33results about How to "Retain structure" patented technology

The invention relates to the technical field of semantic segmentation, in particular to a night semantic segmentation method based on low illumination enhancement and edge optimization, and the method comprises the steps: inputting an image into a low-light enhancement repair network based on the Retinex theory, local contrast enhancement and adaptive feature fusion, obtaining a denoised and enhanced intermediate image, and carrying out the edge optimization of the intermediate image; inputting the intermediate image into a semantic segmentation network to obtain a category distribution diagram of each pixel; inputting a discriminator embedded with a channel attention module according to the category distribution diagram of each pixel, and optimizing a generator composed of a low light enhancement repair network and a semantic segmentation network through a multi-task joint optimization loss function; and inputting a night image to be detected and segmented into the optimized generator, and outputting a segmentation result. By adopting the method, the low-light enhancement repair network is combined with a local contrast enhancement and channel feature fusion mechanism, so that the overall brightness of the image is improved, the details and edge structures of the image are reserved, and the perception capability and robustness of the model are improved.

The invention discloses a polarization degree and light intensity image fusion method based on bidirectional cross attention. The method belongs to the technical field of polarization imaging, computational imaging and image processing. The technical problems that in the prior art, dynamic balance between structural detail keeping and polarization saliency enhancing is generally difficult to achieve, and particularly under the condition of a complex background or a low signal-to-noise ratio, texture missing or noise artifacts are prone to occurring in a fusion result are solved. A fusion mechanism capable of establishing a two-way information interaction relationship between a light intensity image and a polarization degree image is provided, and further processing is carried out aiming at a noise problem, so that more stable and more reliable polarization fusion imaging is realized.

The application discloses a Bayer image denoising method, which comprises the following steps: obtaining a Bayer image and performing a demosaicing operation to obtain an RGB image; converting the format of the RGB image to obtain a YUV image; performing image denoising on the Y channel of the YUV image to obtain Ynr; performing downsampling on the Bayer image to obtain four chrominance subgraphs R1, Gr1, Gb1 and B1; performing downsampling on the Ynr image to obtain four luminance subgraphs YR1, YGr1, YGb1 and YB1; performing guided filtering on the chrominance subgraphs by using the luminance subgraphs to obtain four chrominance filtered images R1nr, Gr1nr, Gb1nr and B1nr; performing guided filtering on each level of the luminance subgraphs and the chrominance subgraphs by using a Gaussian pyramid; and performing upsampling and weighted fusion on each level of the four chrominance filtered images and restoring the images to the original positions to output a Bayer denoised image. The application takes into account low-frequency and high-frequency signals, improves the image signal-to-noise ratio, and retains more image details.

This invention discloses a virtual immunohistochemical staining method and system based on relation generative adversarial networks (GANs) and sequence slide data, comprising the following steps: Step S1, constructing a sequence slide dataset, wherein each H&E-stained slide is followed by three consecutive Ki67-stained slides; Step S2, constructing a relation generative adversarial network (GAN) model, the model comprising two ResNet generators and two relation discriminators, used to train and evaluate the credibility of the sequence relationship between image pairs; Step S3, performing two-stage training on the relation generative adversarial network model, namely, basic style pre-training and Z-axis alignment relation adaptation. This invention fundamentally solves the Z-axis alignment problem, resulting in higher image fidelity and pathological accuracy, significantly improving clinical application value, and providing new research resources and evaluation paradigms.

This invention discloses a specular removal method based on grouped enhanced convolutional attention feature fusion, belonging to the field of image processing technology. The method constructs a specular removal model based on a CycleGAN network, which consists of a generator, a discriminator, and a loss function optimization module. The generator includes detail enhancement depth downsampling, upsampling, and grouped enhanced convolutional attention fusion modules, effectively capturing image details, reducing checkerboard artifacts, and enhancing feature fusion and capture capabilities. The discriminator uses a PatchGAN structure to improve local detail capture capabilities. The optimized loss function integrates multiple losses to improve model performance. Experimental results show that compared with various traditional and deep learning methods, the method of this invention outperforms in PSNR and SSIM metrics, exhibits strong adaptability, and can provide high-quality image data for subsequent detection of elevator door components.

