78 results about "Scatter correction" patented technology

The invention discloses a method for detecting the nitrogen content of organ tissues of different varieties of oilseed rapes based on visible near-infrared spectroscopy, which comprises the following steps: firstly, collecting oilseed rape leaf, shell, stalk and root system samples under the conditions of multiple varieties and multiple nitrogen fertilizer levels, and acquiring spectral data within the range of 430-2500nm by using a visible near-infrared spectroscopy; and a Kjeldahl method is synchronously adopted to measure the real nitrogen content as a reference value. Preprocessing the spectral data, including de-noising, standard normal variable transformation, multivariate scatter correction and derivative processing, so as to weaken the influence of scattering and baseline drift; characteristic wavelengths related to the nitrogen content are screened through stepwise regression, variable projection importance, competitive self-adaptive reweighted sampling, a continuous projection algorithm and other methods, and a random forest model, a support vector machine model, a partial least squares discriminant analysis model, a partial least squares regression model, a support vector regression model, an XGBoost model and other models are combined. And respectively constructing a classification identification model and a regression prediction model.

Aiming at the problems of low efficiency, strong subjectivity, insufficient modeling ability, complex hyperspectral data noise, weak waxiness spectrum difference and the like of a traditional rice grain waxiness discrimination method, the invention provides a hyperspectral rice grain waxiness discrimination method based on multi-branch collaborative modeling. The method comprises the following steps: S1, acquiring glutinous and non-glutinous rice grain images by using a hyperspectral imaging system to obtain spectral data; s2, preprocessing spectral data through a combined method of SG smoothing, an asymmetric weighted penalty least square method and multivariate scatter correction to reduce noise and interference; s3, constructing a multi-branch modeling architecture which comprises a CNN local feature extraction module, an SRU spectrum sequence modeling module and a HorNet global high-order modeling module, and outputting a discrimination result through a classifier after multi-branch features are subjected to fusion and pooling attention weighting; and S4, evaluating the performance. According to the method, high-precision lossless discrimination of the waxiness is realized, and a technical support is provided for germplasm screening and quality evaluation in rice breeding.

A method for training an algorithm for machine learning for a correction of recordings obtained by an imaging apparatus. A number of output images are recorded. Moreover a smaller number of first, high-quality scattered radiation images are simulated from the output images. A corresponding number of second, low-quality scattered radiation images is further simulated, wherein the simulation is undertaken with a number of photons reduced by at least an order of magnitude. The algorithm is trained with the second, low-quality scattered radiation images as input data and the first, high-quality scattered radiation images as output data.

The invention discloses an accurate X-ray radiation dose measurement system based on artificial intelligence, which comprises the following steps: acquiring CT image data of a patient and pulse irradiation parameters of radiotherapy equipment through terminal equipment, and calculating a patient specific anatomical structure dynamic influence factor set, a pulse waveform time domain characteristic factor and a tissue interface scattering correction factor; all the parameters are substituted into a dose measurement formula to obtain an accurate X-ray radiation dose value; through a deep fusion image recognition technology (such as a high-resolution anatomical structure, tissue density gradient and heterogeneous region distribution provided by CT / MRI), a secondary particle scattering effect of a junction region of a tumor and a normal tissue can be accurately represented; and dynamic anatomical information (such as respiratory movement and organ displacement) in the image is combined to correct the time characteristic (such as dynamic change under an instantaneous high dose rate) of the pulse radiation field in real time, so that the dose calculation accuracy of a fast algorithm in a complex scene is remarkably improved.

The invention discloses an image processing system and method suitable for a turbid water area, and belongs to the technical field of image processing. The system comprises an underwater camera component for collecting original image data of a turbid water area in real time and an embedded image processing unit. The embedded image processing unit comprises a preprocessing module used for carrying out noise suppression and image calibration processing on original image data, and a physical parameter estimation module used for estimating environment background light intensity of a current water area, a water body attenuation coefficient and a distance map of an object in a scene and calculating initial transmissivity. The scattering correction module is used for removing back scattering and preliminarily compensating a color channel based on a physical imaging model in combination with physical parameters, and the deep learning enhancement module is used for improving image local contrast, adjusting brightness and finely correcting colors by adopting a lightweight deep convolutional neural network. According to the image processing system and method suitable for the turbid water area, the image definition and color of the turbid water area can be improved, the real-time performance is high, and the system and method are suitable for different water area environments.

