pixel size.

[0041] The rotary shaft of the rotary table 5 is perpendicular to the opening length direction of the linear target 2; the hyperspectral camera to be measured 4 is connected with the rotary table 5 and placed on the supporting and leveling platform 6 through the rotary table 5, so as to ensure that the optical axis of the hyperspectral camera to be measured 4 is always in the same plane as the optical axes of the halogen lamp integrating sphere 1, the linear target 2 and the parallel light pipe 3 during rotation around the rotary shaft through the supporting and leveling platform 6; the optical axes of the halogen lamp integrating sphere 1, the linear target 2, the parallel light pipe 3 and the hyperspectral camera to be measured 4 in the initial state are all on the same straight line, and the linear target 2 is placed at the focal plane of the parallel light pipe 3, so as to ensure that the wide spectral range light emitted by the halogen lamp integrating sphere 1 is sequentially incident to the hyperspectral camera to be measured 4 after forming a parallel light beam through the linear target 2 and the parallel light pipe 3, and is sequentially passed through the telescopic objective 8, the high spectral camera slit 9 and the high spectral camera light splitting assembly 10, and finally forms discrete point images on the detector 11 according to wavelength separation;

[0042] The rotary table control system and the data acquisition and analysis unit 7 are used to control the whole mechanism after the to-be-tested hyperspectral camera 4 is connected with the rotary table 5, the to-be-tested hyperspectral camera 4 is driven to rotate relative to the rotary shaft in a preset angle range while image acquisition is carried out, and the difference of the centroid coordinates of different wavelengths is calculated according to the acquired images to obtain the color distortion variable.

[0043] In the embodiment, the width of the opening of the linear target 2 is d' The calculation is specifically as follows:

[0044]

[0045] In the formula, f 1 is the focal length of the collimator 3; f 2 is the focal length of the to-be-tested hyperspectral camera 4; d is the slit width of the hyperspectral camera slit 9.

[0046] In the embodiment, the support leveling platform 6 is a three-point support leveling platform.

[0047] In the embodiment, the air floating optical platform is further included, the halogen lamp integrating sphere 1, the linear target 2 and the collimator 3 are fixedly installed on the air floating optical platform; the air floating optical platform is suspended by an air film to isolate and eliminate external vibration, and the relative position stability between the halogen lamp integrating sphere 1, the linear target 2 and the collimator 3 is maintained.

[0048] The application provides a color dispersion type hyperspectral camera color distortion measurement method, and the method is realized by using any one of the color dispersion type hyperspectral camera color distortion measurement devices. Figure 3 The method specifically includes the following steps:

[0049] S1, the halogen lamp integrating sphere 1 is turned on, the wide-spectrum light emitted by the halogen lamp integrating sphere 1 forms linear bright light after passing through the linear target 2, and then the linear bright light forms a parallel light beam by the collimator 3 and is incident to the to-be-tested hyperspectral camera 4;

[0050] S2, the rotary table 5 is controlled to drive the to-be-tested hyperspectral camera 4 to rotate and scan to cover the full field of view (at this time, a series of point images of different wavelengths can be observed to move along the spatial dimension on the hyperspectral camera detector, and when the point image scans to the edge of the slit, the edge of the field of view is obtained), and the hyperspectral camera is synchronously triggered to collect n frames of spectral-space two-dimensional images under the same field of view at each position; after the collection is completed, the entrance pupil of the to-be-tested hyperspectral camera 4 is shielded, m frames of images are continuously collected, and all original image data are preprocessed; in the embodiment, n=10 and m=50.

[0051] S3, the average value of the m frames of images collected after the entrance pupil of the to-be-tested hyperspectral camera 4 is shielded is taken as the average noise brightness value of the image plane.

[0052] S4. For the n frames of images collected under the same field of view after preprocessing, random noise is eliminated by time domain median filtering, and the spatial dimension centroid coordinates of the point image of each wavelength are extracted in combination with the average noise brightness value of the image plane. The color distortion value is calculated to generate the full-band color distortion distribution curve.

[0053] In this embodiment, the preprocessing specifically involves performing bandpass filtering and pixel response non-uniformity correction on the original image data.

