22 results about "Scattering correction" patented technology

This invention discloses a highly sensitive detection method for carbendazim based on terahertz metamaterial resonance enhancement. The invention proposes a terahertz metamaterial sensor with a "+" shaped composite double-peak structure. Transmission spectra were acquired using terahertz spectroscopy, and the differences in terahertz spectra at different concentrations were investigated. Analysis revealed that the amplitude of the transmission peak in the terahertz transmission spectrum of this structure decreases with increasing carbendazim solution concentration. Data preprocessing was performed using multivariate scattering correction (MSC), feature extraction was conducted using the iterative information-preserving variable (IRIV) method, and a quantitative detection model for carbendazim terahertz transmission spectra was established based on LS-SVM. This provides a new approach for the quantitative detection of carbendazim. Experimental results show that the model has a high R-value. P The assay yielded a value of 0.9825, an RMSEP of 0.2001, and a detection limit (LOD) of 0.672 μg / ml. This invention provides a novel method for the rapid, non-destructive, and highly sensitive detection of carbendazim in food.

Methods for identifying micron-scale features below a surface of a tissue include emitting from one or more emitters at the surface of the tissue electromagnetic signals capable of penetrating tissue of different densities; obtaining from a sensor positioned at the surface of the tissue electromagnetic signals reflected from or transmitted through tissues of different densities as one or more obtained images; inputting the one or more obtained images into a machine learning model, wherein the machine learning model is trained to: identify scattering of certain electromagnetic signals comprising each of the one or more obtained images; correct for the scattering to create one or more images tissues at one or more subsurface depths; and identify micron-scale features of interest within the one or more images of tissues at the one or more subsurface depths.

The application discloses a kind of ocean polarization lidar multiple scattering correction methods based on machine learning, comprising: based on polarization Monte Carlo method, construct ocean polarization lidar radiation transfer numerical model, set multiple groups of lidar hardware and water body environment parameters, obtain a large number of simulation data, utilize machine learning method to construct the polarization multiple scattering correction model of different detection system, and use non-uniform water body to verify the effectiveness of model.According to the multiple scattering model of the effective attenuation coefficient of the measured signal and the lidar of different platforms, the signal depolarization ratio and water body parameter are calculated, and the polarization multiple scattering correction model is brought into the polarization multiple scattering correction model combined with the system parameters of lidar, to obtain seawater backscattering depolarization ratio.Compared with the direct seawater backscattering depolarization ratio of lidar signal depolarization ratio, the error between the seawater backscattering depolarization ratio corrected by the method and the in-situ value is significantly reduced.

This invention relates to a method and system for detecting fish lysate protein content based on near-infrared spectroscopy. The method includes: acquiring near-infrared spectral data of a fish lysate sample, and constructing a sample dataset based on the near-infrared spectral data and corresponding protein content labels; and performing partitioned scattering correction and characteristic peak enhancement processing on the near-infrared spectral data based on prior information of protein characteristic peaks to obtain an enhanced near-infrared spectral vector. Compared with existing technologies, the advantages of this invention are that it effectively preserves and enhances the key spectral features of fish lysate proteins, improves signal identification, eliminates interference from irrelevant components at the source, and significantly improves the prediction stability and repeatability accuracy of the model in complex industrial scenarios involving fish lysate.

This invention proposes a method for rapid determination of the content of Phellodendron amurense charcoal slices processed at different temperatures using near-infrared spectroscopy. Near-infrared spectroscopy is used to collect spectral information of Phellodendron amurense charcoal slices processed at different temperatures. Background interference is eliminated through preprocessing methods such as baseline correction, standard normal variable transformation, multivariate scattering correction, and first derivative. Combined with competitive adaptive reweighted sampling, CARS combined with iterative retention of information variables, and iterative combination optimization, a partial least squares regression quantitative prediction model is established. This model achieves accurate prediction of the content of five key components: 5-O-FQA, 3-O-FQA, berberine, berberine, and berberine, with prediction correlation coefficients all exceeding 0.90. This invention combines spectral information with chemometric methods to establish a digital content prediction model based on near-infrared spectroscopy for online monitoring of the Phellodendron amurense charcoal slice processing process. It has the advantages of objectivity, speed, non-destructive nature, and reliability, effectively solving the technical problem of lack of objectivity in existing technologies based on human judgment.

