Line confocal spectral detection device based on reflection hyperspectral and raman spectrum conjugation

Abstract

The application discloses a kind of line confocal spectral detection devices based on reflection hyperspectral and raman spectrum conjugate, can realize the reflection hyperspectral and raman spectrum information of target sample along a detection line simultaneously acquisition.It can adopt two kinds of implementation, respectively applicable to static and dynamic application scenarios.The first kind includes line confocal imaging light path, anastigmatic spectrometer, wavelength range selection system, light source switching system;The second kind includes line confocal imaging light path, hyperspectral imager, anastigmatic spectrometer, light source system.The application not only breaks through the limitation of single structure of prior art, more with its comprehensive advantage of low cost and high performance, static and dynamic detection capability in one, provides a kind of powerful, general and efficient spectral imaging tool for agriculture, industry, biological medicine and scientific research and other fields, has very high market value and application prospect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spectral analysis, in particular to a line confocal spectral detection device based on reflection hyperspectral and Raman spectrum conjugation, which is suitable for various scenes requiring precise chemical composition and physical property analysis in material science, biomedical science, pharmacy, agriculture and industrial sorting. BACKGROUND

[0002] As a powerful substance identification and analysis method, spectral analysis technology has been widely used in various fields of scientific research and industry. According to the different physical principles of interaction with light, molecular spectroscopy mainly includes Raman spectroscopy and infrared absorption spectroscopy. Raman spectroscopy is derived from the phenomenon of inelastic scattering of light and molecular vibration / rotation interaction in linear optical media. When monochromatic light (usually laser) is irradiated onto the sample, most of the light is elastically scattered (Rayleigh scattering), and the frequency does not change; but a small amount of light (about one millionth) is inelastically scattered, i.e. energy exchange occurs between photons and molecules: Stokes line (Stokes): the photon transfers part of the energy to the molecule, and the scattered light frequency decreases (wavelength becomes longer); Anti-Stokes line (Anti-Stokes): the molecule transfers energy to the photon, and the scattered light frequency increases (wavelength becomes shorter). These frequency shifts (unit: cm -1 ) are called Raman shifts, which correspond to the internal vibrational energy level differences of the molecule. Different molecules have completely different vibrational energy level systems due to their unique atomic composition, chemical bond strength and molecular spatial structure, and when interacting with monochromatic light, the molecules will produce scattered light with specific Raman shifts, thereby forming unique peak position, peak intensity and peak shape characteristics. It is this high specificity that makes Raman spectroscopy a powerful "molecular fingerprint" identification tool - just as every person has a unique fingerprint, different substances also have their own exclusive Raman spectrum, which can be used for precise qualitative identification and quantitative analysis of substances, and is widely used in the fields of chemistry, pharmacy and biomedical science.

[0003] On the other hand, hyperspectral imaging (HSI) is based on the selective absorption and reflection of light by molecules. This technique captures the reflection, scattering, or fluorescence signals of each spatial pixel across tens to hundreds of consecutive narrow bands, and correlates these signals with known molecular absorption characteristic spectra to generate a "three-dimensional data cube" (two spatial dimensions and one spectral dimension) containing both spatial and spectral information. The two spatial dimensions refer to the spatial coordinates of the image on a two-dimensional plane, typically represented by the X-axis (row direction) and Y-axis (column direction). These two dimensions together form the "pixel grid" in traditional images, with each point (x, y) corresponding to a spatial location in the image, i.e., a "spatial pixel." The one spectral dimension refers to the spectral information recorded at each spatial pixel location across multiple consecutive narrow bands. Each band corresponds to a specific wavelength (or wavelength range), typically extending from visible light to near-infrared, short-wave infrared, etc. Each spatial pixel is no longer just a grayscale value or an RGB three-channel value, but rather contains reflectance / radiance values ​​across tens to hundreds of bands. These values ​​are arranged in wavelength order, forming the "spectral curve" or "spectral feature" of that pixel. The three-dimensional data cube structure can be visualized as a "cube": the base is the XY plane—a two-dimensional spatial image; the height direction is λ (wavelength)—the spectral dimension; the whole is a three-dimensional matrix of X × Y × λ. From this cube, we can extract: a "single-band grayscale image" (spatial image) from any band; a "spectral curve" (reflecting the spectral characteristics of the substance at that point) from any pixel; and a set of spectral data from any spatial region, used for substance classification or composition analysis, etc.

