Calculation spectral imaging system based on spectral response enhancement of color detector

Abstract

The invention discloses a spectral imaging calculation system based on spectral response enhancement of a color detector, and belongs to the technical field of spectral imaging, the spectral imaging calculation system comprises a birefringence light splitting module and a color detector, the birefringence light splitting module comprises a polarizer, a birefringence crystal and a polarization analyzer, the polarizer, the birefringent crystal, the polarization analyzer and the color detector are sequentially and coaxially arranged along an optical axis; the color detector is also electrically connected with a signal processing module; incident light is converted into linearly polarized light through the polarizer, the linearly polarized light is decomposed into o light and e light through the birefringent crystal, and the o light and the e light are combined into emergent light through the polarization analyzer. The emergent light passes through a color detector to obtain a measurement signal; and the signal processing module receives the measurement signal and reconstructs a spectral image in combination with the spectral response matrix of the birefringence light splitting module and the spectral response function of the color detector. The spectral response characteristics of multiple color channels of the color image sensor are combined with the birefringence color development polarization principle, the rank of a spectral response matrix is expanded, and the spectral resolution and the imaging precision are improved, so that the spectral information of a substance is more accurately acquired and analyzed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spectral imaging technology, and in particular to a computing spectral imaging system based on color detector spectral response enhancement. BACKGROUND

[0002] A spectral imaging system is a device that combines imaging and spectral technology to obtain spatial and spectral information of a target simultaneously. It is based on the difference in response characteristics of different substances or elements to different wavelengths of light, and cooperates with optical imaging systems, spectral devices, detectors, etc. to obtain a spectral image data cube. Depending on the type of spectral element, it can be divided into dispersive, interferometric, and filter types. The advantages of a spectral imaging system are high information content, multi-field application, and non-contact measurement. However, it also has limitations such as large data volume, complex system, and trade-off between spectral and spatial resolution. Today, spectral imaging systems have been widely used in remote sensing, biomedical, industrial detection, environmental monitoring, and agriculture, providing key support for the development of various fields.

[0003] In a traditional spectral imaging system based on the birefringence principle, the propagation difference of different polarized light by birefringent crystals or materials is mainly used to realize spectral splitting and imaging. Specifically, natural light is converted into linearly polarized light by a polarizer, and the linearly polarized light is decomposed into ordinary light (o light) and extraordinary light (e light) by a birefringent element. The propagation direction and speed of the two beams of light are different in the element, and the difference in the propagation direction is related to the wavelength of the light. When the optical axis is parallel to the incident plane, the o light and e light have only different optical paths when they exit. After passing through an analyzer, the o light and e light are combined and interfere. Because the optical path differences of o light and e light of different wavelengths are different, the exit light intensities of different wavelengths are different. Rotating the polarizer or analyzer and changing the optical axis angle of the birefringent element will change the composition of the exit wavelength. By calibrating the spectral composition of the exit light in different states, the spectral response matrix of the spectral imaging system can be obtained, and thus the spectral distribution of any incident light can be inversely solved.

[0004] However, the spectral response matrix obtained by this method is often not full rank or even has a low matrix rank. For a computing spectral imaging system, although the spectral resolution can be improved using compressed sensing theory, the low rank of the spectral response matrix still brings problems such as unstable solution and large error of the solution, which limits the practical application of the computing spectral imaging system based on the birefringence principle. SUMMARY

[0005] The present application aims to overcome the deficiencies in the prior art, and provide a kind of based on the spectral response enhancement of color detector computational spectral imaging system, by combining the spectral response characteristics of multiple color channels of color image sensor with birefringent color development polarization principle, expand the rank of spectral response matrix, improve spectral resolution and imaging accuracy, to more accurately obtain and analyze the spectral information of material.

[0006] To achieve the above object, the present application is implemented by using the following technical solutions:

[0007] The present application provides a kind of based on the spectral response enhancement of color detector computational spectral imaging system, including birefringent spectrometer module and color detector, the birefringent spectrometer module includes polarizer, birefringent crystal and polarimeter, the polarizer, birefringent crystal, polarimeter and color detector are coaxially arranged along optical axis in turn;Signal processing module is also electrically connected on the color detector;

[0008] Incident light is converted into linearly polarized light by the polarizer, and the linearly polarized light is decomposed into o light and e light by the birefringent crystal, and the o light and e light are combined into outgoing light by the polarimeter;The outgoing light is measured by the color detector to obtain a signal;

[0009] The signal processing module receives the measurement signal and combines the spectral response matrix of the birefringent spectrometer module and the spectral response function of the color detector to reconstruct a spectral image.

