A device and method for detecting chromatic aberration in a black lens based on hyperspectral imaging

Abstract

This invention discloses a hyperspectral-based device and method for detecting chromatic aberration in black lenses, relating to the field of chromatic aberration detection in black lenses. The device includes a support frame, a hyperspectral camera, a ring light source, a laser rangefinder, a lens fixture, a drive motor, and a main control board. The method involves system initialization and acquisition of a dark background; fixing the lens and a low-signal-value standard white board; driving the fixture to move and simultaneously acquiring multi-exposure spectral data and distance data; performing dark background subtraction and distance-light source compensation; synthesizing the multi-exposure data into high-quality fused spectral data using an intelligent fusion algorithm; calculating the relative reflectivity of the lens under test based on the fused data and the known reflectivity of the standard white board; and finally converting it into CIELAB chromaticity values ​​and outputting the results. This invention effectively overcomes the problems of weak signal, high noise interference, and poor measurement repeatability caused by the low reflectivity of black lenses through multi-exposure fusion and distance compensation, achieving automated detection of chromatic aberration parameters.

Description

Technical Field

[0001] This invention relates to the field of chromatic aberration detection of black lenses, and more specifically, to a device and method for chromatic aberration detection of black lenses based on hyperspectral imaging. Background Technology

[0002] Color difference in black lenses directly affects the uniformity of product appearance and brand reputation. It is a critical aspect of production batch control and after-sales paint matching. Excessive color difference can lead to visual inconsistencies within the same batch of lenses and even raise questions about product accuracy from users. Therefore, accurate measurement is crucial. Currently, the industry mainly uses hyperspectral testing technology, combined with portable colorimeters or spectrophotometers for color evaluation. Specifically, by collecting the spectral information of the sample and calculating ΔL, Δa, Δb, and the comprehensive color difference ΔE value according to standard color spaces such as CIE LAB, the color difference is quantified and the consistency of paint layer color is ensured.

[0003] However, in the existing technology, the above-mentioned existing technical solutions have obvious limitations and defects for the special materials of black lenses, mainly in the following aspects:

[0004] First, the inherent characteristics of black materials limit their ability to reflect visible light, which is often less than 3%. The effective light signal received by hyperspectral cameras is extremely weak and easily affected by equipment noise, resulting in large fluctuations in ΔL, Δa, and Δb data. In particular, minute changes in ΔL are difficult to capture accurately, yet they can still create visual color differences. Furthermore, the high light absorption allows light to penetrate the surface layer. If the inner layer color is uneven, the surface reflection signal cannot reflect the true color difference. At the same time, differences in surface gloss, texture, and flatness can lead to large deviations in multiple measurements at the same location.

[0005] Second, the ambient light interference is strong. Black materials are much more sensitive to laboratory LED lights and natural light than light-colored materials. A small amount of stray light superimposed on the test light source will change the reflected spectrum.

[0006] Third, existing equipment and methods have limitations. Portable colorimeters have poor data repeatability for low reflectivity materials, and spectrophotometers require special parameter settings. Improper parameters can reduce accuracy. Furthermore, small apertures may not provide sufficient signals, while large apertures are prone to defects.

[0007] The aforementioned factors make it difficult for existing technologies to meet the increasingly stringent quality control requirements of modern production and after-sales processes in measuring the color difference of black lens coatings. Therefore, developing a high-precision, high-stability color difference detection solution for black lenses that can effectively overcome these shortcomings has become an urgent technical problem to be solved in this field. Summary of the Invention

[0008] To address the problems in related technologies, this invention provides a hyperspectral-based device and method for detecting color difference in black lenses. This solves the problems of weak signals and large data deviations caused by the low reflectivity, high absorbance, and surface condition differences of the black lens coating, as well as the sensitivity to ambient light and susceptibility to interference, poor repeatability of portable colorimeters, and conflicting parameters of spectrophotometers, all of which affect the accuracy of black lens coating color difference detection.

