485 results about "Reflection spectrum" patented technology

The invention relates to the technical field of medicine quality detection, and particularly discloses an optical detection method and system for the content of vitamin tablets. The method comprises the following steps: driving an optical fiber probe array to carry out three-dimensional multi-angle near-infrared scanning through a multi-axis mechanical arm, and collecting reflection spectrum data; a weighted spectrum is generated through derivative spectrum analysis and double evaluation, and auxiliary material noise is corrected and separated in combination with subspace projection and a dynamic kernel function; performing spectral vector projection and regression coefficient iterative optimization through an online PLS model, and synthesizing an error coefficient based on a four-level index retrieval standard library to perform dual-channel feedback; and finally, fusing the credibility weight to output a detection result of the binding confidence. The method realizes nondestructive and high-precision detection of the vitamin tablets, has the beneficial effects of multi-dimensional data fusion, dynamic error compensation and high model adaptability, and remarkably improves the detection accuracy and reliability.

The invention relates to the related technical field of semiconductor wafers, in particular to a full-automatic wafer film thickness detection method and system based on spectral reflection, and the method comprises the steps: obtaining a wafer surface reflection spectral signal, and extracting spectral feature information and wavelength parameters; identifying a polarization characteristic parameter and an interface interference characteristic parameter, introducing the wavelength parameter as an adjusting variable, determining a wavelength dependence term, and generating a global film thickness distribution map, thereby solving the problems that in a detection scene with a multi-layer film or a rough interface, the influence of an incident light polarization state on a reflection signal is not fully considered, the reflection signal characteristic extraction is distorted, and the detection accuracy is poor. According to the method, the technical problem that the film thickness obtained through reflection characteristic backstepping deviates from the actual film thickness is solved, polarization characteristic parameters and interface interference characteristic parameters are recognized, the thickness of each detection point is determined after the film refractive index is corrected, the thickness detection precision is effectively improved, and the influence of spectral signal noise and interface interference is reduced; the technical effect of accurately presenting the global film thickness distribution of the wafer film is achieved.

The invention provides a wavelength deviation compensation method and system for rapid wavelength analysis of a far-end fiber bragg grating, and relates to the technical field of fiber bragg grating sensing, and the method comprises the steps: collecting grating reflection spectrum signals of different transmission distances, carrying out frequency domain transformation to separate phase components, and building a mapping relation between the transmission distance and dispersion cumulant; calculating an initial wavelength offset estimation value; identifying a refractive index section, and establishing a segmentation compensation coefficient; constructing a wavelength-strain-temperature three-dimensional space, and optimizing a compensation coefficient according to the trajectory curvature; and updating the compensation coefficient according to the residual gradient. The wavelength analysis precision of the far-end fiber bragg grating can be improved, and the influence of dispersion on measurement is reduced.

The invention relates to an infectious bacterium detection method and system based on double effects of a reflection spectrum and autofluorescence, and belongs to the field of biomedical detection. The method comprises the following steps: acquiring hyperspectral images of different samples under different light sources; performing data preprocessing on the collected sample hyperspectral image, realizing pixel-level alignment of the dual-effect image based on feature matching and image registration, and constructing a data set; the method comprises the following steps: constructing a wound infection bacterium detection model based on double effects of a reflection spectrum and autofluorescence, and comprising a double-branch feature extraction module which comprises a reflection branch and a fluorescence branch which are respectively used for extracting reflection hyperspectral image features and fluorescence hyperspectral image features; the redundant feature screening module screens the features extracted by the reflection branch and the fluorescence branch through a channel attention mechanism; and the feature fusion module adopts a cross-branch cross attention mechanism to fuse the features of the double branches. According to the method, learning is carried out from two physicochemical characteristics, and the effectiveness and adaptability of the technology are improved.