[0010] Optionally, the computational spectral imaging system further comprises an imaging lens; the imaging lens is arranged on the front side or rear side of the polarizer or the polarimeter.

[0011] Optionally, the birefringent crystal is made of calcite, iceland spar or liquid crystal.

[0012] Optionally, the polarizer or the polarimeter is provided with a rotation driver for driving the rotation of the polarizer or the polarimeter along the optical axis.

[0013] Optionally, the calibration process of the spectral response matrix of the birefringent spectrometer module is as follows:

[0014] The rotation angle of the polarizer or the polarimeter is taken as the first variable, and the wavelength of the incident light is taken as the second variable, and the spectral composition ratio of the outgoing light generated by the birefringent spectrometer module under the first variable and the second variable is obtained by the spectrometer to construct the spectral response matrix.

[0015] Optionally, the calibration process of the spectral response function of the color detector is as follows:

[0016] The monochromator scans each wavelength and outputs uniform collimated light signals of red, green and blue single colors respectively, and the color detector obtains measurement signals of red, green and blue channels.

[0017] The measurement signal is proportional to the uniform collimated light signal to generate spectral response functions of the red, green and blue channels corresponding to each wavelength.

[0018] Optionally, the signal processing module first performs enhancement processing on the measurement signal after receiving it, and the enhancement processing includes signal amplification, signal filtering and signal denoising.

[0019] Optionally, the signal processing module receives the measurement signal and reconstructs a spectral image in combination with the spectral response matrix of the birefringence spectrometer module and the spectral response function of the color detector, including:

[0020] Constructing an extended spectral response matrix according to the spectral response matrix and the spectral response function :

[0021]

[0022]

[0023]

[0024]

[0025] Where, Respectively wavelength The corresponding spectral response functions of the red, green, and blue channels are, For the The wavelength is The proportion of spectral components under different rotation angles; , is the number of rotation angles, , is the number of wavelengths; Respectively The wavelength is The proportion of spectral components of red, green and blue channels under different rotation angles;

[0026] The signal processing module receives the measurement signal to obtain the measurement matrix of the red, green and blue channels at each rotation angle. :

[0027]

[0028]

[0029]

[0030]

[0031] In the formula, are the measured vectors of the red, green and blue channels respectively under each rotation angle, are the measured vectors of the red, green and blue channels respectively under the first rotation angle;

[0032] Based on the extended spectral response matrix and the measurement matrix , the spectral image of the incident light is solved by using the least square method and the Adam gradient descent method , is the spectral signal value corresponding to the first wavelength of the incident light , and the optimization problem equation to be solved is:

[0033] .

[0034] Compared with the prior art, the beneficial effects achieved by the present application are:

[0035] The present application provides a kind of computing spectral imaging system based on color detector spectral response enhancement, 1) synergistic combination of color detector multichannel spectral response and birefringent color development polarization principle, effectively expand the rank of spectral response matrix, improve spectral resolution and imaging accuracy, can more accurately obtain the spectral information of material, provide more reliable data support for material composition analysis etc. Application makes full use of the multichannel spectral characteristics of color detector, without additional complex optical elements, to a certain extent, simplify the system structure, reduce cost, while improving the integration and portability of system. Least square method is used to combine Adam gradient descent method for spectral recovery calculation, improve the calculation efficiency and stability, can quickly and accurately reconstruct high-resolution spectral image, meet the requirements of spectral analysis speed and accuracy in practical application. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 It is the structure schematic diagram of the computing spectral imaging system based on color detector spectral response enhancement provided by the embodiment of the present application;

[0037] Figure 2 It is the contrast schematic diagram of original birefringent spectral response matrix and extended spectral response matrix combined with RGB channel provided by the embodiment of the present application;

[0038] Figure 3 It is the contrast schematic diagram of recovery effect before and after spectral response matrix combined with RGB channel using same solving algorithm and iteration number provided by the embodiment of the present application. DETAILED DESCRIPTION

[0039] The application will be further described below with reference to the drawings. The following examples are only used to more clearly illustrate the technical solutions of the application, and cannot be used to limit the protection scope of the application.