The application discloses a rumor detection framework based on adaptive data enhancement and adversarial training, and aims to overcome the problems that the prior art only considers structure and ignores features when facing data noise problems, and does not consider the influence of adversarial attacks on rumor detection problems. The steps of the method are as follows: 1. processing the data set; 2. using an adaptive data enhancement method to weaken the influence of noise; 3. using a graph convolutional neural network to generate a graph representation; 4. using an adversarial generation module to generate an adversarial representation to improve the robustness of the model; and 5. constructing a joint loss function to optimize the rumor detection task.

The present application relates to a kind of based on carbonized wood natural structure dopamine electrochemical sensor and its preparation method and application, the working electrode of the electrochemical sensor is modified with carbonized wood-ZnO nanowire-ZIF-8.The electrochemical sensor of the present application uses carbonized wood as substrate, after cutting carbonized wood whole piece, using conductive carbon paste is pasted on working electrode, while ensuring conductivity, also retain the three-dimensional structure of nanomaterial not be destroyed, guarantee the performance of sensor, with good application prospect.

The present application relates to a kind of processing method for prescription data, electronic equipment and storage medium, it is related to data processing technical field, including the following steps, to electronic prescription data is carried out semantic entity extraction, obtain prescription sensitive data, and prescription sensitive data is carried out risk classification mark, obtain sensitive feature label;Prescription sensitive data is carried out random difference disturbance calculation based on sensitive feature label, obtain disturbance numerical sequence;Based on prescription sensitive data, disturbance numerical sequence is carried out feature cross and non-linear mapping, obtain confused prescription data;Confused prescription data is carried out node comparison based on pre-set drug knowledge graph, obtain graph matching node;Confused prescription data is carried out risk assessment and numerical update based on graph matching node, obtain target prescription data, it solves how to protect the technical problem of the premise of prescription data privacy, guarantee the availability of data to support risk assessment and knowledge graph association as main problem.

This invention relates to the field of polishing equipment technology, and discloses a multi-faceted polishing fixture for watch cases, including a processing box. A silicone mesh plate is located in the center of the inner side of the processing box, and a placement seat is provided on the silicone mesh plate. Multiple processing chambers for individually processing the watch case are equidistantly arranged on the placement seat. By adding and setting a front-end processing mechanism, which uses alkaline cleaning water immersion combined with ultrasonic vibration and a circulating filtration structure, various oil stains and impurities such as cutting fluid and release agents remaining on the surface and crevices of the watch case can be thoroughly removed, eliminating polishing slippage and surface dulling caused by oil residue. Secondly, relying on constant-temperature liquid immersion, the thin-walled watch case is heated evenly throughout, thoroughly and evenly releasing residual metal stress from previous processing, avoiding subsequent polishing deformation caused by uneven stress release. Simultaneously, it optimizes the polishing process, reduces grinding head vibration, and eliminates wavy lines and light / dark layer defects on the watch case surface.

The application discloses a preparation method of enteric sustained-release defensin mesoporous silica nanoparticles. The enteric sustained-release defensin mesoporous silica nanoparticles prepared by the method have good defensin loading capacity, good responsiveness to trypsin and good enteric targeting and sustained-release defensin performance.

The invention discloses a fiber ceramic-based catalytic filter element for denitration of industrial flue gas and a preparation method of the fiber ceramic-based catalytic filter element, and belongs to the technical field of catalytic filter elements. According to the preparation method, firstly, a chelating agent is introduced in the preparation process of TiO2 sol to improve the dispersity and particle size uniformity of TiO2 nanoparticles, the chelating agent can effectively adjust the hydrolysis rate of butyl titanate by means of the strong coordination acting force of the chelating agent and titanium ions, and high-dispersity and small-particle-size TiO2 sol is constructed; then, the fiber ceramic base material is immersed in the TiO2 sol, a firm TiO2 coating is formed after calcination, the firm TiO2 coating serves as a carrier with a high specific surface area, and the binding force between the firm TiO2 coating and the base material is enhanced; the preparation method comprises the following steps: firstly, preparing a TiO2 coating, then dipping a solution containing vanadium and tungsten precursors onto the composite carrier, and calcining to form a main active component V2O5 and a cocatalyst WO3 which are highly dispersed on the surface of the TiO2 coating to form selective catalytic reduction active sites, thereby finally obtaining the fiber ceramic-based catalytic filter element which is good in blocking resistance, water-resistant, sulfur-resistant and high-temperature-resistant and can be used for industrial flue gas denitration.