The invention relates to a data analysis technology, and discloses a spectral feature and artificial intelligence fused grain heavy metal detection method and system.The method comprises the steps that a near-infrared diffuse reflection spectrum, a laser-induced breakdown spectrum and a surface enhanced Raman spectrum of a grain sample are obtained, a matrix accumulation area thermodynamic diagram is constructed according to the near-infrared diffuse reflection spectrum, and the matrix accumulation area thermodynamic diagram is obtained; the method comprises the following steps: optimizing a laser-induced breakdown spectrum based on a thermodynamic diagram of a matrix accumulation area, carrying out dual-channel spatial-temporal feature extraction on a surface enhanced Raman spectrum and the optimized laser-induced breakdown spectrum to obtain a fusion feature vector, carrying out weighted fusion optimization on the fusion feature vector based on a signal-to-noise ratio of multispectral data to obtain an optimized fusion feature vector, and finally obtaining a fusion feature vector. And carrying out migration adaptation on a pre-trained grain detection model by using a pre-marked new grain sample, and carrying out heavy metal content identification on the grain sample by using the adaptive grain detection model according to the optimized fusion feature vector to obtain a heavy metal detection result. The precision of grain heavy metal detection can be improved.

The invention relates to the technical field of textile detection, in particular to a computer-aided textile color fastness prediction method, which comprises the following steps: acquiring multi-source data before and after a textile color fastness test through computer control equipment, including digital images, reflection spectrum data, process parameters and dye characteristic component data, carrying out pretreatment; extracting color difference features and spectrum similarity features, and screening by adopting a dimension reduction algorithm to form a core feature vector; a machine learning algorithm is adopted to construct a color fastness prediction model, parameters are optimized in combination with cross validation and an early stop mechanism, model fine tuning is executed, and a final optimization model is obtained; and outputting a color fastness grade prediction result, and carrying out spectrum similarity secondary verification. According to the method, objective, efficient and high-precision prediction of the color fastness grade of the textile is realized, and the problems of high subjectivity and low efficiency of traditional manual evaluation are solved.

The invention discloses a degraded cultivated land soil organic matter three-dimensional mapping method fusing remote sensing, near-earth sensing and historical data, and relates to the technical field of digital soil mapping, the method is based on an INLA-SPDE three-dimensional model framework, remote sensing, near-earth sensing and historical data are fused, and high-precision three-dimensional mapping is achieved. The method comprises the following steps: collecting organic matter data and near-sensing visible near-infrared reflection spectrum data of different depths of a soil profile at each sampling point in a research area, then screening visible near-infrared reflection spectrum characteristic wavebands significantly related to organic matters, collecting and integrating historical data, and completing standardization processing on remote sensing and near-earth sensing data; dividing a modeling set and a verification set, calculating a range value based on a semi-variable function, determining triangular mesh parameters, and constructing an SPDE model; the near sensing data and the remote sensing data serve as covariables and are fused into the INLA-SPDE three-dimensional model, and a soil organic matter three-dimensional space distribution diagram is generated after model fitting and uncertainty is quantified. According to the method, the problem that the vertical change of the soil attribute cannot be accurately simulated by a traditional method is solved, and high-precision and high-efficiency three-dimensional mapping of the regional scale soil organic matter in the horizontal and vertical dimensions is realized.

The invention provides a semiconductor epitaxial growth control method and system, a control device and a related product, and relates to the technical field of semiconductor detection.The semiconductor epitaxial growth control method comprises the steps that reflection spectrum data is obtained in real time and analyzed to obtain an optical feature set, then the optical feature set is input into an optical transmission model for iterative optimization, and a predicted growth rate is obtained; and comparing the predicted growth rate with the target growth rate, generating deviation information, and outputting a process adjustment control instruction, so that the main control system performs closed-loop correction on the process parameters in the next process period. According to the invention, the optical feature set can dynamically represent the influence of the interface transition layer on the reflection spectrum, so that the inversion precision and robustness of the real epitaxial structure state are improved; closed-loop correction is performed on the process parameters based on the predicted growth rate, so that the growth rate drift can be inhibited in time, the process disturbance can be adaptively compensated, the manual calibration and intervention requirements are reduced, and the production stability, repeatability and batch-to-batch consistency are improved.