[0009] Therefore, the specific technical solution adopted by the present invention is as follows:

[0010] A hyperspectral-based black lens chromatic aberration detection device, comprising:

[0011] Support frame;

[0012] A hyperspectral camera, mounted on the support frame, is used to acquire spectral data of the black lens under test and a low-signal-value standard white board;

[0013] A ring light source is fitted around the lens of the hyperspectral camera and fixed on the support frame. The ring light source is equipped with a light-diffusing plate.

[0014] A laser rangefinder, mounted on the support frame, is used to measure the distance between the black lens to be measured and the hyperspectral camera;

[0015] A lens fixture is used to fix the black lens to be tested and the low signal value standard white board;

[0016] A drive motor is used to drive the stage to move the lens fixture.

[0017] The main control board is electrically connected to the hyperspectral camera, the laser rangefinder, and the drive motor, and is used to control the coordinated operation of each component and process spectral data to calculate color difference.

[0018] This device utilizes built-in auxiliary fixtures and positioning devices to achieve repeated positioning, controlling the coaxiality of the lens optical axis and the detection axis within a reasonable range, thereby eliminating visual chromatic aberration introduced by posture deviation; it uses a level to monitor the platform tilt in real time, avoiding geometric chromatic aberration introduced by oblique beams; and it uses laser ranging closed-loop adjustment to control the working distance fluctuation within a reasonable range, preventing signal fluctuations caused by distance differences and ensuring data comparability.

[0019] Furthermore, the device also includes a built-in level for real-time monitoring of the device's horizontal status; the lens fixture has a multi-station structure, with each station neatly arranged in the horizontal and vertical directions.

[0020] Furthermore, the low signal value standard white board reflectivity characteristics are adapted to the reflectivity of a black lens.

[0021] Further, the device is arranged in a darkroom environment or equipped with a dark box structure.

[0022] On the other hand, a method for detecting the chromatic aberration of a black lens based on hyperspectral is provided. The method includes the following steps:

[0023] System initialization, collecting multiple frames of dark background data under different exposure times, and calculating the average dark background value;

[0024] Placing the black lens to be measured and a low-signal-value standard whiteboard on the lens fixture, adjusting the height using a laser rangefinder and locking it through the lens fixture;

[0025] Controlling the driving motor to drive the lens fixture to move along a preset path, while the hyperspectral camera respectively collects the original spectral data at three different preset exposure times A, B, and C, where C < B < A, and the laser rangefinder synchronously collects the distance data;

[0026] Performing dark background subtraction and distance-based light source compensation on the original spectral data to obtain the compensated data;

[0027] Fusing the compensated data to obtain the fused spectral data D(i, j, λ);

[0028] Based on the fused spectral data D(i, j, λ) and the known reflectance r(λ) of the low-signal-value standard whiteboard, calculating the relative reflectance r(i, j, λ) of the black lens to be measured;

[0029] Calculating the L*, a*, and b* values in the CIE LAB color space based on the relative reflectance r(i, j, λ), and outputting the detection result.

[0030] This method uses the sequential exposure superposition technology to enhance the weak reflection signal through multiple exposure signal integrations, while suppressing the equipment noise, to obtain a weak chromatic aberration image with a high signal-to-noise ratio; uses a ring halogen light source配合匀光片 in the effective aperture to achieve uniform light intensity distribution, avoiding chromatic aberration misjudgment caused by local bright spots; uses a darkroom structure to prevent spectral distortion caused by environmental light influence; uses a light source height compensation algorithm to compensate for the error caused by different light source intensities at different heights of the lens. A complete process from system initialization, multi-exposure data collection, data preprocessing, fusion to chromatic aberration calculation. This systematic method flow ensures that under the condition of extremely low reflectivity of the black lens, reflectivity < 3%, a set of coherent and automated operations can systematically solve the core problems such as weak signals, large noise interference, and environmental light sensitivity, avoiding the measurement deviation caused by step disconnection and parameter mismatch in the traditional method. Through a standardized step sequence, the consistency of each detection process is ensured, significantly improving the detection efficiency and data comparability, and is particularly suitable for batch detection and quality control in industrial production.

