A real-time color measurement system and method with spectral level calibration

Abstract

The present application relates to a kind of spectral level calibration real-time color measurement system and method, it is related to color measurement technical field, system includes viewfinder lens group, shutter, light beam spectrometer module, multi-channel light filter detection unit, dispersion module, main control board and angle biasing device;Method includes calibration stage, acquisition stage and measurement stage.By obtaining monochromatic light image sequence and light intensity data, calculate sensor response curve, filter transmittance curve and dispersion module response curve;Using reference board to calculate camera correction coefficient;To be measured object acquisition image and apply correction coefficient, combined with dynamic calibration compensation light source fluctuation, finally output high-precision color value.The present application solves the problem of prior art by light source interference, bench difference, high cost and low efficiency, realizes the real-time color measurement of low cost, high efficiency, high precision and high resolution, it is suitable for the online detection of textile, electronic display and other industries.

Description

A real-time color measurement system and method with spectral-level calibration Technical Field

[0001] This invention relates to the field of color measurement technology, and more specifically, to a real-time color measurement system and method with spectral-level calibration. Background Technology

[0002] In industries such as textile printing and dyeing, electronic display, and plastics processing, accurate color measurement is a core element in ensuring product quality. As manufacturing precision requirements continue to rise, color measurement technology not only needs to be cost-effective and widely applicable, but also needs to meet the real-time monitoring needs of modern production lines in terms of measurement efficiency, accuracy, and spatial resolution.

[0003] Currently, the colorimetric solutions widely used in the industry mainly include three types: ordinary RGB cameras, area array hyperspectral analyzers, and single-point spectrometers. However, all of these solutions have significant technical limitations:

[0004] While ordinary RGB cameras are relatively inexpensive, they lack spectral-level calibration and are susceptible to fluctuations in the light source spectrum and differences between equipment, leading to color interpretation distortion. These devices are only suitable for scenarios with low precision requirements and cannot meet the stringent color consistency requirements of high-end manufacturing fields such as medical consumables and precision electronics. Area array hyperspectral analyzers offer high-precision measurement capabilities across the entire wavelength range, but their high cost (typically hundreds of thousands of yuan per unit) and the massive amounts of data they generate, coupled with complex processing algorithms, make it difficult to achieve the second-level area detection response speed required on production lines. This limits their large-scale deployment in high-speed applications such as online fabric defect detection and display panel color difference analysis. Single-point spectrometers offer some cost advantages, but their point-by-point approach...

[0005] The mechanical measurement method of scanning results in extremely low efficiency in area detection. It usually takes several minutes to complete the color analysis of a region. In addition, it cannot obtain spatial color distribution information, so it is not applicable in scenarios that require high spatial resolution, such as pixel-level color difference detection of 4K / 8K display panels.

[0006] In summary, existing color measurement technologies suffer from irreconcilable contradictions regarding cost, efficiency, accuracy, and spatial resolution, lacking a comprehensive solution that can balance all four aspects and is suitable for high-speed, high-resolution industrial scenarios. Therefore, the industry urgently needs a new color measurement system and method with spectral-level calibration capabilities, real-time processing performance, high spatial resolution, and a reasonable cost structure to overcome existing technological bottlenecks and meet the comprehensive requirements of high-end manufacturing for color quality control. Summary of the Invention

[0007] In view of the problems in related technologies, this invention proposes a real-time color measurement system and method with spectral-level calibration to overcome the aforementioned technical problems existing in the existing related technologies.

[0008] Therefore, the specific technical solution adopted by the present invention is as follows:

[0009] A real-time color measurement system with spectral-level calibration, comprising:

[0010] The viewfinder lens group acquires the scene light signal of the target under test;

[0011] The shutter is located between the viewfinder lens group and the beam splitter module, and is connected to the main control board to control the on / off state of the light path and the exposure time.

[0012] A beam splitting module is located between the viewfinder lens group and the multi-channel filter detection unit to split the incident beam into multiple paths.

[0013] The multi-channel filter detection unit includes multiple filters and corresponding mono sensors to collect light signals from multiple color channels;

[0014] The dispersion module includes a beam collection module, a slit, a dispersion grating, and a linear array mono sensor 7 to acquire the spectral signal of the light source;

[0015] The main control board controls the coordinated operation of the shutter, multi-channel filter detection unit, and dispersion module, and processes the acquired optical signal data;

[0016] Angle offset device adjusts the angle of the dispersion module to adapt to light sources from different directions.