The invention discloses a chicken manure crude protein content detection model construction method, a chicken manure crude protein content detection system and a detection method, relates to the technical field of bioinformatics, solves the problems of high cost, serious pollution and low efficiency of an existing chicken manure crude protein content detection technology, and improves the accuracy of chicken manure resource application. The obtained sample set comprises visible-near infrared reflection spectrum and crude protein content of the chicken manure; carrying out pretreatment on the visible-near infrared reflection spectrum by adopting a plurality of pretreatment methods; screening out an optimal preprocessing method and a corresponding preprocessing spectrum by using a partial least square regression model; obtaining a characteristic wave band by using a plurality of characteristic selection methods; and based on the crude protein content and the characteristic wave band, constructing and screening an optimal chicken manure crude protein content detection model. The method provided by the invention is suitable for detecting the content of crude proteins in chicken manure by using manure resources.

The invention discloses a chicken manure crude protein content detection model construction method, a chicken manure crude protein content detection system and a detection method, relates to the technical field of bioinformatics, and solves the problems that an existing chicken manure crude protein content detection technology is serious in pollution in the detection process, and a spectral analysis technology is serious in information redundancy during detection, high in cost and poor in model stability. The sample set comprises visible-near infrared reflection spectrum and crude protein content of chicken manure; obtaining three spectral indexes corresponding to the visible-near infrared reflection spectrum, the MSC pretreatment spectrum and the SNV pretreatment spectrum, and screening out the optimal wavelength position and the corresponding three spectral indexes; constructing a data set by using the three spectral indexes and the crude protein content, and performing fitting through multiple fitting modes; evaluating the plurality of fitting equations, and screening out an optimal fitting equation and a corresponding pretreatment method and wavelength position as a chicken manure crude protein content detection model. The method provided by the invention is suitable for detecting the content of crude proteins in chicken manure by using manure resources.

The invention provides a multi-sensor fusion crop harvesting and pile-dividing cooperative control method and a multi-sensor fusion crop harvesting and pile-dividing cooperative control system. The method comprises the following steps: extracting morphological characteristics of crops from crop image data, and generating a crop maturity identification result according to a difference value between reflection spectrum data of two groups of wavebands; determining a crop quality identification result according to the morphological characteristics and the crop maturity identification result; an opening and closing control instruction is sent to a pneumatic flow guide plate at a pile separation opening corresponding to the crop quality recognition result in a multi-stage pile separation opening in the tail end of the conveying belt, so that the crops on the conveying belt are conveyed to a corresponding pile separation area; and when the pressure value of the pile dividing area is larger than a preset pressure threshold value, the harvesting speed of the crop harvesting and pile dividing all-in-one machine is adjusted according to the pressure value. According to the technical scheme provided by the invention, the shape and maturity of the crops are analyzed through the image and spectral data, piling and harvesting can be intelligently controlled, and accurate and efficient crop quality grading and harvesting management are realized.

The invention relates to a red light therapeutic instrument real-time calibration method based on multi-modal data fusion, which comprises the following steps: firstly, collecting data such as optical power, target surface temperature, internal temperature, reflection spectrum and light source-target surface distance in a time window, and unifying time base alignment; performing anomaly elimination, de-noising and normalization on each modal data, and constructing a standardized data matrix; extracting multi-modal features through power spectral density, wavelet packet decomposition, exponential moving average and Kalman filtering, inputting the multi-modal features into a fusion network containing a modal special encoder and a cross-modal attention unit, and generating a red light treatment state vector; the driving current, the pulse width, the duty ratio, the emission angle and the aging compensation factor are solved in real time through constrained least square optimization and are issued to the driving control module, and the red light source is smoothly adjusted in a soft start mode; the system monitors the output power, calibration is completed if the output power meets a threshold value, and otherwise, the next iteration is started. According to the method, power closed-loop correction can be realized, and dose consistency and thermal safety are improved.

The invention provides an all-dielectric film system structure with visible near-infrared low transmission vegetation simulation, intermediate infrared high reflection and microwave high transmission. The device sequentially comprises, from top to bottom, a visible near-infrared simulated vegetation layer which is composed of first to fourth dielectric layers and is configured to simulate vegetation reflection spectrum characteristics in a wave band of 380-2500 nm and has a low transmission characteristic; the intermediate infrared high-reflection layer is formed by alternately stacking a first dielectric layer and a third dielectric layer, and is configured to form high reflection in a wave band of 3-14 microns; and a substrate layer; the thickness of the whole film layer is far smaller than the microwave wavelength, so that the film system keeps high transmission in the microwave band. According to the invention, the amorphous InSb material with high loss characteristic in visible near-infrared light is introduced, so that the average transmissivity of 1500-2500 nm is reduced to 0.095 while high-precision simulation of the vegetation reflectivity spectrum is realized, thereby realizing simulation of low transmissivity of the wave band vegetation.