[0031] Furthermore, the step of fusing the compensated data to obtain fused spectral data D(i, j, λ) specifically includes:

[0032] Based on the compensated data of the preset exposure time B;

[0033] The system presets an upper limit M for overexposure threshold, a lower limit N for effective signal, and a camera overexposure threshold MD. For each spectral band of each pixel, it determines whether the exposure DN value exceeds the camera overexposure threshold MD. If it does, it replaces the corresponding data of exposure time C.

[0034] Determine whether all the exposure DN values ​​are lower than the effective signal lower limit N. If so, replace them with the corresponding data of exposure time A.

[0035] Exposure time normalization is performed on the spectral curve of each pixel in the replaced fused image.

[0036] Furthermore, the multi-exposure data fusion step further includes: for pixels whose DN value is neither overexposed nor completely underexposed, determining the proportion of the number of bands whose DN value is within the effective signal range [N, M] to the total number of bands Xw of the pixel; if the proportion is lower than a preset threshold, then the corresponding data of exposure time A or exposure time C is used for replacement.

[0037] Furthermore, the distance-based light source compensation uses the following formula:

[0038] DD(i,j,λ) = (L / M(i,j)) 2 * DN(i, j, λ)

[0039] Where DD(i,j,λ) is the compensated data, L is the standard distance, M(i,j) is the actual distance measured by the laser rangefinder, and DN(i,j,λ) is the original spectral data after subtracting the dark background.

[0040] Furthermore, the reflectivity calculation step uses the following formula:

[0041] r(i,j,λ) = D(i,j,λ) / DR(i n j n ,λ) * r(λ)

[0042] Where r(i, j, λ) is the relative reflectance of the lens under test, D(i, j, λ) is the fused spectral data, and DR(i n j n λ) represents the low-signal-value standard whiteboard region data extracted from the fused spectral data, and r(λ) represents the known reflectance of the low-signal-value standard whiteboard.

[0043] Furthermore, the motion mode of the drive motor is a preset curved path, and the frame rate f and field of view height h of the hyperspectral camera are matched with the motion speed v of the drive motor.

[0044] The beneficial effects of this invention are as follows:

[0045] 1. In practical use, this invention uses a built-in laser rangefinder to measure in real time and ensure consistent detection distance each time. Combined with a light source distance compensation algorithm, it accurately compensates for the error caused by the change in light source intensity due to the difference in lens height. The multi-station precise design and positioning limit function of the lens fixture ensure the consistency of the spatial position and angle between the lens under test and the low signal value standard white board. The built-in level monitors the level of the platform in real time, effectively avoiding geometric color difference caused by the tilt of the device. These measures together control factors such as object distance fluctuation and posture deviation within a reasonable range, greatly improving the accuracy of measurement and the comparability and repeatability of data.

[0046] 2. The built-in laser rangefinder measures in real time and ensures consistent distance for each test. Combined with a light source distance compensation algorithm, it accurately compensates for errors in light source intensity caused by differences in lens height. The multi-station precision centering design and positioning limit function of the lens fixture ensure the consistency of spatial position and angle between the lens under test and the low signal value standard white board. The built-in level monitors the platform's level status in real time, effectively avoiding geometric color difference caused by device tilt. These measures together control factors such as object distance fluctuation and posture deviation within a reasonable range, greatly improving the accuracy of measurement and the comparability and repeatability of data.

[0047] 3. This device supports automated continuous detection via drive motor. A single acquisition can cover multiple areas or multiple lenses, ensuring data consistency and efficiency in batch detection. The main control board integrates algorithms, supports custom selection of detection areas, automatically calculates Lab parameters, and is compatible with multiple ΔE algorithms. It can flexibly adapt to different industry standards and customer needs, reducing the operating threshold. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. 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.