This utility model belongs to the field of UAV power motor technology, and discloses an external rotor structure, a brushless motor, and a UAV. The external rotor structure is composed of a shaft, a fan, a housing, a cage, and magnets. The cage is ring-shaped and fixed inside the housing, with multiple protrusions extending downwards on its lower side to hold the magnets in place. The fan has a top plate on its upper side and fan blades on its lower side. The top plate of the fan is fixed to the top surface of the housing, and the shaft is fixed inside the housing via a shaft mounting part. This external rotor structure achieves the installation of the shaft and magnets through the assembly structure of the fan, housing, and cage. Furthermore, the structure of the fan, housing, and cage is relatively simple, allowing the cage and fan to be manufactured using injection molding, and the housing to be manufactured using deep drawing, effectively reducing manufacturing costs and improving the production efficiency of the UAV. At the same time, the fan structure is retained to ensure the motor's heat dissipation performance.

The present application provides a kind of sparse gradient enhanced edge preserving image smoothing method, comprising: introducing structure fidelity term on the basis of existing global optimization model containing data fidelity term, in structure fidelity term, the gradient of input image is preprocessed using the way of combining mapping function and bilateral filtering;And introduce sparse gradient enhancement term on the basis of structure fidelity term, in sparse gradient enhancement term, L2 norm and L1 norm are used to constrain the gradient of output image to enhance sparsity, L1 norm is iteratively solved by combining subgradient method, fast Fourier transform and alternating direction multiplier method, or L p norm is used to constrain the gradient of output image, and L p Norm is converted to L1 norm for solving by iterative reweighting method, and preconditioned conjugate gradient method is used to improve calculation efficiency in solving process.The present application can retain complete semantic information while showing sparse smoothing effect when processing simple and complex non-texture images and texture images.

The application relates to the technical field of aqueous zinc ion batteries, and discloses a green preparation process of a positive electrode material of an aqueous zinc ion battery. The process realizes resource utilization by recovering a manganese source from waste batteries and extracting a vanadium source from vanadium slag. The process realizes carbon-coated, pre-intercalation and nitrogen-doped synergistic optimization by adopting a green complexing agent to replace a toxic chelating agent, and by integrating in-situ doping, carbon compounding and gelation procedures under low-temperature water bath conditions, combining ultrasonic-assisted low-temperature aging and 400-420 DEG C low-temperature carbonization, so that the total energy consumption of the preparation process is significantly reduced compared with a traditional high-temperature process. Finally, the finished product is obtained through ball milling mechanical activation and centrifugal washing, so that the whole process is free of strong acid, strong alkali and toxic waste liquid, the yield is greatly increased, and the positive electrode material prepared through the process has a hierarchical porous structure, a high specific surface area and a nanoscale particle size.

The invention discloses a gamma-ray energy spectrum data processing method and device, equipment and a medium, and belongs to the technical field of element analysis. The method comprises the following steps: acquiring original energy spectrum data; according to the channel address information of each data in the original energy spectrum data, carrying out dimension expansion on the original energy spectrum data to obtain two-dimensional energy spectrum data; performing position compensation on the two-dimensional energy spectrum data according to the position information of each element in the two-dimensional energy spectrum data to obtain target energy spectrum data; and processing the target energy spectrum data through a self-attention network model to obtain an element analysis result of the tested substance. According to the target energy spectrum data in the scheme, the structure of the energy spectrum data can be effectively reserved, data information is enriched, the influence of spectrum drift on system analysis is reduced, and the accuracy of gamma-ray energy spectrum data analysis is improved.

The present application relates to a kind of flame and smoke detection method based on improved YOLO11, comprising: the image of target to be detected is collected, the image of target to be detected is input into improved YOLO11 model, and detection result is obtained;The improved YOLO11 model is obtained using training set training;The training set includes: original target image and flame and smoke label;The improved YOLO11 model includes: replace the conventional down-sampling layer in backbone network by detail-context collaborative down-sampling module, use dynamic double-feature fusion module to reconstruct feature interaction unit in feature fusion network, and use shared regression and asymmetric classification module to replace original unified classification detection head.The present application improves the performance of flame and smoke detection under complex environment.