This invention relates to the technical field of testing or analyzing materials by measuring their physicochemical properties, specifically to a real-time monitoring system for multiple parameters of environmental water quality based on spectral analysis. The system includes optical paths for acquiring ultraviolet-visible (UV-Vis) and Raman spectra. These two optical paths are coaxially coupled via a dichroic mirror and an optical fiber combiner and connected to the same water flow cell. A timing control module enables time-division multiplexing acquisition of the same spatial coordinates. An edge computing module receives the combined spectral data, performs nonlinear scattering correction based on the Beer-Lambert law, separates the baseline drift component representing suspended matter scattering and the transmission component representing absorption characteristics, and inputs the characteristic peak intensity of the Raman spectrum as a priori constraint into the partial least squares inversion model. By qualitatively identifying specific organic compounds through Raman characteristic peaks, the solution space boundary of the corresponding overlapping bands in the UV-Vis spectrum is locked, and the inversion results for chemical oxygen demand (COD), total organic carbon (TOC), and ammonia nitrogen are output.

The application discloses a scattering correction method, a PET imaging method, a device, equipment and a storage medium. The scattering correction method comprises the following steps: obtaining TOF-PET data and an attenuation image of a target object during scanning; obtaining a back projection image of TOF information based on the TOF-PET data; obtaining a scattering estimation of TOF-PET imaging based on the back projection image and the attenuation image; and performing scattering correction on the TOF-PET data based on the scattering estimation of TOF-PET imaging. The application solves the technical problem of long time consumption of scattering correction in TOF-PET imaging in the prior art.

This invention discloses a non-destructive method for detecting strawberry freshness based on hyperspectral imaging and machine learning, belonging to the field of agricultural product quality testing technology. The method includes: acquiring hyperspectral images of strawberries in the 900-1700 nm wavelength range and extracting the average spectrum of the region of interest; performing multivariate scattering correction and Savitzky-Golay second-order derivative filtering preprocessing on the spectrum; using a genetic algorithm to screen a subset of characteristic wavelengths; and inputting the screened characteristic wavelengths into a pre-trained backpropagation neural network classification model to output the freshness grade determination result. This invention, through the technical path of "MSC and SG2 preprocessing—GA feature screening—BPNN classification," achieves rapid, non-destructive, and high-precision detection of strawberry freshness. The model's recognition accuracy exceeds 99% under storage conditions of 4℃ and 20℃, and can be widely applied to post-harvest grading, cold chain monitoring, and online sorting equipment for strawberries.

A multi-energy CT scatter correction method and apparatus based on a Boltzmann transport equation. The method comprises: preprocessing initial multi-energy projection data to be corrected of an object obtained after multi-energy CT scanning, to acquire initial corrected multi-energy projection data; performing material density conversion on the initial corrected multi-energy projection data to obtain material density distribution data; discretizing and labeling an actual scanning energy spectrum of the object during the multi-energy CT scanning, and determining an energy spectrum matrix in a pre-constructed labeled Boltzmann transport equation; calculating multi-energy scatter signal data of the object during the multi-energy CT scanning; and using the multi-energy scatter signal data to perform scatter correction on the initial multi-energy projection data to obtain actual multi-energy projection data.

Methods for identifying micron-scale features below a surface of a tissue include emitting from one or more emitters at the surface of the tissue electromagnetic signals capable of penetrating tissue of different densities; obtaining from a sensor positioned at the surface of the tissue electromagnetic signals reflected from or transmitted through tissues of different densities as one or more obtained images; inputting the one or more obtained images into a machine learning model, wherein the machine learning model is trained to: identify scattering of certain electromagnetic signals comprising each of the one or more obtained images; correct for the scattering to create one or more images tissues at one or more subsurface depths; and identify micron-scale features of interest within the one or more images of tissues at the one or more subsurface depths.

The application discloses a soil heavy metal pollution rapid detection method based on hyperspectral imaging, relates to the field of soil pollution detection hyperspectral technology, and comprises the following steps: obtaining original hyperspectral data of soil in a target region, obtaining standard hyperspectral data through noise removal, light path scattering correction and spectral smoothing denoising pretreatment, extracting a heavy metal characteristic spectral response interval and dividing the heavy metal characteristic spectral response interval into multiple sub-bands, calculating spectral absorption depth, absorption peak symmetry and first-order differential characteristic values of each sub-band to form a multi-dimensional spectral feature set, constructing and training a nonlinear mapping model to obtain a heavy metal content prediction model, inversely calculating the multi-dimensional features of each pixel point by point, and outputting a soil heavy metal content spatial distribution map. The method refines the spectral feature extraction dimension, realizes pixel-level inversion and spatial distribution visualization, and improves the accuracy and spatialization efficiency of soil heavy metal pollution detection.