[0049] Figure 1 This is a schematic diagram of a hyperspectral black lens chromatic aberration detection device according to an embodiment of the present invention;

[0050] Figure 2This is a schematic flowchart of a hyperspectral-based method for detecting chromatic aberration in a black lens according to an embodiment of the present invention.

[0051] In the picture:

[0052] 1. Support frame; 2. Hyperspectral camera; 3. Black lens; 4. Low signal value standard white board; 5. Ring light source; 6. Laser rangefinder; 7. Lens fixture; 8. Drive motor; 9. Main control board; 10. Built-in level. Detailed Implementation

[0053] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Example 1

[0055] like Figure 1 As shown, the present invention discloses a hyperspectral-based black lens color difference detection device, mainly comprising a support frame 1, a hyperspectral camera 2, a ring light source 5, a laser rangefinder 6, a lens fixture 7, a drive motor 8, a main control board 9, and a built-in level 10. The support frame 1 serves as the core support structure, supporting components such as the hyperspectral camera 2, the ring light source 5, and the laser rangefinder 6. The camera mounting bracket can move freely up and down to adapt to the detection needs of black lenses 3 at different heights, ensuring the stability of the spatial layout of each component. The hyperspectral camera 2 is preferably a line-scanning hyperspectral camera 2, fixed on the support frame 1. Its field of view is adapted to the size of the black lens 3 under test and the low-signal-value standard white board 4. The line-scanning mode can continuously capture the linear spectral distribution and spatial color information of the paint on the black lens 3 and the low-signal-value standard white board 4, providing high-resolution data support for low-reflectivity black materials. A ring light source 5, preferably a ring halogen light source, is mounted around the lens of the hyperspectral camera 2. A quartz light-diffusing plate is added to its outer side to ensure that the light can uniformly cover the surfaces of the black lens 3 under test and the low-signal-value standard white board 4, eliminating local light intensity differences. A laser rangefinder 6 is vertically arranged directly above the black lens 3 under test. It is used to automatically measure the distance between the black lens 3 and the hyperspectral camera 2 before each test, ensuring consistent detection height, avoiding spectral acquisition errors caused by distance deviation, and improving data comparability.

[0056] The lens fixture 7 has a multi-station structure, with the centers of all stations aligned both horizontally and vertically. It has both positioning and limiting functions and is used to precisely fix the待测 black lens 3 and the low-signal-value standard whiteboard 4. The low-signal-value standard whiteboard 4 has a reflectivity characteristic specifically adapted to the low-reflection characteristic of the black lens 3 and serves as a color reference benchmark. The drive motor 8 cooperates with the movable stage built into the darkroom to drive the stage to move the lens fixture 7 along a preset S-shaped path, enabling automated continuous detection of multiple regions or multiple lenses. The built-in level 10 is used to monitor the horizontal state of the device in real time. The main control board 9 integrates a control circuit, an algorithm unit, and a data processing module, and is electrically connected to the hyperspectral camera 2, the laser rangefinder 6, the drive motor 8, etc., coordinates the collaborative work of each component, and is responsible for spectral acquisition, data preprocessing, reflectivity calculation, Lab parameter calculation, and result output. The entire device is preferably integrated into a darkroom environment or equipped with a dark box structure such as a black cloth cover to isolate external stray light interference to the greatest extent.

[0057] Through the above technical solutions, through innovative device design and system integration, high-precision, high-repeatability, and high-efficiency detection of the outer paint color difference of the black lens 3 is achieved, effectively solving many technical bottlenecks in the color difference measurement of black low-reflectivity materials in the prior art, and having significant industrial application value.

[0058] Embodiment 2

[0059] According to the present invention, a method for detecting the color difference of a black lens based on hyperspectral is also provided, as Figure 2 shown, including the following steps:

[0060] S1. System initialization:

[0061] Fix the annular light source 5 outside the lower part of the lens of the hyperspectral camera 2, ensure that the light source does not directly irradiate the lens, and install a light homogenizing sheet. Place the system in a dark box. Place the lens fixture 7 on the test bench and fix it. Install and calibrate the level 10 and the laser rangefinder 6. Start each device to complete self-check.