[0004] Hyperspectral imaging can effectively reflect the external physical structure, composition distribution and state information of the measured object, and plays an important role in environmental remote sensing, agricultural product quality sorting, material classification and other fields.

[0005] Although Raman spectroscopy and hyperspectral imaging provide chemical information about substances from two different dimensions—molecular vibrational fingerprints and macroscopic compositional distribution—and are highly complementary, they still face significant challenges in practical applications. While Raman spectroscopy possesses high specificity, its signal is weak and susceptible to fluorescence interference; traditional point scanning methods have low spatial throughput, making it difficult to quickly acquire information on large-area distributions. Hyperspectral imaging, while capable of rapidly acquiring spatial chemical composition distributions, typically exhibits broad and overlapping spectral features, limiting its ability to finely resolve molecular structures and hindering the accurate differentiation of isomers or complex mixtures.

[0006] In many cutting-edge applications, such as biological tissue pathology diagnosis, drug composition and crystal form analysis, and high-end material quality control, it is often necessary to simultaneously obtain molecular fingerprint identification results and the spatial distribution of chemical composition in the same region to make a comprehensive and reliable judgment. Existing technical solutions usually rely on two independent sets of instruments to perform Raman and hyperspectral measurements separately, which suffers from problems such as high cost, difficulty in spatial matching, cumbersome operation, and low efficiency. Although some studies have attempted to integrate the two technologies into a single system, most solutions are still limited by bottlenecks such as insufficient speed and stability due to mechanical switching, mismatch between field of view and focus due to non-conjugated optical paths, or excessive system complexity and cost.

[0007] Therefore, there is an urgent need in this field to develop a new type of spectroscopic detection system that can fully leverage the molecular fingerprinting capabilities of Raman spectroscopy and the spatial chemical composition analysis advantages of hyperspectral imaging, enabling efficient and spatially consistent detection of the same sample region using both technologies, and providing flexible, reliable, and cost-effective solutions for different application needs. Summary of the Invention

[0008] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a line confocal spectroscopy detection device based on the conjugation of reflectance hyperspectroscopy and Raman spectroscopy.

[0009] The technical solution for achieving the objective of this invention is as follows:

[0010] A linear confocal spectroscopy detection device based on the conjugate of reflectance hyperspectral and Raman spectroscopy, suitable for static sample detection, includes:

[0011] A line confocal imaging optical path, wherein the entrance end of the line confocal imaging optical path is disposed on the sample plane, for receiving and transmitting the line spectral signal generated from the sample surface after being excited by a broadband light source or a line laser light source;

[0012] The aberration-corrected spectrometer has its entrance optically connected to the exit of the line confocal imaging optical path, and is used to receive and analyze line spectral signals.

[0013] A wavelength range selection system, located inside the aberration-corrected spectrometer, includes a rotatable reflection grating for selecting the target wavelength band by adjusting the angle, thereby achieving time-division acquisition of high-resolution Raman spectra and wide-range reflectance hyperspectral spectra.

[0014] The light source switching system includes a broadband light source subsystem and a line laser light source subsystem, which are used to switch the illumination source between the reflection hyperspectral detection mode and the Raman spectral detection mode.

[0015] The aberration-correcting spectrometer includes an entrance slit, a first concave mirror, a reflection grating, a toroidal mirror, and a detector. The toroidal mirror works in conjunction with a cylindrical lens to compensate for astigmatism.

[0016] The broadband light source subsystem includes a broadband light source and a 50% beam splitter; the line laser light source subsystem includes a laser, a line laser shaping optical path, and a dichroic mirror; the light source switching system uses a mechanical drive mechanism to move the beam splitter or dichroic mirror into / out of the optical path to achieve light source switching.