[0054] In this embodiment, the opening length direction of the inline target 2 is parallel to the earth, and the slit 9 of the hyperspectral camera is perpendicular to the earth. A high-precision theodolite can be used to calibrate the inline target 2 so that the opening length direction is parallel to the earth; and a high-precision theodolite can be used to calibrate the turntable surface so that the slit 9 of the hyperspectral camera is perpendicular to the earth. Figure 4 The horizontal axis of the acquired image represents the spectral dimension, and the vertical axis represents the spatial dimension. Each row of points represents point images of the same field of view at different wavelengths, and each column of points represents point images of the same wavelength at different fields of view. Wavelengths outside the spectral range of the hyperspectral camera are not considered.

[0055] In this embodiment, the spatial dimension centroid coordinates of the image of each wavelength point are extracted by combining the average noise brightness value of the image plane, and the specific calculation is as follows:

[0056]

[0057] Where, y wavelenth is the centroid coordinate of the image formed by scanning the point image along the slit direction at a certain wavelength, y i For the i The average spatial coordinates of n acquisitions in the field of view, DN i For the i The average spatial pixel brightness value collected n times in the field of view, DN noise is the average noise brightness value of the image plane.

[0058] In this embodiment, the color distortion value is calculated as follows:

[0059]

[0060] Where, y smile is the wavelength λ 1 and wavelength λ Color distortion between 2, and Wavelength λ 1 and wavelength λ 2-dimensional center-of-mass coordinates.

[0061] The application provides a computer readable storage medium storing a computer program, the computer program being executed by a processor to implement any above-mentioned color dispersion type hyperspectral camera color distortion measurement method.

[0062] In conclusion, in view of the problems of the prior art, such as dependence on complex targets, complicated data processing, and weak anti-environmental interference capability, the application realizes efficient and accurate color distortion measurement through high-precision linear targets, uniform illumination systems, turntable scanning, and centroid calculation methods. The device comprises a halogen lamp integrating sphere, a linear target, a collimator, a to-be-measured hyperspectral camera, a turntable, and a data acquisition and analysis unit. The collimator generates an infinite remote complex light image of the linear target. The slit of the color dispersion type hyperspectral camera is orthogonal to the linear target, and the intersection of the slit and the linear target generates a single-point dispersion image. Combined with full-field scanning of the turntable and centroid coordinate difference calculation, the full-band color distortion distribution curve under each spatial field of view is obtained. The advantage of the application lies in reducing the difficulty of point-to-line alignment caused by simultaneous adjustment of pitch and azimuth, and simplifying the alignment requirements of the optical path measurement. Through selection of the target, the measurement difficulty is reduced, and the measurement accuracy is improved. The application can be widely applied to color distortion measurement of color dispersion type high-precision spectral imaging systems.

[0063] The above is the preferred embodiment of the application, and any changes made according to the technical solutions of the application, as long as the generated functions do not exceed the scope of the technical solutions of the application, belong to the protection scope of the application.