[0062] S2. Dark background acquisition:

[0063] Set three exposure times A, B, and C, where C < B < A. For example, A = 50 ms, B = 30 ms, and C = 10 ms. Cover the camera lens with a lens cap with a light shielding rate ≥ 99.9%, and collect multiple frames of dark background data at each exposure time. Calculate the average dark background values avg_A(λ), avg_B(λ), and avg_C(λ) corresponding to each exposure time.

[0064] The calculation method of the average dark background data is as follows: where avg_A, avg_B, and avg_C are the average dark backgrounds of high, medium, and low exposures respectively, and are used for subsequent dark noise deduction.

[0065] ;

[0066] ;

[0067]

[0068] S3. Sample placement and height adjustment:

[0069] Place the black lens group to be tested according to the limiting position of lens fixture 7, and place the low signal value standard white board 4 at the designated position of lens fixture 7. Use laser rangefinder 6 to measure the height of the reference board and the top surface of each black lens 3, adjust them to be roughly the same, and then lock them in place through lens fixture 7. Test run drive motor 8, and use level 10 to confirm that the plane of lens fixture 7 remains horizontal during the movement. After confirmation, reset the motor.

[0070] S4, Multi-exposure data acquisition:

[0071] Set the motion parameters of drive motor 8, and set the X and Y axis acquisition parameters. For the X-axis direction, ensure that the frame rate f, field of view height h, and motion speed v of hyperspectral camera 2 are compatible. The specific calculation formula is: v = f * binV * perOne, where binV is the vertical pixel merging number, perOne is the size of a single pixel, f represents the camera's frame rate, and v represents the actual clicking speed. This ensures that image acquisition is distortion-free and gap-free, accurately reproducing the shape of the actual object being measured. For the Y-axis direction, the single vertical movement distance Sy is calculated using the field of view height h and field of view angle α. The width of a single scan is calculated as Sy = 2*h*tan(α / 2), ensuring that the entire image frame is acquired without gaps between stripes.

[0072] The drive motor 8 is controlled to move along a preset S-shaped path. The hyperspectral camera 2 sequentially acquires raw spectral data DN_low_raw(i,j,λ), DN_mid_raw(i,j,λ), and DN_high_raw(i,j,λ) at low exposure time C, medium exposure time B, and high exposure time A. Simultaneously, the laser rangefinder 6 synchronously acquires the distance array M(i,j). The motor resets after each acquisition.

[0073] S5. Data Matching and Preprocessing:

[0074] By pre-calibrating, the distance array M(i,j) is associated with the raw data of each exposure time by pixel column, ensuring that the signal of each (i,j) pixel corresponds one-to-one with the distance.

[0075] Perform dark background subtraction:

[0076] DN_low(i, j, λ) = DN_low_raw(i, j, λ) - avg_C(λ)

[0077] DN_mid(i, j, λ) = DN_mid_raw(i, j, λ) - avg_B(λ)

[0078] DN_high(i, j, λ) = DN_high_raw(i, j, λ) - avg_A(λ)

[0079] Perform distance-based light source compensation (L is the standard distance):

[0080] DD_low(i, j, λ) = (L / M(i,j))^2 * DN_low(i, j, λ)

[0081] DD_mid(i, j, λ) = (L / M(i,j))^2 * DN_mid(i, j, λ)

[0082] DD_high(i, j, λ) = (L / M(i,j))^2 * DN_high(i, j, λ)

[0083] S6. Multi-exposure data fusion:

[0084] Take the data DD_mid(i, j, λ) after medium exposure compensation as the benchmark. Preset the upper limit M of the overexposure threshold, the lower limit N of the effective signal, and the camera overexposure critical value MD.

[0085] Judge each spectral band of each pixel (i, j):

[0086] If the DN value of any band in DD_mid > MD, then the corresponding pixel data of DD_low is used to replace this pixel.

[0087] If the DN values of all bands in DD_mid < N, then the corresponding pixel data of DD_high is used to replace this pixel.