A method for correcting scatter in a reconstructed medical image comprises inputting the reconstructed volume 302 into a trained machine learning model 304 to output a correction volume 306. The corre

The application provides a diamond wear-resistant floor construction process optimization method, and belongs to the technical field of wind tunnel construction.The application analyzes stress distribution by establishing a three-dimensional finite element model, removes base layer floating slurry by using a vacuum water suction device, sprays a penetrating interface agent to improve interface bonding, monitors interface bonding quality by using an ultrasonic detection device, realizes quantitative diamond scattering by using a variable frequency scattering device equipped with machine vision and proportional-integral-derivative closed-loop control, generates a distribution density three-dimensional cloud map and identifies abnormal areas by using a laser radar scanning device, carries out supplementary scattering correction and screeding treatment on the abnormal areas, and realizes quality determination by constructing a Voronoi diagram to calculate uniformity evaluation index, thereby solving the technical problem that local area wear-resistant performance is not up to standard due to the difficulty in accurately controlling diamond wear-resistant floor scattering uniformity.

This invention discloses a deep learning-based system for optimizing laser grooving process parameters in grinding wheels, belonging to the field of laser processing technology. It includes: a multi-dimensional physical property sensing unit for online measurement of the dynamic thermophysical parameters of the grinding wheel; a physical constraint deep learning prediction engine, whose loss function embeds Fourier's law of heat conduction as a soft constraint and corrects the laser absorptivity of the superhard abrasive through an optical scattering correction factor; a cross-domain small-sample transfer enhancement module, utilizing a physically consistent generative adversarial network and hierarchical multi-kernel maximum mean difference to achieve model transfer; a multi-objective process parameter optimization center, introducing hardware dynamic response constraints and searching for optimal parameters; and a closed-loop monitoring and adaptive adjustment module, integrating a coaxial spectral monitoring device to determine the oxidation level in real time and dynamically adjust the laser parameters. This invention solves the problems of prediction distortion in purely data-driven models and data sparsity in superhard materials, achieving high-precision physically consistent prediction and closed-loop stable control of processing quality.

This application relates to an intelligent X-ray image detection method for defects in solid-state battery electrodes, comprising: S1, acquiring an X-ray transmission image of the solid-state battery electrode to be detected; establishing an X-ray attenuation physical model; using the X-ray attenuation physical model to perform scattering correction calculation and thickness non-uniformity compensation calculation on the X-ray transmission image to obtain a structural correction image; S2, performing layered structural segmentation calculation on the structural correction image to obtain structural feature maps corresponding to the electrode material layer, solid electrolyte layer, and interface transition region, respectively; S3, calculating interface continuity parameters, grayscale statistical parameters, and texture feature parameters from the structural feature maps to form a multi-dimensional numerical feature vector; converting the multi-dimensional numerical feature vector into numerical defect characterization parameters; S4, inputting the numerical defect characterization parameters into a pre-trained defect recognition model for classification and localization calculation, and outputting defect type and defect spatial location information; S5, outputting the electrode defect detection result.

In one aspect, the disclosure relates to a method for correcting distortions or aberrations in a multi-photon microscopy image, the method including at least: (a) illuminating the sample; (b) imaging the illuminated sample with an imaging device having one or more detection elements that receive a signal from the sample; and (c) processing the image using a scattering correction and a phase correction. The phase correction can be a low-order deformable mirror adaptive optics correction with a hardware component and a software component. In some aspects, a spatial light modulator can be employed as a low-order phase correcting device. The scattering correction can be intensity scattering correction, phase scattering correction, or both, and can be performed with a high-order digital micromirror device with binary intensity modulation or a phase or intensity high-order spatial light modulator device. Also disclosed are methods of imaging complex tissue in living subjects.