[0017] The original image acquired by the aberration-corrected spectrometer is two-dimensional data, where the X-axis corresponds to the spatial position information on the sample line and the Y-axis corresponds to the wavelength information after grating dispersion. The device also includes a data processor for performing dark current subtraction, flat field correction, spectral calibration, spatial position extraction, and spectral curve reconstruction on the image to form a spectral data cube distributed along the detection line.

[0018] A linear confocal spectroscopy detection device based on the conjugate of reflectance hyperspectroscopy and Raman spectroscopy, suitable for dynamic sample detection, includes:

[0019] A line confocal imaging optical path, wherein the entrance end of the line confocal imaging optical path is disposed on the sample plane, for receiving and transmitting the line spectral signal generated from the sample surface after being excited by a broadband light source or a line laser light source;

[0020] The hyperspectral imager has its entrance optically connected to the exit of the line confocal imaging optical path for synchronous spatial-spectral imaging of line spectral signals.

[0021] The aberration-corrected spectrometer has its entrance optically connected to the exit of the line confocal imaging optical path and is set in parallel with the hyperspectral imager to synchronously receive line spectral signals to obtain Raman spectra.

[0022] The light source system includes a broadband light source subsystem and a line laser light source subsystem. The broadband light source subsystem and the line laser light source subsystem are conjugately coupled to the line confocal imaging optical path through a beam splitter element, which can provide broadband illumination and line laser excitation simultaneously without mechanical switching.

[0023] The hyperspectral imager includes an entrance slit, a collimating lens group, a prism-grating-prism beam splitting module, and a detector. The prism-grating-prism beam splitting module is used to achieve wide-band high-resolution dispersion.

[0024] The light source system includes a 50% beam splitter and a dichroic mirror, which are used to couple the output beams of the broadband light source and the line laser light source along the same optical axis to the line confocal imaging optical path, so as to realize synchronous output of the illumination source without mechanical switching.

[0025] The prism-grating-prism beam splitting module in the hyperspectral imager includes a first prism, a transmission grating, and a second prism. The first prism is used to deflect the incident light, the transmission grating is used for principal dispersion, and the second prism is used to deflect the light again to focus the target center wavelength onto the center of the detector image plane.

[0026] The aberration-correcting spectrometer includes an entrance slit, a first concave mirror, a reflective grating, a toroidal mirror, and a detector. The toroidal mirror is used to compensate for aberrations, and a cylindrical lens is placed in front of the detector to further correct astigmatism.

[0027] The line confocal imaging optical path includes an objective lens, a slit, and a relay lens group, used to achieve line confocal detection of the sample line.

[0028] The hyperspectral imager and the aberration-corrected spectrometer achieve optical conjugation through the same line confocal imaging optical path, and the detectors acquire data synchronously with a time delay of less than 10ms. Under normal conditions, a single-line detection mode is used to acquire spectral data on a fixed line on the sample. When two-dimensional spectral imaging is required, a scanning mechanism is added. For static samples or samples with low resolution requirements, scanning is completed by driving a displacement stage at the sample. For dynamic samples or samples with high resolution and speed requirements, a galvanometer is added to the relay lens group in the line confocal imaging optical path to complete rapid scanning.

[0029] A spectral data processing method based on the aforementioned device includes the following steps:

[0030] (1) Image preprocessing: Dark current subtraction and flat field correction are performed on the acquired two-dimensional raw spectral image;

[0031] (2) Spectral calibration: Establish the correspondence between the Y-axis pixel position and the wavelength using a standard light source;

[0032] (3) Spatial position extraction: Identify each pixel column along the X-axis, and each column corresponds to the spectral information of a spatial point on the sample line;

[0033] (4) Spectral curve extraction: Extract the light intensity value along the Y-axis for each pixel column to form the reflectance hyperspectral or Raman spectral curve of the corresponding point;

[0034] (5) Data reconstruction: Arrange the spectral curves of all spatial points in spatial order to construct a three-dimensional spectral data cube with a continuous linear distribution.