[0088] If the DD_mid data is neither overexposed nor underexposed in all bands, then judge the proportion of the number of bands with DN values in the interval [N, M] to the total number of bands Xw. If the proportion is less than 50%, then use the non-overexposed DD_high or DD_low data to replace it.

[0089] Divide the spectral curve of each pixel in the fused image by its corresponding exposure time for normalization, and finally output the fused spectral data D(i, j, λ).

[0090] S7. Reflectance calculation:

[0091] Extract the low signal value standard whiteboard 4 region data DR(i) from the fused data D(i,j,λ). n j n ,λ).

[0092] Based on the known reflectivity r(λ) of the low-signal-value standard white board 4, the relative reflectivity of the black lens 3 to be tested is calculated as: r(i,j,λ) = D(i,j,λ) / DR(i n ,j n ,λ)* r(λ)

[0093] S8. Color Calculation and Output:

[0094] 1. Calculate the CIE 1931 tristimulus values ​​X, Y, Z based on the relative reflectance r(i, j, λ) (using the D65 standard illuminator and standard observer function).

[0095]

[0096] in, It is the relative spectral power distribution of a standard illuminator. , , λ is the spectral tristimulus value of the CIE standard observer, Δλ is the wavelength interval, and k is the normalization constant, which is usually chosen to make Y=100 for a perfectly reflective diffuser.

[0097] 2. Convert the X, Y, Z values ​​to L*, a*, b* values ​​in the CIELAB color space.

[0098]

[0099] Where Xn, Yn, and Zn are the tristimulus values ​​of a perfectly reflective diffuser under a standard illuminator, and the function... The definition of is:

[0100]

[0101] in, .

[0102] The output includes a test report containing reflectance curves, Lab values, and raw data, and can further calculate color difference parameters such as ΔE.

[0103] In this embodiment, a complete technical solution is formed through systematic process design, innovative multi-exposure fusion algorithm, precise physical quantity compensation, targeted calibration calculation, and coordinated motion control. This solution effectively overcomes the inherent difficulties in chromatic aberration detection of black lenses, ultimately achieving high-precision, high-stability, and high-efficiency measurement of ΔL, Δa, and Δb values, especially the ΔL value which is sensitive to the human eye, meeting the stringent color quality requirements of industrial production and after-sales service.

[0104] In summary, with the help of the above-mentioned technical solutions of the present invention, in practical use, this method supports custom selection of detection areas, automatically calculates Lab parameters and is compatible with multiple ΔE algorithms, adapting to different industry standards; the camera's autofocus lowers the operational threshold, while supporting the acquisition of multiple lines of data at once, ensuring the consistency of batch detection data and improving efficiency.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A hyperspectral-based black lens chromatic aberration detection device, characterized in that, The device is set in a darkroom environment or equipped with a dark box structure, and includes: A support frame (1); A hyperspectral camera (2), arranged on the support frame (1), for collecting spectral data of a to-be-tested black lens (3) and a low-signal-value standard white board (4); An annular light source (5), sleeved around the lens periphery of the hyperspectral camera (2) and fixed on the support frame (1), and the annular light source (5) is provided with a light homogenizing plate; A laser rangefinder (6), arranged on the support frame (1), for measuring the distance between the to-be-tested black lens (3) and the hyperspectral camera (2); A lens fixture (7), for fixing the to-be-tested black lens (3) and the low-signal-value standard white board (4); A driving motor (8), for driving a stage to drive the lens fixture (7) to move; A main control board (9), electrically connected to the hyperspectral camera (2), the laser rangefinder (6) and the driving motor (8), for controlling each component to work in cooperation and processing spectral data to calculate color difference.

2. The hyperspectral-based chromatic aberration detection device for black lenses according to claim 1, characterized in that, The device further includes a built-in level (10), for monitoring the horizontal state of the device in real time; the lens fixture (7) is a multi-station structure, and each station is neatly arranged in the horizontal and vertical directions.