[0040] Example 1

[0041] The embodiment of the application provides a kind of based on color detector spectral response enhancement's computing spectral imaging system, including birefringent spectrometer module and color detector, birefringent spectrometer module includes polarizer, birefringent crystal and polarimeter, polarizer, birefringent crystal, polarimeter and color detector are coaxially arranged in sequence along optical axis;Color detector is also electrically connected with signal processing module.Birefringent crystal is made of calcite, ice stone, liquid crystal or other birefringent material, the crystal optical axis of birefringent crystal is parallel to the upper and lower surfaces of crystal, and is perpendicular to incident light.The working principle of computing spectral imaging system is:

[0042] Incident light is converted into linearly polarized light by polarizer, linearly polarized light is decomposed into o light and e light by birefringent crystal, and o light and e light are combined into outgoing light by polarimeter;Outgoing light obtains measurement signal by color detector;Signal processing module receives measurement signal and combines the spectral response matrix of birefringent spectrometer module and the spectral response function of color detector to reconstruct spectral image.

[0043] O light (ordinary light) and e light (extraordinary light) are combined by polarimeter, two kinds of polarized light interfere, and white light mixed with multiple wavelengths undergoes color polarization phenomenon after passing through birefringent crystal, and wavelength composition of outgoing light changes with different angles of rotation of polarizer or polarimeter, and the change is calibrated to obtain spectral response matrix.

[0044] Specifically in the embodiment, the calibration process of the spectral response matrix of birefringent spectrometer module is:

[0045] The rotation angle of polarizer or polarimeter is taken as the first variable, and the wavelength of incident light is taken as the second variable, and the spectral composition ratio of outgoing light generated by birefringent spectrometer module under the first variable and the second variable is obtained by spectrometer to construct spectral response matrix.

[0046] Since polarizer or polarimeter needs to be rotated, rotation driver capable of driving polarizer or polarimeter to rotate along optical axis is arranged on the structure of polarizer or polarimeter.

[0047] The computing spectral imaging system also includes imaging lens;Imaging lens is arranged on the front side or rear side of polarizer or polarimeter. Figure 1As shown, in actual setting, a rotation driver is arranged on the polarizer, and the imaging lens is arranged at the rear side of the polarizer for facilitating assembly of the rotation driver. At this time, the polarizer, the imaging lens, the birefringent crystal, the analyzer and the color detector are arranged along the optical axis in sequence to form the spectral imaging system of the embodiment.

[0048] The color detector is usually composed of an image sensor with a multi-color filter. The image sensor has specific spectral response functions for different wavelengths of light in three color channels of red (R), green (G) and blue (B).

[0049] In the embodiment, the calibration process of the spectral response functions of the color detector is as follows:

[0050] The monochromator scans each wavelength and outputs uniform collimated light signals of red, green and blue colors respectively, and the color detector acquires measurement signals of the red, green and blue channels.

[0051] The measurement signals are proportionally generated into the spectral response functions of the red, green and blue channels corresponding to each wavelength.

[0052] In order to ensure the accuracy of the measurement signals acquired by the color detector, the signal processing module first performs enhancement processing on the measurement signals after receiving the measurement signals, and the enhancement processing includes signal amplification, signal filtering and signal denoising.

[0053] Since only based on the birefringence principle, the proportion of different wavelengths of light components only changes with the sine of the rotation angle, so the rank of the matrix is relatively low, which limits the accuracy of spectral recovery. The signal processing module of the embodiment receives the measurement signals and combines the spectral response matrix of the birefringence spectrometer module and the spectral response function of the color detector to reconstruct the spectral image. The process is as follows:

[0054] (1) Constructing an extended spectral response matrix according to the spectral response matrix and the spectral response function :

[0055]

[0056]

[0057]

[0058]

[0059] In the formula, and are the spectral response functions of the red, green and blue channels corresponding to the th wavelength, respectively, is the th wavelength, is the th wavelength in the th channel, and is the th wavelength in the th channel. ​​​​​The spectral component proportion at each rotation angle; , is the number of rotation angles, , is the number of wavelengths; is the spectral component proportion of the first wavelength in the red, green, and blue channels at the first rotation angle;

[0060] In this way, the extended spectral response matrix combines the multi-channel spectral response information of the color detector and the spatial information of the birefringent spectrometer, and the rank is significantly improved, providing a foundation for more accurate spectral recovery.

[0061] As shown in Figure 2 , a comparison diagram of the original birefringent spectral response matrix and the extended spectral response matrix combined with the RGB channel is shown, which intuitively shows the improvement of the rank of the extended matrix and the increase of the information dimension.

[0062] (2) According to the signal processing module receiving the measurement signal, the measurement matrix of the red, green, and blue channels at each rotation angle is obtained :

[0063]

[0064]

[0065]

[0066]

[0067] In the formula, are the measurement vectors of the red, green, and blue channels at each rotation angle, are the measurement vectors of the red, green, and blue channels at the first rotation angle;

[0068] (3) Based on the extended spectral response matrix and the measurement matrix , the spectral image of the incident light is solved by using the least square method and the Adam gradient descent method , is the spectral signal value corresponding to the first wavelength of the incident light , and the optimization problem equation for solving is:

[0069] .