[0035] The beneficial effects of this invention are:

[0036] The line confocal spectral imaging device provided by this invention integrates two configurable optical structures into a single technical solution, bringing significant and multi-dimensional benefits. Its core advantage lies in successfully resolving the long-standing dilemma in traditional spectral imaging technology regarding the conflict between cost, performance, and application scenarios, achieving unprecedented flexibility and practicality. Specifically, the first imaging structure based on mechanical switching, with its simplified components and multiplexed spectrometer structure, significantly reduces the manufacturing cost and maintenance difficulty of the equipment. It provides a highly cost-effective and reliable solution for static sample detection where time resolution requirements are not high, making the widespread application of high-performance spectral detection technology in budget-constrained situations or conventional laboratories possible. The second multi-path conjugate synchronous imaging structure, through its ingenious optical design, achieves sample capture by all spectral channels at strictly simultaneous moments, completely eliminating motion artifacts caused by time-division scanning. This effectively solves the problem of accurate measurement in dynamic samples, rapidly changing processes, or online automated detection, greatly expanding the application boundaries of spectral imaging technology. More importantly, the coexistence of two structures under a single inventive concept empowers users to make the optimal choice based on their specific detection needs—whether pursuing cost-effective static analysis or requiring high-fidelity dynamic capture. This flexibility avoids redundant investment and equipment redundancy, achieving optimal resource allocation. In summary, this invention not only overcomes the limitations of existing single-structure technologies but also provides a powerful, versatile, and efficient spectral imaging tool for multiple fields such as agriculture, industry, biomedicine, and scientific research with its comprehensive advantages of combining low cost and high performance with static and dynamic detection capabilities. It possesses extremely high market value and application prospects.

[0037] Of course, not all of the above-mentioned beneficial effects can be achieved by any of the technical solutions of the present invention. Attached Figure Description

[0038] Figure 1a The first embodiment of the line confocal spectroscopy detection device based on the conjugate of reflectance hyperspectroscopy and Raman spectroscopy is shown, which is suitable for the detection of static samples.

[0039] Figure 1b The image shows a second embodiment of a line confocal spectroscopy detection device based on the conjugation of reflectance hyperspectroscopy and Raman spectroscopy, which is suitable for the detection of dynamic samples.

[0040] Figure 2a The Raman spectrum of ABS microplastic particles is shown.

[0041] Figure 2b The Raman spectrum of LLDPE microplastic particles is shown.

[0042] Figure 2cThe Raman spectrum of POM microplastic particles is shown.

[0043] Figure 2d The Raman spectrum of PET microplastic particles is shown.

[0044] Figure 3 Images of four microplastics are shown: acrylonitrile-butadiene-styrene copolymer (ABS), linear low-density polyethylene (LLDPE), polyoxymethylene (POM), and polyethylene terephthalate (PET).

[0045] In the figure, the components are: 1. Aberration-corrected spectrometer; 2. First concave mirror; 3. Toroidal mirror; 4. Cylindrical lens; 5. Detector; 6. Reflection grating; 7. Spectrometer entrance slit; 8. Line laser source subsystem; 9. Laser; 10. Line laser shaping optical path; 11. Dichroic mirror; 12. Broadband source subsystem; 13. Line confocal imaging optical path; 14. Sample to be tested; 15. Hyperspectral imager; 16. Detector; 17. Prism-grating-prism beam splitting module; 18. Hyperspectral imager entrance slit; 19. Collimating and focusing lens group; 20. Wavelength range selection system. Detailed Implementation

[0046] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0047] Example 1: Detection of Microplastics in Liquids

[0048] This embodiment provides a first implementation of a line spectral detection device based on a conjugate system of reflectance hyperspectral and Raman spectroscopy, such as... Figure 1a As shown.

[0049] The detection device includes an aberration-corrected spectrometer 1, a wavelength range selection system 20, a line laser source subsystem 8 and a broadband source subsystem 12, a line confocal imaging optical path 13 and a sample to be tested 14.

[0050] Figure 1a The structure of the aberration-corrected spectrometer is also shown in the image. The aberration-corrected spectrometer 1 includes an entrance slit 7, a first concave mirror 2, a reflection grating 6, a toroidal mirror 3, a cylindrical lens 4, and a detector 5. This structure improves imaging quality by synergistically compensating for optical astigmatism through the toroidal mirror and the cylindrical lens.