The application discloses a hyperspectral rapeseed detection system, method and device. The system comprises a hyperspectral data acquisition unit for acquiring the reflected light of rapeseed samples in the short-wave infrared of 900-1700 nm; a multi-dimensional pretreatment fusion unit for synchronously implementing multiple pretreatments such as scattering correction, derivation and smoothing on the spectrum and cascading and fusing the results to construct a high-dimensional feature space; and a stacked ensemble learning unit for adopting XGBoost, a support vector machine and a random forest as heterogeneous base learners to preliminarily predict, and then adopting a Ridge regression as a meta learner to secondarily fuse, so as to output the rapeseed variety category and oil content. The device integrates a closed light-shielding cavity, a linear array push-broom InGaAs spectrometer, a uniform illumination module and an embedded control system, and realizes on-site rapid nondestructive detection. Through the cooperation of hardware and algorithms, the application realizes high-precision and high-stability real-time detection of the rapeseed variety and oil content.

This invention discloses a PET scattering correction and image reconstruction method based on time-of-flight (TOF) and GPU, belonging to the field of medical imaging technology. This invention uses a TOF histogram image as the initial radiation source distribution for scattering estimation for scattering correction. The computational overhead of the TOF histogram image is extremely low, and the initial scattering estimation can be executed in parallel with the sensitivity image calculation process, thereby effectively improving the computational efficiency of the overall reconstruction process. This invention has strong flexibility; it can be modularly integrated into EM-type reconstruction algorithms without requiring significant modifications to existing reconstruction frameworks, facilitating implementation. An atomically driven multi-GPU parallel MC computing framework solves the load balancing problem in multi-GPU computing, fully leveraging the performance advantages of multi-GPU computing resources. While meeting clinical needs in terms of computational efficiency, it can accurately estimate the scattering distribution, including multiple scattering coincidence events. The method of this invention is simple and easy to understand. This invention is applied to PET scattering correction and image reconstruction.

The application provides a contrast enhancement method for improving X-ray imaging quality of thick-wall GIS components, and belongs to the technical field of X-ray inspection of GIS components. After a unified three-dimensional imaging coordinate system is established, a Compton scattering point spread function is calculated by using a Monte Carlo scattering estimator, and a scattering correction image is obtained by frequency domain separation through a scattering deconvolution network. The scattering correction image is input into a pyramid attention fusion network to perform multi-scale feature extraction and defect saliency enhancement. A thermal diffusion contrast adjustment algorithm based on a local gradient structure tensor is performed on a preliminary enhanced image. A graph cut multi-region segmentation algorithm is used to independently optimize the contrast parameters for homogeneous regions. A closed-loop optimization is formed by adjusting the pyramid feature fusion weight coefficient according to the global contrast evaluation value feedback, thereby solving the technical problems of low X-ray imaging contrast of thick-wall GIS components and the inability to adaptively enhance different regions.

The application discloses a kind of oral cavity CBCT projection domain scattering suppression method and system based on double-flow deep learning network, it is related to oral cavity CBCT scattering suppression technical field, including creating double-flow parallel network architecture, inference process is carried out based on double-flow parallel process;Design high-fidelity training data preparation subsystem.The application improves the precision and image fidelity of scattering correction;Innovation uses size FOV physical scanning to obtain high-fidelity paired training data, ensures that model learns real scattering physical law, greatly enhances generalization and clinical robustness;The whole scheme does not need any hardware modification, only through software deployment can realize the soft upgrade of existing CBCT equipment, with very low implementation cost and wide popularization potential;The high-quality projection data finally output effectively improves bone density quantitative accuracy, tissue contrast and microstructure display capability, improves the diagnostic value and clinical reliability of CBCT in implantation, periodontal disease and apical lesion and other key diagnosis and treatment scenarios.

This invention discloses a method for optimizing the illumination distribution in oyster farms based on 3D modeling, comprising the following steps: collecting spatial structure data and water flow data of the oyster farm and unifying them to the same coordinate system; performing preprocessing to generate a standardized 3D modeling scene of the oyster farm; calculating the propagation distribution of light energy in the medium to obtain an initial photon energy density field; performing scattering correction on the initial photon energy density field; generating an ecologically corrected photon energy density field based on the photosensitivity feedback data of the oyster population; generating a time-balanced photon energy density field according to the temporal variation law of the ecologically corrected photon energy density field; calculating the photon energy density differences in different light source areas and adjusting the light source power and illumination angle; and generating illumination distribution results and a photon energy density heatmap. This invention utilizes 3D modeling and multi-source data fusion technology to achieve intelligent optimization of illumination distribution in oyster farms, possessing the advantages of accuracy, efficiency, and ecological adaptability.