The invention relates to the technical field of cone-beam computed tomography, and discloses a scattering correction method and system for cone-beam computed tomography. The method comprises the following steps: acquiring original projection data and main scattering data acquired by a first flat panel detector; wherein the first flat panel detector is arranged in the emission direction of the radiation source; acquiring secondary scattering data acquired by a second flat panel detector; wherein the second flat panel detector is arranged in the scattering direction of the radiation source; determining target projection data based on the original projection data, the main scattering data and the auxiliary scattering data; the target projection data is used for cone beam computed tomography imaging. According to the method, scattering correction can be completed on the premise of not depending on special hardware, not additionally increasing radiation dose and not losing imaging view, and the imaging precision can be effectively improved.

The invention provides an infrared spectrum-based material component analysis and defect detection evaluation method, which is characterized by comprising the following specific steps of: setting a scanning range by virtue of an instrument, and collecting infrared spectrum data in a sample; a self-adaptive iteration reweighted penalty least square method is sequentially adopted to carry out baseline correction, wavelet transform noise suppression and multivariate scatter correction standardization to process spectral data, a multi-dimensional characteristic parameter system is combined with a machine learning model, accurate identification of defect types and degrees is achieved, and the limitation that a traditional method can only carry out qualitative judgment is broken through; and through a component-defect linkage evaluation model, quantitative judgment of the overall performance of the material is realized, and the comprehensive requirement of an industrial scene for quality control is met.

The invention provides an image correction method and a double-projection imaging system, and relates to the technical field of images, in particular to the technical field of image correction. According to the method, a first projection image and a second projection image of a target object under a target projection angle can be acquired, scattering correction is performed on the first projection image, and a corrected image under the target projection angle is determined based on the first projection image after scattering correction and the second projection image of the target object under the target projection angle. Wherein the target projection angle is a projection angle within a target projection angle range, the first projection image comprises a shadow area formed after shielding of the first BSA, and the target projection angle range is a projection angle range where the first projection group and the second projection group are overlapped. According to the method, the phenomenon of shielding artifacts of the image in the reconstruction process can be effectively reduced.

SOLUTION: An X-ray imaging apparatus and associated method are provided that receive measured projection data from a wide aperture scan of a wide body-axis direction and a narrow aperture scan of a narrow body-axis direction within the wide body-axis direction and determine an estimated scatter in the wide body-axis direction using an optimized scatter estimation technique. The optimized scattering evaluation technique is based on the difference between the measured scattering in the narrow body-axis direction and the evaluated scattering in the narrow body-axis direction. The kernel-based scatter evaluation / scatter correction approach can be fitted to minimize scatter differences in a narrow body-axis direction. Thereafter, the fitted (optimized) kernel-based scatter evaluation / correction is applied to the wide body-axis direction region. Optimization may occur in the projection data domain or in the reconstruction domain. An iterative process is also utilized.SELECTED DRAWING: None

The invention relates to the technical field of cosmetic component detection and analysis, and discloses a method for rapidly recognizing arbutin in cosmetics based on surface enhanced Raman spectroscopy, AgNPs sol is prepared by an ascorbic acid oxidation-reduction method, and an SERS sensing interface is constructed through aspartic acid electro-polymerization modification; for different dosage forms of cosmetics, a to-be-detected solution is obtained through methanol extraction and centrifugal filtration purification, the to-be-detected solution and AgNPs sol are mixed and adsorbed, then a Raman spectrum is collected, pretreatment is conducted through combination of multivariate scatter correction and standard orthogonal transformation, characteristic wavelengths are screened through a genetic algorithm-combined interval partial least square method, and detection is achieved through combination of qualitative and quantitative models. Through cooperation of sensing interface specific modification and a data processing technology, the arbutin signal recognition capability is enhanced, matrix interference is eliminated, rapid and accurate detection of arbutin in cosmetics is realized, and the effects of high sensitivity, high specificity and simple and convenient operation are achieved.

The application relates to the technical field of image processing, and discloses a PET image enhancement method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a PET image to be reconstructed and measurement data, wherein the measurement data refers to sinogram data formed after signal data collected by a PET scanner is subjected to attenuation and scattering correction; and finally, based on the PET image, the sinogram data, a trained convolutional neural network and an EM fidelity algorithm, the PET image is reconstructed to obtain a target PET image. According to the application, a convolutional neural network for denoising is first trained based on a diffusion model theory on a high-quality (a high-quality PET image of a conventional dose) image, and then based on the trained model, high-quality image reconstruction is performed under the guidance of the PET image to be reconstructed and an EM fidelity algorithm.