[0051] Figure 1a The composition of the wavelength range selection system is also shown. The system consists of a mechanical rotary table and a reflective grating 6 installed on it. By rotating the grating to switch the detection band, the system can achieve segmented acquisition and stitching of high-resolution Raman signals and wide-range hyperspectral data.

[0052] Figure 1a also shows the structure of the light source switching system, including a line laser light source subsystem 8 and a broadband light source subsystem 12. The line laser light source subsystem consists of a laser 9, a line laser shaping optical path 10, and a dichroic mirror 11; the broadband light source subsystem consists of a broadband light source and a beam splitter. The optical path switching between the two illumination modes is achieved through mechanical switching elements.

[0053] To verify the practical application effect of the present invention, Raman spectroscopy was used to detect microplastics extracted from sea salt solution. The steps are as follows:

[0054] 1. Sample preparation: Microplastic particles were extracted from sea salt samples. A certain amount of sea salt sample was dissolved in ultrapure water, and the solution was fully dissolved by stirring or sonication, so that the microplastic particles encapsulated inside the salt crystals were released into the solution. Subsequently, the solution was filtered in stages using a vacuum filtration device and a series of filter membranes with different pore sizes (500 μm, 100 μm and 10 μm, respectively) to gradually separate plastic particles of different sizes. After filtration, the filter membranes that finally retained the target microplastics were carefully rinsed with ultrapure water and dispersed in deionized water to form a uniform microplastic solution.

[0055] 2. Detection process: Place the microplastic solution in the detection device and select the Raman spectroscopy detection mode. Activate the line laser source subsystem 8 through the light source switching system. The laser beam is focused onto the microplastic particles in the solution through the line laser shaping optical path 10, exciting their Raman scattering signals.

[0056] 3. Spectral Acquisition: The reflected light signal generated after the sample is excited is collected and collimated by the linear confocal imaging optical path 13. This collimated light then passes through an optional filter and is subsequently guided into the spectrometer entrance slit 7. The light entering the spectrometer is first collimated into parallel light by the first concave mirror 2, and then illuminates the core dispersive element—the reflective grating 6—achieving precise dispersion and spectral separation of different wavelengths. The dispersed beam is converged by the toroidal mirror 3, ultimately forming a clear spectral image on the high-sensitivity detector 5 (such as a CCD or CMOS sensor), thus completing the high-resolution Raman spectral data acquisition.

[0057] 4. Spectral range selection: The wavelength range selection system 20 uses a precision motor to drive and rotate the reflective grating 6 to switch the center band, ensuring that the system can perform high-resolution acquisition in different spectral ranges.

[0058] 5. Data Analysis: After image preprocessing, spectral calibration, spatial location extraction, and spectral curve extraction, the acquired two-dimensional spectral data were used to generate Raman spectral curves for the microplastic particles. By comparing these curves with a standard Raman spectral library, four types of microplastics were successfully identified, such as... Figures 2a-2dThe images shown are of four microplastics: acrylonitrile-butadiene-styrene copolymer (ABS), linear low-density polyethylene (LLDPE), polyoxymethylene (POM), and polyethylene terephthalate (PET). Figure 3 As shown.

[0059] Data processing includes the following steps:

[0060] Image preprocessing: Dark current subtraction and flat field correction are performed;

[0061] Spectral calibration: Establishing the pixel-wavelength correspondence using a mercury lamp or other standard light source. The calibration formula can be expressed as:

[0062] ;

[0063] in Lambda For wavelength, p For pixel position, a , b , c These are the scaling factors;

[0064] Spatial location extraction: Each spatial pixel column is identified along the X-axis, and each pixel column corresponds to the complete spectral information of a specific spatial point on the sample line;

[0065] Spectral curve extraction: For each spatial pixel, extract the light intensity values ​​of all wavelength channels along the Y-axis to form the reflectance hyperspectral or Raman spectral curve of that point.

[0066] Example 2: Dynamic Detection System for Fluorescence and Raman Spectroscopy of Microfluidic Chips Based on the Second Embodiment

[0067] This embodiment provides an implementation method for rapid dynamic process detection, and its structure is as follows: Figure 1b As shown, it is suitable for rapid, time-resolved detection of fluorescence and surface-enhanced Raman spectroscopy (SERS) signals in self-driven microfluidic chips.