The invention discloses a comprehensive evaluation method for drought resistance of corn varieties, which relates to the technical field of corn variety evaluation, and comprises the following steps: obtaining multi-temporal multi-band reflection spectrum data under the condition of continuous drought of corn, and performing structure coupling modeling on the spectrum data in combination with the structural characteristics of corn leaf epidermis wax, so as to obtain the drought resistance of the corn varieties; constructing a time sequence curve capable of representing the dynamic change of leaf wax distribution as a dynamic spectrum monitoring baseline; differential analysis between continuous time points is carried out based on the dynamic spectrum monitoring base line, spectrum abrupt change points caused by leaf wax migration are identified, and space consistency inspection is combined. According to the method, a dynamic spectrum monitoring baseline and an energy migration fingerprint are constructed, wax migration interference is accurately identified, a spectrum stability hypothesis is broken through, high precision, self-adaption and consistency of evapotranspiration rate inversion under a drought condition are realized through multi-level correction and a closed-loop updating strategy, and scientificity and comparability of drought resistance evaluation are improved.

The invention relates to the technical field of material detection, in particular to a cross-linked polyethylene insulation material impurity full-detection device and system for a high-voltage cable and application, the detection device carries out detection based on single-row continuous flowing insulation material particles, and the detection device comprises a near infrared spectrum imaging unit, a visible light imaging unit and an image processor, and identifying black spot impurities and heterochromatic particles. The full-detection system provides a detection channel for installing full-detection equipment and comprises a current limiting pipe body and a camera obscura structure, and reflection spectrum and surface color information are collected in the camera obscura structure; the full-inspection system further comprises a buffer stock bin, a finished product stock bin, a pneumatic sorting module, a cyclone separator, a cooling fluidized bed and a metal separator. Impurity identification reliability is guaranteed through double optical detection, full-process online detection of insulation material impurities can be achieved, limitation of traditional sampling detection is broken through, full-detection coverage of insulation material particles is achieved, the impurity missing detection risk is eliminated, and the strict requirement of an ultrahigh-voltage cable for insulation material cleanliness is fundamentally met.

The embodiment of the invention provides a dynamic multifunctional optical metasurface design method and system, and the method comprises the steps: obtaining function demands in different application scenes, and setting a target reflection spectrum according to the function demands; the target reflection spectrum is input into a reverse retrieval network, and super-unit structure prediction design parameters are obtained through encoder compression and decoder parameter prediction; the super-unit structure prediction design parameters are input into a forward prediction network, and a prediction reflection spectrum is obtained through network forward propagation calculation; based on the matching degree between the predicted reflection spectrum and the target reflection spectrum, establishing a bidirectional relation between a design domain and a physical domain by using a phase recovery algorithm, and predicting design parameters by continuously adjusting the super-unit structure, so that the predicted reflection spectrum gradually approaches the target reflection spectrum, and obtaining ideal design parameters of the super-unit structure; and according to ideal design parameters of the super-unit structure, super-units are arranged in an operation space, and the super-surface structure capable of realizing dynamic multifunctional regulation and control is formed.

The invention relates to the technical field of laser cutting, in particular to a laser control method and system of laser cutting equipment. The method comprises the following steps: carrying out reflection spectrum detection and feature extraction on the surface of a to-be-processed material to obtain a surface spectrum reflection feature parameter; performing state classification on the surface of the to-be-processed material based on the surface spectral reflection characteristic parameters to obtain surface state classification data; performing fusion risk analysis on the to-be-processed material according to the surface state classification data to obtain a material fusion risk assessment result; generating a laser power control signal according to a material melting risk assessment result; and performing power-plasma closed-loop monitoring and state analysis on the laser-material interaction process based on a laser power control signal to obtain a material overheating sign recognition result. The problems of machining instability and defects caused by dynamic changes of material surface characteristics in laser cutting are solved.