[0017] Through the collaborative operation of the multi-channel filter detection unit and the dispersion module, the optical signals of multiple color channels and the spectral signals of the light source can be acquired. By using a planar array mono sensor combined with a beam splitter for multi-channel parallel acquisition, multi-band color information of the area under test can be acquired instantly. The integrated angle offset device of the system enables the dispersion module to flexibly adapt to light sources in different directions, enhancing the deployment flexibility of the system. It can effectively compensate for the spectral fluctuations of the light source over time and the differences between different devices, ensuring the measurement consistency and stability of the system during long-term operation in complex industrial environments.

[0018] Furthermore, the multi-channel filter detection unit includes three sets of independent filters and three sets of independent mono sensors, which correspond to the red, green and blue color bands respectively.

[0019] Furthermore, the main control board integrates a high-speed processing unit to process multi-channel spectral data in real time and execute color space conversion algorithms.

[0020] Furthermore, the system is suitable for online color detection in the textile, electronic display, and plastics processing industries, achieving high-precision, high-resolution real-time color measurement through spectral-level calibration and dynamic compensation mechanisms.

[0021] On the other hand, a real-time color measurement method with spectral-level calibration includes the following steps:

[0022] Calibration phase: Obtain monochromatic light image sequences and corresponding light intensity data of the system in the states of no filter and with filter, and calculate the response curve of the mono sensor, the transmittance curve of the filter and the response curve of the dispersion module.

[0023] Acquisition phase: Use a reference board to acquire image data under the current light source and calculate the camera correction coefficient;

[0024] Measurement stage: Image acquisition of the object to be measured, and calculation of its color value based on the correction coefficient.

[0025] The calibration phase systematically acquires and calculates sensor response curves, filter transmittance curves, and dispersion module response curves, establishing a precise and reliable internal optical characteristic database for the entire measurement system. During the acquisition phase, a reference plate with known reflectivity is used, combined with real-time light source spectral information, to calculate camera correction coefficients for the specific measurement environment. This ensures that measurement results are not limited by specific environments or equipment conditions, significantly improving measurement accuracy and scene adaptability. In the measurement phase, the system directly utilizes the previously established correction coefficients to rapidly process the image of the object under test, calculating the spectrally corrected true color value.

[0026] Furthermore, the calibration phase includes:

[0027] Without installing the filter, acquire the image sequence P(λ) of monochromatic light from 380nm to 780nm using a monochromatic sensor;

[0028] Simultaneously acquire the light intensity sequence E(λ) measured by the illuminometer and the light intensity sequence L(λ) measured by the dispersion module;

[0029] After installing the filter, the image sequence Q(λ) is obtained again;

[0030] Calculate the response curve A(λ) = P(λ) / E(λ), the transmittance curve T(λ) = Q(λ) / P(λ), and the dispersive module response curve R(λ) = L(λ) / E(λ).

[0031] Furthermore, the acquisition phase includes:

[0032] Using a reference plate with known reflectivity r(λ)(x,y), three-wavelength images I1, I2, and I3 are captured under a uniform light source;

[0033] Obtain the spectral signal L(λ) collected by the dispersion module, and calculate the current light source intensity E(λ) = L(λ) / R(λ);

[0034] The camera correction coefficients K1, K2, and K3 are calculated based on the combined calculations of I1, I2, I3 with E(λ), r(λ), A(λ), T(λ) and the human visual function Fx(λ), Fy(λ), Fz(λ), respectively.

[0035] Furthermore, the measurement phase includes:

[0036] Three-wavelength images I4, I5, and I6 were captured on the object under test;

[0037] For each pixel (x, y) in the image, calculate the corrected R, G, and B values ​​using correction coefficients K1, K2, and K3;

[0038] Convert the corrected R, G, B values ​​to CIELAB color space values.

[0039] Furthermore, real-time color measurement methods also include a dynamic calibration step:

[0040] Periodically collect the spectral signal L(λ,t) of the light source;

[0041] Update the dynamic correction coefficients K1(t), K2(t), and K3(t) based on L(λ,t);

[0042] Dynamic correction coefficients are used to compensate for light source fluctuations in real-time acquired image data.