The invention discloses a peanut kernel producing area identification method based on a near infrared spectrum. The method comprises the following steps: (1) collecting near infrared diffuse reflection spectrum data of peanut kernel samples from different producing areas by using a portable near infrared spectrometer; (2) carrying out preprocessing on the spectral data by utilizing a multiple scatter correction (MSC) method and a Savitzky-Golay smoothing filter; (3) extracting identification information of the spectral data by using a fuzzy identification analysis method; and (4) classifying by using a K neighbor classifier. According to the method, linear identification information of the near infrared spectrum value domain space and the null space of the peanut kernels can be extracted, the K-nearest neighbor classifier is adopted for classification, and the peanut kernel producing area can be rapidly and accurately identified. The peanut kernel producing area identification method based on the near infrared spectrum can be used for accurately identifying the peanut kernel producing area.

The invention discloses a CBCT scattering correction method based on dual-domain interaction and multi-scale wavelet transform, and the method comprises the steps: inputting original scattering-containing CBCT projection data into a preliminary projection correction network, and obtaining preliminary de-scattering projection data; performing three-dimensional reconstruction on the two types of projection data to obtain an original scattering-containing image and a preliminary correction image; fusing the features through an image domain feature fusion network to generate a fusion correction image; performing forward projection on the fusion correction image to obtain reference projection data, and calculating a difference value between the reference projection data and the original projection data to obtain differential projection data; performing multi-scale two-dimensional discrete wavelet transform on the differential projection data, and optimizing coefficients; performing inverse transformation to obtain a refined scattering diagram; and subtracting the refined scattering image from the original projection data, and reconstructing to obtain a final non-scattering CBCT image. According to the method, depth interaction between the projection domain and the image domain is realized, signals are accurately separated in scattering estimation, and the image quality, the detail retention capability and the robustness of the algorithm are effectively improved.

The invention discloses a system for multispectral transmission image breast lesion segmentation, and belongs to the technical field of medical image processing and optical imaging. The system comprises an image preprocessing subsystem and a semi-supervised segmentation network subsystem. The preprocessing subsystem adopts modulation and demodulation, frame accumulation and a scattering correction technology based on spatial pyramid matching, so that the signal-to-noise ratio and the contrast ratio of an image are remarkably improved; the segmentation network subsystem proposes a multi-head consistency semi-supervised network model, integrates a multi-head decomposition module and a cross supervision module, and effectively utilizes unlabeled data to carry out uncertainty modeling and consistency constraint. Under the condition that only 20% of labeled data is used, the segmentation Dice coefficient on the self-built mammary gland phantom data set reaches 91.47% and is obviously superior to that of an existing method, and a complete solution from image enhancement to intelligent segmentation is provided for early diagnosis of the breast cancer.

In order to reduce a computing effort for correcting initial images from an imaging apparatus, a scatter model is used to simulate scattered-radiation images, and corresponding initial images are selected from the original initial images. These are used to train a neural network. The neural network may be used to calculate scattered-radiation single images that may be used to correct each individual initial image into a corrected initial image.

Systems, methods, and devices for generating a corrected image are provided. A first robotic arm may be configured to orient a source at a first pose and a second robotic arm may be configured to orient a detector at a plurality of second poses. An image dataset may be received from the detector at each of the plurality of second poses to yield a plurality of image datasets. The plurality of datasets may comprise an initial image having a scatter effect. The plurality of image datasets may be saved. A scatter correction may be determined and configured to correct the scatter effect. The correction may be applied to the initial image to correct the scatter effect.

Systems, methods, and devices for generating a corrected image are provided. A first robotic arm may be configured to orient a source at a first pose and a second robotic arm may be configured to orient a detector at a plurality of second poses. An image dataset may be received from the detector at each of the plurality of second poses to yield a plurality of image datasets. The plurality of datasets may comprise an initial image having a scatter effect. The plurality of image datasets may be saved. A scatter correction may be determined and configured to correct the scatter effect. The correction may be applied to the initial image to correct the scatter effect.

The application provides a scattering correction method and device, a digital device and a computer readable storage medium. The method comprises: obtaining a down-sampling response line based on detection data; obtaining an estimated target function corresponding to each down-sampling response line by using a matrix estimation method based on a maximum likelihood-expectation maximization iterative algorithm; solving the estimated target function to obtain a number of scattering coincidence events corresponding to each down-sampling response line; performing up-sampling processing on the down-sampling response line to obtain an up-sampling response line; and calculating the number of scattering coincidence events corresponding to each up-sampling response line. When the embodiment of the application performs scattering correction, it does not depend on activity images and attenuation images, and the method does not require a hyperparameter, so that the hyperparameter no longer needs to be set in a determined range, and the method is simple and accurate.