The invention discloses a PVC film coating uniformity detection method based on spectral analysis, and relates to the technical field of optical detection, and the method comprises the following steps: S001, adopting a controllable incident illumination system with a multi-angle rotation function to carry out continuous angle scanning on the surface of a PVC film, collecting reflection spectrum intensity data in a corresponding wave band range according to different incident angles, and generating a spectrum data matrix containing interference characteristics; and S002, carrying out continuous wavelet transform processing on the spectral data matrix, carrying out multi-scale local decomposition on a spectral response signal, extracting and separating a periodic interference characteristic signal, and obtaining a purified spectral matrix after interference reduction. According to the method, through multi-angle spectrum collection and interference feature stripping, accurate identification of a coating thickness abnormal area and combination of strength indexes, spraying parameters are regulated and controlled, a detection-compensation-reinspection closed-loop mechanism is constructed, interference misjudgment is effectively avoided, and the accuracy and regulation and control efficiency of uniformity detection of the PVC film coating are improved.

The invention discloses a water quality monitoring device and a monitoring method thereof, belongs to the technical field of water quality monitoring, and solves the problems that trace hydrogen sulfide generated by sulfate reducing bacteria in a deposited layer at the bottom of a water tank which is not cleaned for a long time in a low-oxygen environment reacts with a platinum membrane electrode of a residual chlorine sensor to generate a platinum sulfide covering layer; and the electrode reading is continuously virtually high and drifts slowly. Comprising an optical transmitting module, an optical receiving module, a spectrum separation module, a signal conditioning module, a lock-in amplification module, a spectrum feature extraction module and a hydrogen sulfide recognition engine. Residual chlorine is monitored in a non-contact mode, deep ultraviolet pulse light is emitted to penetrate through a water body, a reflection spectrum is received, second derivative spectral characteristics are extracted through multichannel signal processing, and multivariable regression compensation is executed to calculate a real residual chlorine value after hydrogen sulfide interference characteristics are recognized. Meanwhile, the drift trend is quantified based on the double-time-constant integral, graded early warning is triggered, industrial standard signals are finally output, and the problem of electrode contact pollution is avoided in the whole process.

The invention provides a semiconductor epitaxial growth parameter inversion method, a semiconductor epitaxial growth parameter inversion system, a semiconductor epitaxial growth parameter detection device and a related product, and relates to the technical field of semiconductor detection. The gas phase transmittance of a detection light path is further obtained, quantitative compensation is carried out on the actually measured reflection spectrum in the semiconductor epitaxial growth process, the compensated target reflection spectrum is closer to the real optical response of a semiconductor epitaxial film, and therefore spectral shape distortion and system errors caused by gas phase absorption superposition are effectively weakened. And the accuracy and the stability of the inverted growth parameters are obviously improved. Furthermore, the quantitative compensation value can be adaptively updated according to the fluctuation of the working condition, the inversion deviation introduced by the drift of the working condition is reduced, the consistency and comparability of the process are improved, and more reliable end point control and parameter closed-loop optimization can be conveniently realized.

The invention relates to the technical field of color detection, in particular to a color difference calculation optimization method and system for archaized linen viscose reactive dyeing. Comprising the following steps: monitoring a fabric motion state in real time, calculating a motion speed and displacement, and determining a spectrum acquisition opportunity and a position compensation parameter; illuminating the surface of the to-be-detected fabric at an acquisition position through a polarized light source, and performing polarized spectrum data acquisition in the normal direction of the surface of the fabric to obtain reflection spectrum information; exciting the surface of the to-be-detected fabric by adopting an ultraviolet light source, collecting fluorescence spectrum data, and identifying different fiber areas; calculating fiber density distribution in the fiber area, and constructing a layered weight system; calculating chromatic aberration of the fiber area based on the reflection spectrum information, extracting chromatic aberration characteristics, and measuring and calculating an antique degree quantification result; setting a color difference threshold value standard according to an antique degree quantification result, and adjusting a color difference threshold value according to an evolution trend; and based on the layered weight system and the chromatic aberration threshold, performing weighted calculation on the chromatic aberration value of the fiber region to obtain a comprehensive chromatic aberration evaluation result.