[0043] By capturing the light source spectrum L(λ,t) in real time, the system can perceive changes in spectral energy distribution that are imperceptible to the human eye and traditional cameras, thus eliminating color interpretation errors caused by them in principle. The introduced dynamic calibration step, by periodically acquiring the light source spectral signal and updating the dynamic correction coefficient accordingly, endows the system with continuous adaptive anti-interference capability and long-term stability.

[0044] Furthermore, all image acquisition steps include dark noise correction, which involves acquiring a dark image before shooting as a noise baseline and subtracting it from subsequent calculations.

[0045] The beneficial effects of this invention are as follows:

[0046] 1. In practical use, this invention acquires the spectral signal of the light source in real time through the dispersion module, and performs dynamic spectral-level correction on the RGB three-channel data acquired by the multi-channel filter detection unit based on this. This fundamentally eliminates color interpretation distortion caused by factors such as fluctuations in the color temperature of the light source and changes in ambient light, significantly improving measurement accuracy. At the same time, by using a monochromatic area sensor combined with a beam splitting and filtering scheme, parallel light signal acquisition of a large 4K / 8K area is achieved. Combined with the high-speed operation of the main control board, real-time color measurement of the area is achieved, taking into account both spatial resolution and time response speed.

[0047] 2. In practical use, compared with traditional colorimetric methods, this invention introduces a reference plate calibration and dynamic time calibration mechanism. The reference plate provides the equipment's reference spectrum, eliminating inter-stage differences between different devices. Through algorithm optimization and the combination of dispersion modules and mono sensors, it replaces expensive hyperspectral imaging components, enabling the system to meet the precision requirements of high-end manufacturing while possessing significant cost advantages for large-scale deployment and application.

[0048] 3. This system utilizes the characteristics of an area array detector to acquire the spectral correction color information of each pixel on the surface of the object under test, achieving pixel-level high spatial resolution color analysis. Furthermore, the system can periodically monitor and compensate for the attenuation and fluctuation of the light source over time. By updating the dynamic correction coefficient, it ensures the long-term stability and reliability of the measurement results under long-term continuous operation, giving it advantages for large-scale application. Attached Figure Description

[0049] 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.

[0050] Figure 1 is a schematic diagram of a real-time color measurement device according to an embodiment of the present invention;

[0051] Figure 2 is a calibration process for a real-time color measurement device according to an embodiment of the present invention;

[0052] Figure 3 is a flowchart of a real-time color measurement device according to an embodiment of the present invention.

[0053] In the picture:

[0054] 1. Viewfinder lens group; 2. Shutter; 3. Beam splitter module; 4. Beam collector module; 5. Slit; 6. Dispersion grating; 7. Linear array mono sensor; 8. Main control board; 9. Angle offset device; 10. First filter; 11. Second filter; 12. Third filter; 13. First mono sensor; 14. Second mono sensor; 15. Third mono sensor. Detailed Implementation

[0055] 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.

[0056] Example 1

[0057] As shown in Figure 1, the present invention provides a real-time color measurement system with spectral-level calibration. The system includes a viewfinder lens group 1, a shutter 2, a beam splitting module 3, a multi-channel filter detection unit, a dispersion module, a main control board 8, and an angle offset device 9.

[0058] The viewfinder lens group 1 is located at the front of the system and is responsible for acquiring the scene light signal of the target under test and converging it into a collimated beam for subsequent optical paths. The optical parameters of the viewfinder lens group 1, such as focal length and field of view, can be configured according to the actual detection distance and range.

[0059] Shutter 2 is installed in the optical path between viewfinder lens group 1 and beam splitter module 3, and is connected to main control board 8 via cable. Main control board 8 controls the opening and closing of shutter 2 and the exposure time, achieving precise exposure control, and supports the acquisition of dark field noise data when the light source is turned off, providing a reference for subsequent signal correction and ensuring the accuracy and stability of color detection from the source.

[0060] The beam splitter module 3 is positioned in the optical path between the viewfinder lens group 1 and the multi-channel filter detection unit. Preferably, it employs a semi-transparent mirror structure to split the single incident beam from the viewfinder lens group 1 into three paths. The three output optical axes of the split beams are precisely aligned with the three sets of optical path input axes in the multi-channel filter detection unit and are collinear with the optical axis of the viewfinder lens group 1, ensuring the accuracy of optical path transmission.