3. The hyperspectral-based black lens chromatic aberration detection device according to claim 1, characterized in that, The reflectivity characteristics of the low-signal-value standard white board (4) are adapted to the reflectivity of the black lens.

4. A method for detecting chromatic aberration in a black lens based on hyperspectral imaging, employing the apparatus as described in any one of claims 1-3, characterized in that, The method includes the following steps: System initialization, collecting multiple frames of dark background data at different exposure times, and calculating an average dark background value; Placing the to-be-tested black lens (3) and the low-signal-value standard white board (4) on the lens fixture (7), adjusting the height by using the laser rangefinder (6) and locking through the lens fixture (7); Controlling the driving motor (8) to drive the lens fixture (7) to move along a preset path, and simultaneously the hyperspectral camera (2) respectively collects original spectral data at three different preset exposure times A, B, C, wherein C < B < A, and the laser rangefinder (6) synchronously collects distance data; Performing dark background subtraction and distance-based light source compensation on the original spectral data to obtain compensated data; Fusing the compensated data to obtain fused spectral data D(i, j, λ); Based on the fused spectral data D(i, j, λ) and the known reflectivity r(λ) of the low-signal-value standard white board, calculating the relative reflectivity r(i, j, λ) of the to-be-tested black lens; Calculating the L*, a*, b* values in the CIE LAB color space based on the relative reflectivity r(i, j, λ), and outputting a detection result; Dark background collection includes: setting three exposure times A, B, C, wherein C < B < A, covering the camera lens with a lens cap, respectively collecting multiple frames of dark background data at the three exposure times, and calculating the average dark background values avg_A(λ), avg_B(λ), avg_C(λ) corresponding to each exposure time; The hyperspectral camera (2) sequentially collects original spectral data DN_low_raw(i, j, λ), DN_mid_raw(i, j, λ), DN_high_raw(i, j, λ) at a low exposure time C, a medium exposure time B, and a high exposure time A; Dark background removal includes: ； The distance-based light source compensation uses the following formula: ； Where DD(i,j,λ) is the compensated data, L is the standard distance, M(i,j) is the actual distance measured by the laser rangefinder (6), and DN(i,j,λ) is the original spectral data after subtracting the dark background; The specific steps of fusing the compensated data to obtain the fused spectral data D(i, j, λ) include: Based on the compensated data of the preset exposure time B; The system presets an upper limit M for overexposure threshold, a lower limit N for effective signal, and a camera overexposure threshold MD. For each spectral band of each pixel, it determines whether the exposure DN value exceeds the camera overexposure threshold MD. If it does, it replaces the corresponding data of exposure time C. Determine whether all the exposure DN values ​​are lower than the effective signal lower limit N. If so, replace them with the corresponding data of exposure time A. Exposure time normalization is performed on the spectral curve of each pixel in the replaced fused image.

5. The method for detecting chromatic aberration in a black lens based on hyperspectral imaging according to claim 4, characterized in that, The multi-exposure data fusion step further includes: for pixels whose DN values ​​are neither overexposed nor completely underexposed, determining the proportion of the number of bands whose DN values ​​are within the effective signal range [N, M] to the total number of bands Xw of the pixel. If the proportion is lower than a preset threshold, the corresponding data of exposure time A or exposure time C is used for replacement.

6. The method for detecting chromatic aberration in a black lens based on hyperspectral imaging according to claim 4, characterized in that, The reflectance calculation step uses the following formula: ； Where r(i, j, λ) is the relative reflectance of the lens under test, D(i, j, λ) is the fused spectral data, DR(in, jn, λ) is the low signal value standard white board region data extracted from the fused spectral data, and r(λ) is the known reflectance of the low signal value standard white board.

7. The method for detecting chromatic aberration in a black lens based on hyperspectral imaging according to claim 4, characterized in that, The motion mode of the drive motor (8) is a preset curved path, and the frame rate f and field height h of the hyperspectral camera (2) are matched with the motion speed v of the drive motor (8).

Images