The invention provides a uniform light mixing algorithm for an omni-directional light-emitting lamp. The uniform light mixing algorithm comprises the steps of obtaining a reflection spectrum of a target object, dynamically calculating a target SPD based on the data, driving a multi-channel LED to synthesize output, and achieving closed-loop control through continuous spectrum monitoring. According to the algorithm, a perception-decision-execution-feedback loop is constructed according to a visual effect of an optimization target after object-light interaction from a light-emitting surface static parameter to match the environment and requirements in real time, and the problems of color development distortion and insufficient comfort of static light mixing in a dynamic scene are solved.

The invention discloses a physicochemical index detection method for canned food production, and relates to the technical field of food detection, and the method comprises the following steps: deploying annularly arranged near infrared spectrum, eddy current electromagnetic and biological impedance three-mode sensing units, and collecting tank reflection spectrum, eddy current loss and complex impedance response signals in a non-contact manner; extracting characteristic parameters of sulfur dioxide, heavy metals and microorganisms through wavelet packet decomposition, Hilbert transform and principal component analysis; constructing a ternary coupling correction equation set to perform cross correction on the original data, and outputting an accurate detection result; and a self-calibration mechanism and an abnormal product automatic sorting function are integrated. By constructing four core technology systems of multi-modal synchronous acquisition, cross-modal feature coupling, dynamic threshold decision and online self-calibration, lossless, efficient and multi-index parallel detection is realized, the precision is stable, the cost is reduced by 60%, and the real-time quality control and tracing requirements of a full-automatic production line are met.

The invention relates to the technical field of semiconductor detection, in particular to a method and a system for measuring the thickness of a semiconductor epitaxial layer based on multi-beam interference correction. Acquiring infrared reflection spectrum data of the wafer under different incident angles, and performing double-stage smoothing preprocessing on the data by using a moving least square fitting algorithm; an interference extreme point is accurately identified by adopting a self-adaptive extreme value feature extraction strategy; the method comprises the following steps of: calculating the dynamic refractive index and the initial thickness estimation value of the epitaxial layer, constructing a multi-beam interference intensity distribution model, and finally providing a hybrid nonlinear optimization algorithm to invert the accurate thickness d of the epitaxial layer, thereby overcoming the measurement error caused by neglecting the multi-beam effect and the refractive index change in the traditional double-beam interference, and improving the accuracy of the multi-beam interference. And the precision and the reliability of epitaxial layer thickness measurement are remarkably improved through the combination of data preprocessing and a nonlinear optimization algorithm. The objective of the invention is to solve the problem of how to correct measurement errors caused by multi-beam effect and refractive index change.

A thermal measurement system enables high-resolution sub-surface temperature monitoring during additive manufacturing processes through machine learning demodulation of chirped fiber Bragg grating (C-FBG) sensors. An optical sensing subsystem includes a C-FBG sensor that encodes spatial temperature information in wavelength for high-temperature operation. A neural network model transforms complex reflection spectra into spatial temperature profiles with micrometer-scale resolution, overcoming limitations of traditional demodulation methods. A calibration subsystem generates synchronized spectral and thermal imaging data for training the neural network using controlled thermal profiles. The system captures steep thermal gradients and rapid cooling rates during laser powder bed fusion operations. A fiber embedding technique maintains the sensor in a strain-free condition at controlled sub-surface depths. The integration of high-temperature C-FBG sensors with machine learning signal processing achieves significant improvement in spatial resolution compared to traditional fiber optic thermal measurement approaches.

The invention discloses a transformer winding temperature and strain double-parameter monitoring method, system and device and a medium, and relates to the technical field of transformer monitoring, and the method comprises the steps: arranging a temperature sensing unit and a strain sensing unit on an optical fiber; a first reflection spectrum signal and a second reflection spectrum signal of the transformer winding are obtained and demodulated, and a first temperature signal of the transformer winding and a mixed signal containing a second strain signal and a second temperature signal are obtained; and taking the first temperature signal as compensation, and decoupling the mixed signal to obtain a second strain signal of the transformer winding. The fiber bragg grating technology is adopted, a plurality of sensors are inscribed on one optical fiber, synchronous measurement of temperature and strain is achieved, and strain data are compensated through winding temperature changes. According to the invention, the defects that only a single physical quantity can be monitored and the strain measurement accuracy is influenced by the temperature in the prior art are overcome, and accurate, convenient and efficient monitoring of the temperature strain of the transformer winding is realized in combination with online monitoring software.