The invention discloses a high-energy cone beam industrial CT scattering correction method, and belongs to the field of industrial nondestructive testing. According to the method, on the basis of a CAD model and material attribute information of a target industrial component, a total projection image and scattering distribution image pairing data set is generated through Monte Carlo analog simulation. And constructing an improved U-Net architecture convolutional neural network model, preprocessing a total projection image, inputting the preprocessed total projection image into the model for training, and performing optimization by a composite loss function. And inputting an actually acquired total projection image into the trained network after same preprocessing, outputting a predicted scattering distribution image, and correcting the predicted scattering distribution image in a projection domain. And multi-component scatter correction generalization is realized through a transfer learning strategy. According to the method, Monte Carlo simulation and deep learning technologies are combined, the defects of a traditional method under the high-energy condition are overcome, scattering correction efficiency and accuracy are improved, and the method has good engineering application value.

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.

Apparatus for measuring tissue health and methods of use. The device includes a plurality of sensors for detecting events related to positron decay emitted from drug radionuclides in tissue, and a processor. The processor measures a first count rate indicative of a three-photon emission associated with a positron first decay mode; measuring a second count rate indicative of two-photon emission associated with a positron second decay mode; applying a scatter correction factor to each of the first count rate and the second count rate to account for scattered and attenuated annihilation photons in the tissue; determining a ratio of the first count rate to the second count rate; determining the decay lifetime of the positive-positron element (o-Ps) based on the ratio; and determining the health condition of the tissue based on the decay life of the o-Ps.

The application discloses a kind of self-adapting direction and step of multi-energy spectrum CT iterative reconstruction method with scattering correction, comprising: step 1, setting multi-energy spectrum CT projection reconstruction model with scattering correction;Step 2, using iterative algorithm to solve out the increment of base material distribution projection value angle by angle, update the current base material image to new base material image;Wherein, through self-adapting direction and step to quickly obtain convergent base material projection image.The scheme provided by the application is used to remove scattering, and the result of base material decomposition is more accurate, and the reconstruction result artifact is less, which lays a foundation for quantitative analysis.

The invention belongs to the technical field of crop cultivation and breeding, and particularly discloses a system and method for predicting the germination rate of cucumber seeds based on a hyperspectral imaging technology. The method comprises the following steps: acquiring materials in Dai nationality autonomous prefecture in Xishuangbanna of Yunnan province, performing seed reproduction in a specified base to obtain experimental seeds, performing serial number scanning on the seeds harvested in 2021-2023 to obtain spectral information, and acquiring images with the wavelength of 400-1000nm by using a specific hyperspectral imaging system. The method comprises the following steps: constructing a data set after marking the germination condition through a germination experiment, preprocessing a spectrum by applying at least one method such as multivariate scatter correction, extracting a characteristic wavelength by adopting an algorithm such as non-information variable elimination, predicting by virtue of a KNN or LogitBoost model, and evaluating by using indexes such as accuracy and the like, thereby effectively predicting the germination rate of the Xishuangbanna cucumber seeds.

The invention relates to a ray inspection system and a scattering correction method. The ray inspection system comprises: a ray source (1) configured to generate a ray beam (10); at least part of the first detector array (2) is located in the coverage range of the ray beam (10); the second detector array (3) and the first detector array (2) are located on the same side of an inspection object (4) of the ray inspection system, the second detector array (3) is located outside the coverage range of the ray beam (10), and the second detector array (3) is configured to receive scattered signals of the ray beam (10) in the process of penetrating through the inspection object; and the processor is in signal connection with the first detector array and the second detector array, and is configured to convert the scattering signals according to a preset relationship between the scattering signal detection capabilities of the first detector array and the second detector array, and perform scattering correction on the received signals of the first detector array according to the converted scattering signals.

The invention relates to a scatter correction method based on a side irradiation scintillator. The cuboid scintillators are sliced into piles along the long sides in an equal-thickness mode; collimated beams matched with the sheet thickness only penetrate through the first sheet from the side face to irradiate, the camera directly faces for imaging, and the brightness is extracted one by one; fitting a scattering attenuation relation based on an illuminated experiment to obtain a scattering correction factor; and deducting scattering of the actually measured brightness item by item according to layers, recovering the real brightness, and obtaining three-dimensional dose distribution according to brightness-dose calibration. According to the method, initial scintillation light and interlayer scattering can be separated, the structure is simple, calibration and measurement are consistent, and the three-dimensional dose measurement precision and reproducibility are remarkably improved.