[0061] The multi-channel filter detection unit comprises three independent "filter-mono sensor" combinations. In this embodiment, the first filter 10, the second filter 11, and the third filter 12 are designed as narrowband filters targeting the three key color bands of red (R), green (G), and blue (B), respectively. A mono sensor is installed behind each filter, namely the first mono sensor 13, the second mono sensor 14, and the third mono sensor 15. The optical axis of the optical center of the filter is strictly aligned with the optical axis of the corresponding optical path of the mono sensor.

[0062] Overlap. The mono sensor is a black and white photosensitive chip, which has higher light intake and sensitivity compared to color sensors. It can efficiently convert the filtered monochromatic light signal into an electrical signal and upload it to the main control board 8 via cable, thereby providing three independent raw spectral data of red, green and blue, realizing the parallel acquisition of multi-dimensional color information, and laying the data foundation for subsequent color space conversion.

[0063] The dispersion module includes a beam collection module 4, a slit 5, a dispersion grating 6, and a linear mono sensor 7. The beam collection module 4, located on the system's beam input side, is responsible for collecting light signals from a specific area or angle and guiding them to the slit 5. This provides a concentrated and stable incident beam for the dispersion grating 6's beam splitting process, ensuring the signal strength and stability of the spectral detection. It also provides a real-time reference for the entire system to cope with spectral fluctuations of the light source over time, assisting the main control board 8 in dynamically correcting color interference from the light source.

[0064] The slit 5 is installed after the beam collection module 4 to spatially restrict the incident beam, so that the beam enters the dispersive grating 6 in the form of a narrow slit, thereby improving the spectral resolution and ensuring that the subsequent linear array mono sensor 7 accurately detects the subdivided spectral information.

[0065] The dispersion grating 6 is placed in the optical path before the linear array mono sensor 7. Utilizing the dispersion characteristics of the dispersion grating 6, the incident composite light is decomposed into monochromatic light according to wavelength, providing a basis for the linear array mono sensor 7 to perform high spectral resolution detection, thereby supporting high-precision and high-repeatability color detection and analysis.

[0066] The linear array mono sensor 7 receives the spectral signal after it has been dispersed by the dispersion grating 6. Based on the linear array photosensitive structure, it converts light signals of different wavelengths into electrical signals in sequence and uploads them to the main control board 8 via cable. This provides the system with high-resolution spectral distribution data to assist in the fine correction and analysis of color detection.

[0067] Angle offset device 9 is positioned between the dispersion module and the overall system frame, fixing the dispersion module to the overall system frame and allowing the dispersion module to rotate to adapt to different light source directions, thus enhancing the system's deployment adaptability.

[0068] The main control board 8 serves as the core of the system's control and processing. It preferably integrates high-speed processing units such as FPGA or DSP. It connects to and controls modules such as shutter 2, multi-channel filter detection unit, and linear array mono sensor 7 via cables. This enables synchronous control of shutter 2 exposure, data acquisition from various sensors, analog-to-digital conversion, dark noise subtraction, and subsequent color algorithm calculations, outputting real-time color detection results.

[0069] Example 2

[0070] This invention provides a real-time color measurement method with spectral-level calibration. The method mainly includes three stages: system calibration, acquisition and correction, and real-time measurement. It also includes a dynamic calibration step. Before acquiring any image, a dark field image must be acquired to perform dark noise correction. Subsequent calculations are all performed on the basis of deducting dark noise.

[0071] 1. System Calibration Phase

[0072] As shown in Figure 2, the system calibration phase aims to obtain the system's own characteristic parameters.

[0073] First, without installing the filter, perform the following steps:

[0074] The camera is controlled to sequentially capture images of monochromatic light in the wavelength range of 380nm-780nm, resulting in three sets of images from the mono sensor.

[0075] Unfiltered image sequences P1(λ), ​​P2(λ), P3(λ).

[0076] Simultaneously, the intensity of monochromatic light in the range of 380nm to 780nm was measured sequentially using a photometer to obtain the light intensity sequence E(λ).