[0068] The detection device includes an aberration-corrected spectrometer 1, a hyperspectral imager 15, a line laser source subsystem 8, a broadband source subsystem 12, a line confocal imaging optical path 13, and a sample to be tested 14.

[0069] Figure 1b The image also shows the structure of the hyperspectral imager 15, which mainly includes a hyperspectral imager entrance slit 18, a collimating and focusing lens group 19, and a prism-grating-prism beam splitting module 17. The prism-grating-prism structure achieves wide-band high-resolution beam splitting by sequentially passing through a first prism pre-deflection, a transmission grating dispersion, and a second prism re-deflection.

[0070] The device includes an aberration-corrected spectrometer 1, a hyperspectral imager 15, a line laser source subsystem 8, a broadband source subsystem 12, and a line confocal imaging optical path 13. The line laser source subsystem 8 integrates a line laser source, and the broadband source subsystem 12 integrates a broadband source, achieving millisecond-level illumination switching between different spectral excitation modes through high-speed electronic control. The line confocal imaging optical path 13 employs a beam splitting and relay design, simultaneously imaging the fluorescence and scattering signals formed at the microfluidic chip detection window onto the slits of the aberration-corrected spectrometer 1 and the hyperspectral imager 15. The Raman line spectral data acquisition process is the same as step 3 in Embodiment 1. The hyperspectral imager uses a prism-grating-prism PGP beam splitting structure, whose optical path sequentially includes an entrance slit 18, a collimating and focusing lens group 19, a first prism, a transmission grating, and a second prism forming a prism-grating-prism beam splitting module 17, ultimately forming a spectral image on the area array detector. PGP (Platelet-Glass Spectroscopic Probe) structures can achieve high-bandwidth, high-throughput spectral dispersion, and their dispersion linearity can be approximated by the following formula:

[0071] ;

[0072] Where θ is the diffraction angle, λ is the wavelength, m is the diffraction order, and σ is the grating constant. This system coordinates light source switching and detector acquisition through a synchronous controller, achieving millisecond-level timing response and enabling near-synchronous tracking of fluorescence and Raman signals within the same detection line region of a microfluidic chip. For high-concentration samples, a broadband light source is used for excitation, and a hyperspectral imager is employed to acquire time-varying fluorescence intensity information. For low-concentration, difficult-to-detect samples, a laser light source is switched for excitation, and an aberration-corrected spectrometer is used to acquire high-resolution surface-enhanced Raman (SERS) time-series signals. This system has no moving mechanical parts, boasts a high spectral acquisition rate, and is particularly suitable for rapid dynamic analysis and process monitoring of multi-target, multi-modal spectral information in microfluidic environments.

[0073] Example 3: Rapid Cell Hyperspectral and Raman Spectroscopy Imaging System Based on Targeted Localization

[0074] This embodiment provides an implementation method for rapid, targeted multimodal spectral imaging of biological samples (such as cell smears, tissue sections, etc.). Its core lies in combining the rapid wide-field recognition capability of hyperspectral imaging with the high specificity of Raman spectroscopy. Through image recognition and precise positioning of the displacement platform, it achieves efficient Raman spectral acquisition of specific target areas (such as cells of specific morphology), thereby significantly shortening the search and total detection time for rare or key targets.

[0075] This solution is based on the second embodiment described above, and integrates a high-precision electric displacement platform at the sample to be tested 14. The device includes an aberration-corrected spectrometer 1, a hyperspectral imager 15, a line laser source subsystem 8, a broadband source subsystem 12, and a line confocal imaging optical path 13.

[0076] Its typical workflow is as follows:

[0077] (1) Rapid scanning and target identification stage: The fixed cell sample is placed on a high-precision displacement platform. First, the control system shuts down the line laser source subsystem 8 and turns on the broadband source subsystem 12, switching to hyperspectral imaging mode. Then, the control displacement platform drives the sample 14 to be tested to perform a rapid one-dimensional scan, and the hyperspectral imager 15 simultaneously and continuously acquires data, obtaining a hyperspectral data cube of the entire sample area in a short time. After acquisition, the hyperspectral image is processed by the built-in image recognition algorithm (e.g., a cell classification algorithm based on morphological features or specific spectral features), automatically identifying and locating the spatial coordinates X of the target cell of interest.