The invention discloses a hyperspectral detection method for the sugar degree and hardness of tomatoes. The method comprises the following steps: firstly, acquiring a hyperspectral reflection or transmission spectrum of tomatoes in a range of 400-1000nm, and carrying out pretreatment such as standard normal transformation, multivariate scatter correction, Savitzky-Golay smoothing and orthogonal signal correction on the original spectrum; then, characteristic wavelength screening is carried out on the preprocessed spectrum through a competitive adaptive reweighted sampling (CARS), an irreducible variable elimination (UVE) or a UVE-CARS joint algorithm, and a characteristic spectrum matrix is obtained; a Kennard-Stone (KS) algorithm is further adopted to divide a training sample into a correction set and a prediction set, a tomato sugar degree model and a tomato hardness model are constructed based on partial least squares regression (PLSR), and an optimal main factor number is determined through cross validation; and inputting the characteristic spectrum of the tomato to be detected into the model, and outputting the predicted values of sugar degree and hardness. The method has the advantages of no damage to fruits, high modeling stability, good prediction precision and the like, and is suitable for the fields of tomato quality evaluation, grading, postharvest treatment and the like.

The present application relates to a kind of vortex electromagnetic wave target high-frequency scattering calculation method based on equivalent edge electromagnetic current correction, belong to electromagnetic calculation field.Method includes: constructing target triangulation model and extracting convex split edge;Bessel vortex wave is decomposed into plane subwave superposition;The main scattering field under the irradiation of subwave is calculated using physical optics PO;Based on physical diffraction theory, the diffraction field of subwave in convex split edge is calculated using PTDEEC model of equivalent edge electromagnetic current;The main scattering field and edge diffraction field are superposed to obtain the modified total field, and the radar cross section (RCS) is calculated.The present application effectively corrects the diffraction error of traditional PO method at the edge of target by equivalent edge electromagnetic current, significantly improves the scattering calculation precision of large-size edge target under complex vortex beam irradiation, and provides more reliable high-frequency algorithm for radar target recognition and stealth design.

The invention provides a marine steel laser-arc hybrid welding arc spectral feature extraction method which comprises the following steps: step 1, laser-arc hybrid welding spectral signal acquisition: a high-resolution spectrometer is matched with double optical fiber probes to synchronously acquire signals of corresponding channels, and cross interference of different wavebands is eliminated through grating light splitting; step 2, noise reduction processing of the spectral signal: adopting a multiplicative scatter correction (MSC) algorithm to carry out noise reduction processing; step 3, spectral signal feature extraction: extracting electron density and electron temperature features respectively; according to the marine steel laser arc hybrid welding arc spectral feature extraction method, the arrangement of the extraction method can be optimized, and the reliability and accuracy of spectral feature extraction are enhanced.

The application relates to a PET image scattering correction method and system, which comprises the following steps: modeling a PET acquisition process; alternately solving radioactive activity distribution and detector space attenuation coefficients based on an ADMM algorithm; realizing mapping of PET detection data and detector space attenuation coefficients to accurate scattering distribution estimation based on a deep learning network G; keeping the radioactive activity distribution and the detector space attenuation coefficients constant, and utilizing an unfitted scattering distribution estimation to maximize a log-likelihood function to obtain a scattering correction factor; and utilizing the detector space attenuation coefficients and the accurate scattering distribution after fitting of the scattering correction factor to obtain a PET image after scattering correction. The method has the beneficial effects of avoiding image quality decline caused by inaccurate scattering evaluation, greatly reducing scattering correction time consumption, and greatly improving stability for high-noise data.

The invention discloses a jackfruit quality evaluation method based on microminiature hyperspectrum combined with machine learning, and the method comprises the steps: obtaining hyperspectral signals of jackfruits with different maturity degrees under the wave band of 713-920 nm, determining physicochemical indexes according to a traditional method, preprocessing the collected spectral signals through the methods of multivariate scatter correction, standard normal variable transformation and the like, and calculating the quality of the jackfruits with different maturity degrees. A jack fruit quality evaluation model is constructed by combining machine learning methods such as partial least squares, a support vector machine and a one-dimensional convolutional neural network on the basis of characteristic wave band screening algorithms such as a continuous projection algorithm and competitive adaptive reweighted sampling, and effective prediction of the jack fruit storage quality is realized by only using 10 characteristic wave band parameters. According to the method, the maturity and key quality indexes of the jackfruit pulp can be nondestructively detected only by collecting the microminiature hyperspectral signals of the jackfruit pulp, and in the links of planting, storing, processing and selling the jackfruit, the technology can remarkably improve the efficiency and accuracy of quality monitoring and support on-site rapid detection.