[0077] Simultaneously, the camera's dispersion module is controlled to capture monochromatic light in the range of 380nm to 780nm, and the light intensity sequence L(λ) measured by the dispersion module is obtained.

[0078] Next, install the red, green, and blue filters into their respective optical paths, and then perform the following steps again:

[0079] The camera is controlled to sequentially capture monochromatic light images in the wavelength range of 380nm to 780nm, resulting in three image sequences Q1(λ), ​​Q2(λ), and Q3(λ) after the filter is installed.

[0080] Finally, based on the calibration data, the response curves of each mono sensor under the current system architecture are calculated: A1(λ) = P1(λ) / E(λ).

[0081] A2(λ) = P2(λ) / E(λ),

[0082] A3(λ) = P3(λ) / E(λ).

[0083] Where P(λ) is the camera image data at wavelength λ without a filter, E(λ) is the incident light intensity at wavelength λ, and A(λ) reflects the response capability of the mono sensor to light at wavelength λ under the current system structure. Because there is also inter-sensor difference, the response curves A1(λ), ​​A2(λ), and A3(λ) of the first mono sensor 13, the second mono sensor 14, and the third mono sensor 15 can be obtained.

[0084] Calculate the transmittance curves for each filter based on the calibration data:

[0085] T1(λ) = Q1(λ) / P1(λ),

[0086] T2(λ) = Q2(λ) / P2(λ),

[0087] T3(λ) = Q3(λ) / P3(λ).

[0088] Where P(λ) is the camera image data at wavelength λ without the filter, Q(λ) is the camera image data at wavelength λ after the filter is installed, and T(λ) reflects the proportion of light transmitted by the filter at wavelength λ. The three filters represent the RGB bands respectively, and the transmittance curves T1(λ), ​​T2(λ), and T3(λ) of the three filters can be calculated based on the corresponding P(λ) and Q(λ).

[0089] The response curve of the dispersion module is calculated based on the calibration data: R(λ) = L(λ) / E(λ). Where L(λ) is the light intensity at wavelength λ measured by the dispersion module, E(λ) is the incident light intensity at wavelength λ, and R(λ) characterizes the response characteristics of the current system's dispersion module to light at wavelength λ. After completing the above calculation, the calibration process ends.

[0090] 2. Data Acquisition and Calibration Phase

[0091] As shown in Figure 3, after completing the system calibration, a correction is performed before on-site data collection.

[0092] The system enters the acquisition phase, where a reference plate with a known reflectivity r(λ)(x,y) is laid in the area to be measured. r(λ)(x,y) represents the reflectivity of the reference plate at position (x,y) for wavelength λ light. This is a parameter that is pre-calibrated or provided by the manufacturer. A uniform surface light source is used for illumination to ensure that the light intensity distribution on the surface of the reference plate is uniform, providing stable illumination conditions for optical signal acquisition.

[0093] The camera is controlled to capture images of the reference board, resulting in three images, I1, I2, and I3, which correspond to the light signals of the RGB channels, providing basic data for subsequent color calculations.

[0094] The dispersion module is aligned with the light source to capture the current spectral signal L(λ).

[0095] Calculate the actual intensity E(λ) of the current light source: E(λ) = L(λ) / R(λ). Where L(λ) is the mixed signal measured by the dispersion module, R(λ) is the response curve of the dispersion module obtained during the calibration stage, and E(λ) reflects the actual light intensity of the light source at wavelength λ.

[0096] Based on the calibration parameters and reference board data, calculate the camera correction coefficients K1, K2, and K3 independently for each pixel (x, y) in the image:

[0097] K1(x ,y) = I1(x ,y) / [ Σ ( E(λ) * r(λ)(x ,y) *A1(λ) * T1(λ) ) * Σ( Fx(λ)) ]

[0098] K2(x ,y) = I2(x ,y) / [ Σ ( E(λ) * r(λ)(x ,y) * A2(λ) * T2(λ) ) * Σ( Fy(λ) ) ]

[0099] K3(x ,y) = I3(x ,y) / [ Σ ( E(λ) * r(λ)(x ,y)* A3(λ) * T3(λ) ) * Σ( Fz(λ)) ]

[0100] Where I1, I2, and I3 are the reference board image data corresponding to the three RGB color channels captured by the camera; E(λ) is the current light source intensity; r(λ)(x,y) is the reflectivity of the reference board; A1(λ), ​​A2(λ), and A3(λ) are the response curves of the first mono sensor 13, the second mono sensor 14, and the third mono sensor 15 obtained during the calibration phase, respectively; T1(λ), ​​T2(λ), and T3(λ) are the first filter 10, the second filter 11, and the third filter 12 obtained during the calibration phase, corresponding to the red, green, and blue channels, respectively.