[0070] The Adam gradient descent method can effectively avoid the problems of slow convergence, falling into a saddle point or local optimal value in the traditional iterative method, and improve the calculation efficiency. In the solving process, the spectral vector is updated by continuous iteration, so that the objective function converges to the minimum value, thereby obtaining the restored spectral image.

[0071] As shown in Figure 3 The left graph is the effect of the scheme in this paper, and the right graph is the spectral recovery effect without color detector multi-channel spectral response enhancement.

[0072] After solving, the performance of the calculated spectral imaging system can be further verified and optimized. a). Test the system using standard samples with known spectra, compare the recovered spectra with the standard spectra, and evaluate the spectral resolution and accuracy of the system. b). According to the test results, optimize and adjust the system, such as optimizing the optical path design of the birefringence spectrometer module, adjusting the working parameters of the RGB color channel detection module, improving the signal processing algorithm, etc., to further improve the performance and reliability of the system.

[0073] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" and the like are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined with "first", "second" and the like can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0074] In the description of the present application, it should be noted that, unless otherwise specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication between the two elements inside. For those skilled in the art, the specific meaning of the above terms in the present application can be understood through specific circumstances.

[0075] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims

1. A computational spectral imaging system based on color detector spectral response enhancement, characterized in that: The invention comprises a birefringence spectroscopic module and a color detector, wherein the birefringence spectroscopic module comprises a polarizer, a birefringence crystal and an analyzer, wherein the polarizer, the birefringence crystal, the analyzer and the color detector are coaxially arranged along the optical axis in sequence; the color detector is also electrically connected to a signal processing module; The incident light is converted into linearly polarized light by the polarizer, the linearly polarized light is decomposed into o light and e light by the birefringent crystal, the o light and e light are combined into outgoing light by the analyzer; the outgoing light is passed through the color detector to obtain a measurement signal; The signal processing module receives the measurement signal and reconstructs a spectral image by combining the spectral response matrix of the birefringence spectroscopic module and the spectral response function of the color detector.

2. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The computational spectral imaging system further includes an imaging lens; the imaging lens is arranged in front of or behind the polarizer or the analyzer.

3. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The birefringent crystal is made of calcite, Iceland spar or liquid crystal.

4. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The polarizer or the analyzer is provided with a rotation driver for driving the polarizer or the analyzer to rotate along the optical axis.

5. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The calibration process of the spectral response matrix of the birefringence spectrometer module is as follows: The rotation angle of the polarizer or analyzer is used as the first variable, and the wavelength of the incident light is used as the second variable. The spectral component ratio of the outgoing light generated by the birefringence spectrometer under the first and second variables is obtained by a spectrometer to construct a spectral response matrix.

6. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The calibration process of the spectral response function of the color detector is: The monochromator scans each wavelength and outputs uniform collimated light signals of red, green and blue monochromatic colors respectively, and the color detector obtains the measurement signals of the red, green and blue channels; The measurement signal is proportional to the uniform collimated light signal to generate spectral response functions of the red, green and blue channels corresponding to each wavelength.

7. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: After receiving the measurement signal, the signal processing module first performs enhancement processing on the measurement signal, and the enhancement processing includes signal amplification, signal filtering and signal denoising.

8. The computational spectral imaging system based on color detector spectral response enhancement according to claim 1, characterized in that: The signal processing module receives the measurement signal and reconstructs a spectral image by combining the spectral response matrix of the birefringence spectrometer module and the spectral response function of the color detector, including: Constructing an extended spectral response matrix according to the spectral response matrix and the spectral response function : Where, Respectively wavelength The corresponding spectral response functions of the red, green, and blue channels are, For the The wavelength is The proportion of spectral components under different rotation angles; , is the number of rotation angles, , is the number of wavelengths; Respectively The wavelength is The proportion of spectral components of red, green and blue channels under different rotation angles; The signal processing module receives the measurement signal to obtain the measurement matrix of the red, green and blue channels at each rotation angle. : Where, are the measurement vectors of the red, green and blue channels at each rotation angle, Respectively The measurement vectors of the red, green and blue channels under the rotation angles; Based on the extended spectral response matrix and the measurement matrix , using the least squares method and Adam gradient descent method to solve the spectral image of the incident light , The incident light wavelength The corresponding spectral signal value, the optimization problem equation to be solved is: 。

Images