The invention discloses a springback type composite sheet and a preparation method thereof, and belongs to the technical field of composite sheets. The sheet comprises an inner layer, a springback layer, a blocking layer and an outer layer which are sequentially arranged in a stacked mode. The base material of the rebound layer is polyester or polyamide, the tensile strength of the rebound layer is limited to be larger than or equal to 150 MPa, the elongation at break of the rebound layer is larger than or equal to 50%, the thickness of the rebound layer is 10-30 microns, meanwhile, the melt flow rate of the polyester under the condition of 275 DEG C / 2.16 kg is controlled to be 5-20 g / 10 min, and / or the melt flow rate of the polyamide under the condition of 235 DEG C / 2.16 kg is controlled to be 5-18 g / 10 min. The springback layer further comprises functional auxiliaries such as maleic anhydride grafted polyethylene and oxazoline grafted polyolefin, and polyester or polyamide is preferably subjected to two-way stretching treatment. The preparation method comprises the following steps: performing corona treatment on each layer of substrate film, sequentially compounding each layer at different temperatures and pressures by adopting a step-by-step dry-method compounding process, and finally performing curing treatment. The method is used for improving the rebound resilience of the composite sheet packaging.

The application discloses the technical field of high polymer synthesis and functional material modification, and relates to a preparation method of low-molecular-weight polyacrylamide. The method comprises the following steps: firstly, preparing a double-metal doped multi-level pore carbon nanotube-covalent organic framework composite modification material, and the preparation process comprises the following steps: surface functionalization of multi-walled carbon nanotubes, growth of a covalent organic framework layer, doping of double-metal ions, and subsequent reduction and calcination treatment; then, adding deionized water, acrylamide and the modification material into a three-necked flask, stirring under nitrogen to dissolve the acrylamide, preparing an oxidation-reduction initiator solution, and dropping the initiator solution into the system under the protection of nitrogen, and polymerization reaction is carried out in a constant-temperature water bath at 19-21 DEG C. After the polymerization is completed, the product system is washed, filtered, and finally, low-molecular-weight polyacrylamide is obtained by using a vacuum freeze-drying method. The polyacrylamide prepared by the method has low and controllable molecular weight, and has good solubility and salt resistance.

The invention discloses a single-cell multi-omics dimension reduction method based on a Gaussian process hidden variable model, and aims to solve the problems that single-omics and multi-omics data analysis requirements are difficult to consider and high sparsity and technical noise of single-cell data cannot be effectively handled in the prior art. According to the method, a shared potential space probability generation framework is constructed, variational inference and a sparse Gaussian regression strategy are combined, data of different modalities are non-linearly mapped to a unified low-dimensional space, and meanwhile, an adjustable modal weight is introduced, so that the model can perform efficient dimensionality reduction on single-cell transcriptome and multi-omics data. When the method is used for carrying out dimension reduction on single cell sequencing data, low-dimensional representation learned by the model can be used for completing downstream tasks such as cell clustering, visualization and differential expression analysis, cell heterogeneity and potential biological regulation signals are accurately revealed, and the method has a good application prospect.

This invention discloses an enhanced antistatic polyetheretherketone (PEEK) composite material and its preparation method. The composite material is prepared from the following raw materials in parts by weight: 60-85 parts of PEEK resin, 5-15 parts of a first synergistic component, 3-10 parts of a second synergistic component, 2-8 parts of a third synergistic component, and 1-5 parts of a high-temperature processing aid. The first synergistic component is a mixture of a metal-based conductive filler and a high-temperature resistant antistatic synergist; the second synergistic component is ionic liquid-functionalized modified graphene; the third synergistic component is a composite of carbon nanotubes and conductive nanofibers; and the high-temperature processing aid includes a high-temperature lubricant and a high-temperature antioxidant. The preparation method includes steps such as preparing various premixes, melt blending, extrusion granulation, and supercritical fluid in-mold foaming. This invention constructs a stable three-dimensional conductive network through the synergistic effect of multi-scale conductive fillers, giving the composite material excellent antistatic properties.

The invention belongs to the technical field of exosome extraction equipment, and discloses an exosome extraction device which comprises a fixing base, a filtering assembly, an extraction assembly and a heating assembly, and the heating assembly comprises a heating base, a spring, a resistance sliding block and a resistance base body; the extraction assembly comprises an extraction cylinder and a collection box, and a sand core and gel filler are arranged on the inner side of the extraction cylinder; the filter assembly comprises a connecting seat, a filter cartridge, a vibration unit and a back flushing unit, the filter cartridge is connected with a filter screen and a filter membrane, the vibration unit comprises a first motor, a first cam and a first sliding block, and the back flushing unit comprises a second motor, a liquid drawing cartridge, a second cam, a second sliding block, a piston rod group, a liquid inlet pipe, a liquid outlet pipe, a liquid guide pipe and a flushing nozzle. The biological body fluid is filtered through the filtering assembly, the exosome is extracted through the extraction assembly, the heating assembly is used for heating the extraction assembly, the exosome extraction efficiency is high, and the activity of the extracted exosome is good.