[0101] The transmittance curves; Fx(λ), Fy(λ), and Fz(λ) are the visual functions of the human eye's x, y, and z visual channels, describing the human eye's sensitivity to light of different wavelengths, and are standard physiological optical parameters.

[0102] 3. Measurement Phase

[0103] Remove the reference plate, place the object to be tested in the measurement area, and under the same uniform light source illumination, control the camera to photograph the object to be tested, obtaining three wavelength images I4, I5, and I6. The three images are the light signal images of the object to be tested under the same illumination conditions.

[0104] For each pixel (x, y) in the image, use the correction coefficients calculated above to calculate the spectrally corrected color value:

[0105] R(x,y) = I4(x,y) * K1(x,y)

[0106] G(x,y) = I5(x,y) * K2(x,y)

[0107] B(x,y) = I6(x,y) * K3(x,y)

[0108] Wherein, I4(x,y), I5(x,y), and I6(x,y) are the original R, G, and B data of images I4, I5, and I6 at coordinates (x,y), respectively, and K1(x,y), K2(x,y), and K3(x,y) are camera correction coefficients. The calculation results are the R, G, and B signals of the object under test at this point after spectral signal correction.

[0109] 4. Dynamic calibration steps

[0110] To address the fluctuations in light source spectra over time (t) in industrial environments, this method also includes dynamic calibration:

[0111] The dispersion module is periodically controlled to align with the light source and capture real-time spectral signals L(λ, t). L(λ, t) represents the spectral intensity distribution of the light source measured by the dispersion module at time t, and is used to correct fluctuations in the light source over time.

[0112] Calculate the camera dynamic calibration coefficients K1(t), K2(t), and K3(t):

[0113] K1(t) = I1(x ,y) / Σ (L(λ ,t) / R(λ) * r(λ)(x ,y) * A1(λ) * T1(λ))*Σ (Fx(λ))

[0114] K2(t) = I2 (x ,y) / Σ (L(λ ,t) / R(λ) * r(λ)(x ,y) * A2(λ) * T2(λ)) *Σ (Fy(λ))

[0115] K3(t) = I3 (x ,y) / Σ (L(λ ,t) / R(λ) * r(λ)(x ,y) * A3(λ) * T3(λ)) *Σ (Fz(λ))

[0116] Where L(λ,t) is the real-time spectral signal collected periodically, and the meanings of the other symbols are the same as in step (6). Since the other parameters fluctuate little over time after the first collection, they can be regarded as fixed data. Therefore, the variables over time are only K1(t)(x,y), K2(t)(x,y), K3(t)(x,y) and L(λ,t). Real-time compensation for light source fluctuations is achieved through this calculation.

[0117] For any point (x, y) in space, calculate the values ​​of its R, G, and B channels after light source fluctuation compensation:

[0118] R(x,y) = I4(x,y) * K1(t)(x,y)

[0119] G(x,y) = I5(x,y) * K2(t)(x,y)

[0120] B(x,y) = I6(x,y) * K3(t)(x,y)

[0121] Dynamic calibration coefficients K1(t), K2(t), and K3(t) are used to correct the original signal during the measurement process, eliminating the influence of light source fluctuations on color. This effectively compensates for the effects of light source fluctuations and ensures the long-term stability of the measurement results.