Claims

1. A method for measuring chromatic aberration of a dispersive hyperspectral camera, the method comprising: The color distortion measurement method is realized by a color dispersion type hyperspectral camera color distortion measurement device, and includes a halogen lamp integrating sphere (1), a one-dimensional target (2), a collimator (3), a to-be-measured hyperspectral camera (4), a rotary table (5), a supporting and leveling platform (6), and a rotary table control system and data acquisition and analysis unit (7); The to-be-measured hyperspectral camera (4) includes a telescope objective (8), a hyperspectral camera slit (9), a hyperspectral camera light splitting assembly (10), and a detector (11); the hyperspectral camera slit (9) is arranged orthogonally to the opening length direction of the one-dimensional target (2), and the overlapping area of the hyperspectral camera slit (9) and the opening of the one-dimensional target (2) is one pixel size; The rotary shaft of the rotary table (5) is perpendicular to the opening length direction of the one-dimensional target (2), the to-be-measured hyperspectral camera (4) is connected with the rotary table (5) and placed on the supporting and leveling platform (6) through the rotary table (5), so as to ensure that the optical axis of the to-be-measured hyperspectral camera (4) is always in the same plane as the optical axes of the halogen lamp integrating sphere (1), the one-dimensional target (2) and the collimator (3) during rotation around the rotary shaft; the optical axes of the halogen lamp integrating sphere (1), the one-dimensional target (2), the collimator (3) and the to-be-measured hyperspectral camera (4) in the initial state are all on the same straight line, and the one-dimensional target (2) is placed at the focal plane of the collimator (3), so as to ensure that the wide spectrum light emitted by the halogen lamp integrating sphere (1) enters the to-be-measured hyperspectral camera (4) after forming a parallel light beam by passing through the one-dimensional target (2) and the collimator (3) in turn, and then passes through the telescope objective (8), the hyperspectral camera slit (9) and the hyperspectral camera light splitting assembly (10) in turn, and finally forms discrete point images on the detector (11) according to wavelength separation; The rotary table control system and data acquisition and analysis unit (7) is used for controlling the whole mechanism after the to-be-measured hyperspectral camera (4) is connected with the rotary table (5), and the to-be-measured hyperspectral camera (4) is driven to rotate relative to the rotary shaft in a preset angle range while image acquisition is performed, and the difference of the centroid coordinates of different wavelengths is calculated according to the acquired images to obtain the color distortion variable; The color dispersion type hyperspectral camera color distortion measurement method specifically includes the following steps: S1, the halogen lamp integrating sphere (1) is turned on, the wide spectrum light emitted by the halogen lamp integrating sphere (1) forms a one-dimensional bright light after passing through the one-dimensional target (2), and then forms a parallel light beam by passing through the collimator (3) and enters the to-be-measured hyperspectral camera (4); S2, the rotary table (5) is controlled to rotate to drive the to-be-measured hyperspectral camera (4) to rotate and scan to cover the full field of view, and the hyperspectral camera is triggered to acquire n frames of spectral-space two-dimensional images at the same field of view at each position synchronously; after completion, the entrance pupil of the to-be-measured hyperspectral camera is blocked, and m frames of images are continuously acquired; and all original image data is preprocessed; S3, the average value of the m frames of images acquired after the to-be-measured hyperspectral camera entrance pupil is blocked is taken as the average noise brightness value of the image plane; S4, the n frames of images collected in the same field of view after preprocessing are subjected to time domain median filtering to eliminate random noise, and then the spatial dimension centroid coordinates of each wavelength point image are extracted in combination with the average noise brightness value of the image plane, color distortion value calculation is performed, and a full-waveband color distortion distribution curve is generated; The opening length direction of the linear target (2) is parallel to the ground, and the slit (9) of the hyperspectral camera is perpendicular to the ground; the collected image has a horizontal axis representing the spectral dimension and a vertical axis representing the spatial dimension; The spatial dimension centroid coordinates of each wavelength point image are extracted in combination with the average noise brightness value of the image plane, and the specific calculation is as follows: wherein y wavelenth is the centroid coordinate of the image formed after scanning the point image along the slit direction for a certain wavelength, y i is the average spatial dimension coordinate of the i-th field acquired n times, DN i is the average spatial dimension pixel brightness value of the i-th field acquired n times, DN noise is the average noise brightness value of the image surface; The color distortion value calculation is specifically as follows: where y smile is the color distortion between the wavelength λ1 and the wavelength λ2, and are the spatial dimensional centroid coordinates of the wavelength λ1 and the wavelength λ2, respectively.

2. The method of claim 1, wherein, The width d' of the opening of the linear target (2) is calculated as follows: In the formula, f1 is the focal length of the collimator (3), f2 is the focal length of the hyperspectral camera (4) to be measured, and d is the slit width of the hyperspectral camera slit (9).

3. The method of claim 1, wherein, The support leveling platform (6) is a three-point support leveling platform.

4. The method of claim 1, wherein, It also includes an air floating optical platform, the halogen lamp integrating sphere (1), the linear target (2), and the collimator (3) are all fixedly installed on the air floating optical platform; the air floating optical platform is suspended by air film to isolate and eliminate external vibration, and the relative position stability between the halogen lamp integrating sphere (1), the linear target (2), and the collimator (3) is maintained.

5. The method of claim 1, wherein, The preprocessing is specifically to perform band pass filtering and pixel response non-uniformity correction on the original image data.

6. A computer readable storage medium storing a computer program, characterized in that, The computer program, when executed by a processor, implements the color dispersion type hyperspectral camera color distortion measurement method according to any one of claims 1-5.

Images