The invention discloses a digital holographic microscopic imaging coherent noise suppression network model and method based on deep learning. The network model adopts a double-branch encoder-single decoder structure, and a double-branch encoder extracts local details and global noise features of an image in parallel; the innovative double-branch intensive attention fusion module fuses multi-scale features through an adaptive weighting and enhancement mechanism; and the decoder integrates information through jump connection and reconstructs a clear image. The training method generates a high-fidelity data set based on a physical simulation system. According to the method, speckle noise and parasitic fringes can be efficiently suppressed, the phase details of the object are effectively reserved while the peak signal-to-noise ratio and the structural similarity index are remarkably improved, and the method has excellent generalization ability and calculation efficiency and is suitable for real-time high-quality imaging of a digital holographic microscopy system.

This invention discloses a super-resolution method for heart sound signals based on frequency band features and multi-timescale modeling. The method includes: first, acquiring a heart sound image signal corresponding to cardiac motion using defocus speckle imaging technology; mapping the heart sound image signal to a high-dimensional feature space to obtain intermediate feature representations and dividing them into low-frequency structural components and high-frequency detail components; extracting residuals from the low-frequency structural components and high-frequency detail components to obtain local feature representations and enhancing the frequency components; capturing the local correlation features of heart sound events within a single cardiac cycle and the temporal correlation information between different cycles using hierarchical temporal modeling; and reconstructing high-resolution heart sound audio based on the encoded local correlation features. This invention can capture both local feature changes within a cycle and characterize global dependencies across cycles, thus being more suitable for the temporal characteristics of heart sound signals and helping to improve the stability and accuracy of super-resolution reconstruction.

The invention belongs to the field of medical application of birch juice, and discloses application of a birch juice separating medium to improvement of parenchymal hepatic cell injury, the birch juice separating medium is used for improving parenchymal hepatic cell injury, and the birch juice separating medium is prepared according to the following steps: S1, raw material pretreatment; s2, preliminary filtration; s3, performing ultrafiltration; s4, performing nanofiltration concentration; s5, chromatographic purification; s6, vacuum drying; according to the technical scheme, the birch juice separation liquid is applied to improvement of the parenchymal hepatic cell injury, the parenchymal hepatic cell injury is improved through the birch juice separation liquid, the application field of the birch juice is expanded, and experiments show that the birch juice separation liquid prepared through the technical scheme can remarkably reduce transaminase release of the injured parenchymal hepatic cells, so that the survival rate of the injured parenchymal hepatic cells is increased. The traditional Chinese medicine composition has the advantages that the activity of antioxidant enzymes is improved, the content of lipid peroxidation products is reduced, and the traditional Chinese medicine composition has a good improvement effect on parenchymal hepatic cell injury induced by various factors and provides a new natural active substance source for preventing and treating liver injury.

This invention discloses a method, apparatus, and imaging device for correcting metal artifacts using nonlinear projection decomposition. The method includes: Step 1, acquiring original CT scan data containing artifacts to obtain an original CT image with metal artifacts; Step 2, thresholding the CT image with metal artifacts to separate the metal from the reconstructed image, and then obtaining the metal projection region; Step 3, using an interpolation method to complete the metal projection region and outputting an interpolated corrected projection; Step 4, combining the interpolated corrected projection obtained in Step 3 to obtain the projection of the metal portion in the CT scan data, then obtaining fitted metal projection data using a rational fraction fitting method, and finally subtracting the contribution of the metal projection portion from the CT scan data to obtain the corrected projection; Step 5, reconstructing the projection data after subtracting the metal contribution, and fusing it with the metal region obtained in Step 2 to obtain a CT image corrected for metal artifacts. This invention can effectively preserve details and structural information near metal implants while removing metal artifacts.