The invention discloses a terahertz reflectance spectrum-based insulating oil micro-water content and aging detection method, device and equipment and a storage medium, and relates to the technical field of dielectric spectroscopy, and the method comprises the following steps: sampling each standard sample to obtain a to-be-detected sample, placing each to-be-detected sample in a sample bearing vessel to obtain an oil film layer, reflecting terahertz wave beams are utilized to irradiate each oil film layer to obtain a terahertz amplitude spectrum and a terahertz phase spectrum, a target three-layer optical model is utilized to perform inversion to obtain a complex dielectric function frequency spectrum, and a preset sub-dielectric model is utilized to fit the complex dielectric function frequency spectrum to obtain a main resonance characteristic and a characteristic parameter set; determining a reference micro-water content based on the main resonance characteristics, determining training data based on the characteristic parameter set, and constructing a target multi-task regression model; and processing the characteristic parameter set by using the model to obtain an initial prediction result and a confidence interval, verifying the prediction result to obtain a target prediction result, and improving the efficiency of detecting the micro-water content and aging of the insulating oil.

The invention discloses an unmanned aerial vehicle low-altitude target radar detection and recognition system, and belongs to the technical field of radar detection and target recognition. Comprising a quantum entanglement laser emission module, a quantum multi-dimensional receiving module, a multi-source heterogeneous data fusion module, a quantum-classical hybrid calculation module, a self-adaptive anti-interference module and an environment self-adaptive correction module, the quantum entanglement laser emission module is used for generating entanglement photon pairs carrying orbital angular momentum (OAM) and emitting the entanglement photon pairs to a detection airspace; and the quantum multi-dimensional receiving module demodulates the distance, the angle, the OAM attenuation spectrum, the polarization state and the reflection spectrum of the echo signal through coincidence measurement. The quantum entanglement-OAM combined modulation technology is adopted, and the non-locality of entanglement photon pairs and the multi-dimensional information bearing capacity of a 9-path orthogonal OAM mode are improved by more than three times compared with a traditional microwave radar; single-photon-level coincidence measurement is combined with quantum Fourier transform signal processing, the signal-to-noise ratio is increased by 3dB, and stable detection of a weak target is achieved.

The invention provides a polishing end point detection method, chemical mechanical polishing equipment and a storage medium, and belongs to the field of chemical mechanical polishing. The method comprises the steps that a reflection spectrum of the surface of a wafer in the chemical mechanical polishing process is obtained; correcting the reflection spectrum by adopting a preset correction coefficient to obtain a corrected reflection spectrum; dividing the corrected reflection spectrum into a dark noise spectrum and a to-be-measured spectrum based on a comparison result of the corrected reflection spectrum and a corresponding threshold value; determining the reflectivity of the surface of the wafer based on the to-be-measured spectrum, the dark noise spectrum, the reference spectrum and a preset reference spectrum reflectivity; the wafer thickness is determined using the reflectivity of the wafer surface to detect a polishing endpoint based on the wafer thickness. The terminal point detection accuracy is improved.

The invention discloses a multi-mode human body recognition system and method based on photoelectric metasurface and radar fusion, and belongs to the technical field of optoelectronics and radar perception. According to the system, optical edge enhancement is realized through a phase programmable metasurface, a continuous wave radar and a frequency modulation continuous wave laser radar are combined to obtain a micro-Doppler spectrum and a three-dimensional point cloud, and a near-infrared reflection spectrum is collected at the same time. According to the algorithm, a cross-modal Transform-GAT architecture is adopted to fuse multi-source features, edge extraction is optimized through a self-adaptive kernel function, and low-power-consumption edge reasoning is achieved through a Mach-Zehnder interference device or a ReRAM chip. Experiments show that the recognition accuracy of the method in complex scenes such as multi-target shielding and low light is remarkably superior to that of a single-mode scheme, the method has the edge deployment capability with delay smaller than or equal to 25 ms and power consumption smaller than or equal to 4 W, and an efficient solution is provided for multi-scene human body dynamic recognition and health monitoring.