[0122] For any point (x, y) in space, its CIELAB color space value can be calculated based on the corrected RGB values:

[0123] Lab(x,y) = f(R(x,y) , G(x,y) , B(x,y))

[0124] 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 real-time color measurement method with spectral-level calibration, characterized in that, Includes the following steps: Calibration Phase 1: Acquire monochromatic light image sequences and corresponding light intensity data of the system under conditions with and without filters, and calculate the response curve of the mono sensor, the transmittance curve of the filter, and the response curve of the dispersion module; Phase 2: Acquire image data under the current light source using a reference board and calculate the camera correction coefficient; Phase 3: Acquire images of the object under test and calculate its color value based on the correction coefficient. The calibration phase further includes: acquiring the image sequence P(λ) of monochromatic light from 380nm to 780nm using the mono sensor without the filter installed; simultaneously acquiring the light intensity sequence E(λ) measured by the illuminometer and the light intensity sequence L(λ) measured by the dispersion module; acquiring the image sequence Q(λ) again after installing the filter; and calculating the response curve A(λ) = P(λ) / E(λ), the transmittance curve T(λ) = Q(λ) / P(λ), and the dispersion module response curve R(λ) = L(λ). / E(λ); The acquisition phase further includes: using a reference plate with known reflectivity r(λ)(x,y), capturing three-wavelength images I1, I2, and I3 under uniform light source; acquiring the spectral signal L(λ) acquired by the dispersion module, and calculating the current light source intensity E(λ) = L(λ) / R(λ); calculating the camera correction coefficients K1, K2, and K3, respectively based on the comprehensive calculation of I1, I2, I3 with E(λ), r(λ), A(λ), T(λ) and the human eye's visual functions Fx(λ), Fy(λ), and Fz(λ); wherein, the calculation formulas for the camera correction coefficients K1, K2, and K3 are: K1(x,y) = I1(x,y) / [ Σ ( E(λ) * r(λ)(x,y) * A1(λ) * T1(λ) ) * Σ ( Fx(λ)) ]; K2(x,y) = I2(x,y) / [ Σ ( E(λ) * r(λ)(x,y) * A2(λ) * T2(λ) ) * Σ (Fy(λ) ) ]; K3(x,y) = I3(x,y) / [ Σ ( E(λ) * r(λ)(x,y)* A3(λ) * T3(λ) ) * Σ ( Fz(λ)) ]; The measurement stage further includes: capturing three-wavelength images I4, I5, and I6 of the object under test; calculating the corrected R, G, and B values ​​for each pixel (x,y) in the image using correction coefficients K1, K2, and K3; converting the corrected R, G, and B values ​​into CIELAB color space values; Dynamic calibration steps: periodically acquiring the light source spectral signal L(λ,t); updating the dynamic correction coefficients K1(t), K2(t), and K3(t) according to L(λ,t); using the dynamic correction coefficients to compensate for light source fluctuations in the real-time acquired image data.

2. The real-time color measurement method with spectral-level calibration according to claim 1, characterized in that, All image acquisition steps include dark noise correction, which involves acquiring a dark image before shooting as a noise baseline and subtracting it from subsequent calculations.

3. A real-time color measurement system with spectral-level calibration, used to implement the real-time color measurement method with spectral-level calibration as described in claim 1 or 2, characterized in that, include: The viewfinder lens group (1) collects the scene light signal of the target under test; The shutter (2) is located between the viewfinder lens group (1) and the beam splitting module (3), and is connected to the main control board (8) to control the on / off state of the light path and the exposure time; the beam splitting module (3) is located between the viewfinder lens group (1) and the multi-channel filter detection unit to split the incident light beam into multiple paths; the multi-channel filter detection unit includes multiple filters and corresponding mono sensors to collect light signals from multiple color channels; the dispersion module includes a beam collecting module (4), a slit (5), a dispersion grating (6) and a linear array mono sensor (7) to acquire the light source spectral signal; the main control board (8) controls the shutter (2), the multi-channel filter detection unit and the dispersion module to work together and processes the collected light signal data; the angle offset device (9) adjusts the angle of the dispersion module to adapt to light sources in different directions.

4. The real-time color measurement system with spectral-level calibration according to claim 3, characterized in that, The multi-channel filter detection unit includes three independent filters and mono sensors, corresponding to the red, green, and blue color bands, respectively.

5. The real-time color measurement system with spectral-level calibration according to claim 3, characterized in that, The main control board (8) integrates a high-speed processing unit to process multi-channel spectral data in real time and execute color space conversion algorithms.

6. The real-time color measurement system with spectral-level calibration according to claim 3, characterized in that, The system is suitable for online color detection in the textile, electronic display, and plastics processing industries. It achieves high-precision, high-resolution real-time color measurement through spectral-level calibration and dynamic compensation mechanisms.

Images