[0078] (2) Precise localization and Raman acquisition stage: Based on the target cell coordinates obtained in the previous stage, the displacement platform is controlled to precisely move the target cell to the detection line of the line confocal imaging optical path 13. Subsequently, the broadband light source is turned off, the line laser light source subsystem 8 is turned on, and the Raman line spectrum acquisition mode is switched. The aberration-corrected spectrometer 1 performs localized Raman spectral data acquisition on the target cell that is stationary on the detection line. Since the Raman signal is relatively weak, this stage may require a certain integration time to ensure the signal-to-noise ratio.

[0079] (3) Data fusion and analysis: The Raman spectra of the target cells with high chemical specificity are obtained and correlated and fused with their morphology and reflectance / fluorescence spectral information in the hyperspectral image to provide more comprehensive information for cell analysis.

[0080] This embodiment solves the bottleneck problem of slow acquisition speed in Raman spectroscopy imaging, which makes it difficult to quickly screen large-scale samples, by adopting a strategy of "rapid global reconnaissance first, followed by precise point-to-point analysis." It avoids the extremely long time consumed by "blindly scanning" the entire area in traditional Raman imaging, and only measures the key targets identified by the algorithm, greatly improving detection efficiency.

[0081] In summary, this invention provides a line spectral detection solution suitable for different application scenarios through two optical structure designs, effectively achieving technical complementarity and efficient acquisition of reflectance hyperspectral and Raman spectroscopy.

[0082] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications and substitutions should all be covered within the scope of protection of the present invention.

Claims

1. A linear confocal spectroscopy detection device based on the conjugate of reflectance hyperspectral and Raman spectroscopy, suitable for static sample detection, characterized in that, include: A line confocal imaging optical path is provided, with its entrance end positioned on the sample plane to receive and transmit line spectral signals generated from the sample surface after excitation by a broadband light source or a line laser light source. The line confocal imaging optical path includes an objective lens, a slit, and a relay lens group to achieve line confocal detection of sample lines. The aberration-corrected spectrometer has its entrance optically connected to the exit of the line confocal imaging optical path, and is used to receive and analyze line spectral signals. A wavelength range selection system, located inside the aberration-corrected spectrometer, includes a rotatable reflection grating for selecting the target wavelength band by adjusting the angle, thereby achieving time-division acquisition of high-resolution Raman spectra and wide-range reflectance hyperspectral spectra. The light source switching system includes a broadband light source subsystem and a line laser light source subsystem, which are used to switch the illumination source between the reflection hyperspectral detection mode and the Raman spectral detection mode. The aberration-corrected spectrometer includes an entrance slit, a first concave mirror, a reflective grating, a toroidal mirror, and a detector. Light entering the spectrometer is first collimated into parallel light by the first concave mirror, then illuminates the reflective grating to achieve dispersion and separation of different wavelengths. The dispersed light beam is converged by the toroidal mirror, ultimately forming a spectral image on the detector. The toroidal mirror, in conjunction with a cylindrical lens, is used to compensate for astigmatism. The broadband light source subsystem includes a broadband light source and a 50% beam splitter; the line laser light source subsystem includes a laser, a line laser shaping optical path, and a dichroic mirror; the light source switching system uses a mechanical drive mechanism to move the beam splitter or dichroic mirror into / out of the optical path to achieve light source switching.

2. The line confocal spectroscopy detection device according to claim 1, characterized in that, The original image acquired by the aberration-corrected spectrometer is two-dimensional data, where the X-axis corresponds to the spatial position information on the sample line and the Y-axis corresponds to the wavelength information after grating dispersion. The device also includes a data processor for performing dark current subtraction, flat field correction, spectral calibration, spatial position extraction, and spectral curve reconstruction on the image to form a spectral data cube distributed along the detection line.

Images