This application discloses a composite high-flowability, high-conductivity conductive masterbatch and its preparation method, relating to the technical field of conductive materials. The masterbatch uses a fiber-forming polymer as a matrix, compounded with conductive carbon black, graphene, silver nanosheets, carbon nanotubes, and modified MXene nanosheets, along with coupling agents, dispersants, antioxidants, and lubricants. The modified MXene nanosheets are modified with an amino-functionalized modifier and a polysaccharide mixture of sodium alginate and agar. Amino grafting improves interfacial compatibility, while polysaccharide coating inhibits filler agglomeration and oxidation, optimizing the dispersion of the conductive filler in the matrix. This masterbatch constructs a multi-level continuous conductive network, significantly improving melt flowability, conductive stability, and mechanical properties, effectively solving the problems of easy agglomeration, poor flowability, and insufficient spinnability of traditional conductive masterbatches.

This invention proposes a fuzzing method and system for WebAssembly engines, relating to the field of software testing. It involves acquiring a large number of WebAssembly binary samples as initial seed samples; parsing the initial seed samples into an attribute abstract syntax tree (API) with node position and type information; applying a mutation method to the API using a type-aware mutation strategy and a structure-aware instruction mutation method to generate a new API with valid structure and type; pre-validating the mutated API and quickly correcting invalid node positions; and generating WebAssembly binary samples from the pre-validated API, which are then input into the target engine for fuzzing. This invention, by using an API based on attribute abstract syntax trees and a WebAssembly mutation method with pre-validation and rapid correction, more effectively generates WebAssembly module mutations with correct binary format, legality, and compliance with security constraints, enabling efficient fuzzing of the WebAssembly engine.

The invention relates to a method for decolorizing waste cotton-polyester blended fabric through low-temperature plasma synergistic chemical oxidation. Through the synergistic effect of low-temperature plasma and an acidic chemical oxidant (composed of inorganic acid, hydrogen peroxide, a penetrant and a hydrogen peroxide stabilizer), color development and auxochrome groups of dye on the waste polyester-cotton blended fabric are effectively destroyed, and efficient decolorization of the waste polyester-cotton blended fabric is achieved. According to the technical scheme, the decolorizing efficiency is improved, the original performance of polyester and cotton fibers can be reserved, the secondary utilization performance of the recycled waste polyester-cotton mixed textile fabric is guaranteed, and an effective means is provided for resource utilization of the waste polyester-cotton mixed textile fabric. Compared with a traditional chemical oxidation technology, the method has the advantages that the required acid amount and treatment time are remarkably reduced, the temperature required by the reaction is greatly reduced, the environmental protection performance is remarkably improved, the energy consumption is reduced, and an economic and green new way is provided for recycling the waste polyester-cotton mixed textiles.

The application provides a ZSM-5 molecular sieve, a preparation method and application thereof, a hydroprocessing catalyst, a hydrodewaxing catalyst and application thereof. The pyridine infrared total acid amount of the ZSM-5 molecular sieve is 0.03-0.40 mmol / g, the di-tert-butyl pyridine infrared total acid amount is 0.002-0.02 mmol / g; the mesopore volume of the ZSM-5 molecular sieve accounts for 10-20% of the total pore volume, and / or the mesopore volume of 2-10 nm in the ZSM-5 molecular sieve accounts for 70-95% of the total mesopore volume. The molecular sieve has obvious technical advantages as a carrier or an active component, for example, the hydrodewaxing catalyst prepared from the ZSM-5 molecular sieve can be used for oil product treatment to simultaneously improve the quality and yield of low-condensation oil products.

The application discloses a reaction device and method for alkaline oxidation continuous step-by-step extraction of lignin, and relates to the technical field of high-efficiency utilization of biomass. Different lignin products in different extraction stages are collected respectively, and serious secondary degradation reaction of the lignin products is avoided. The reaction device comprises an oxidizing alkali solution preparation system, an oxidation extraction system and a product collection system. The oxidizing alkali solution preparation system comprises a high-pressure gas cylinder, a heatable pressure-resistant container, a heatable magnetic stirrer and a gas supply pressure regulating system. The oxidation extraction system comprises a rapid heating box, a heat radiation preheater arranged in the rapid heating box and a flow-through reaction tube. The product collection system comprises an equilocal pump and a sample collection bottle. In the application, the biomass lignin can be continuously extracted and collected in different alkaline oxidation solvents, different time gradients and different temperature gradients, and the secondary reaction of the extracted lignin in the reaction system is significantly weakened.