101 results about "Particle identification" patented technology

The invention relates to the technical field of partial discharge detection of power equipment, and discloses a KNN algorithm-based photoacoustic feature combined particle identification method and system, optical and acoustic signals caused by particles in different shapes under alternating voltage are obtained through experiments, and mathematical modeling is performed on particle motion discharge based on particle motion and discharge theories; and then, time sequence distribution characteristics of optical and ultrasonic signals triggered by the particles are mastered, a particle type identification algorithm based on a KNN model is constructed, and particle shape identification and mixed particle component identification are realized. In conclusion, the method combines an optical detection method and an ultrasonic method to form photoacoustic combined detection, excavates combined feature information and establishes a particle discharge combined detection method, and has important significance on application of a particle discharge optical detection method and multi-parameter combined detection; and the method has outstanding value for maintaining safe and stable operation of power equipment and even a power system.

The invention discloses a shield muck coarse particle gradation real-time identification system and method based on multi-modal machine vision, and relates to the field of muck particle identification, and the system comprises a multi-modal data collection module which is composed of a plurality of groups of sensors and is used for collecting multi-modal data of muck particles; the edge calculation processing unit is used for preprocessing the multi-modal data; the analysis module is used for constructing a deep learning model and realizing grading prediction and mud content prediction; and the control module is used for real-time feedback and control including visualization, decision support and tunneling parameter adaptive adjustment. According to the system, accurate real-time analysis of geometric, image and material features of muck particles is realized through multi-modal sensor fusion and edge calculation efficient preprocessing in combination with feature-level / decision-level fusion and a lightweight deep learning model; by means of closed-loop control, the method has high adaptability and engineering intelligent feedback capacity, and muck grading recognition precision and tunneling parameter self-adaptive adjustment efficiency in shield construction are remarkably improved.

The invention relates to the technical field of drilling waste treatment, in particular to a drilling slag slurry multi-stage circulation treatment system and method.The drilling slag slurry multi-stage circulation treatment system comprises an automatic grading module, a dynamic monitoring module, a slurry purification module, a separation optimization module, a waste slag classification recycling module and an intelligent control module; wherein the automatic grading module comprises a particle identification unit and a variable frequency grading unit; the particle identification unit is used for carrying out real-time detection on the size, the form and the density of particles in the drilling slag and the slurry through a multispectral sensor; the variable-frequency grading unit is used for adjusting the grading speed according to the detection data provided by the particle identification unit and carrying out dynamic grading on the particles; the dynamic monitoring module comprises an online sensor and a data analysis unit; according to the invention, through intelligent monitoring and automatic adjustment technologies, efficient grading, filtering and separation of drilling slag slurry and intelligent classification of waste slag are realized, and efficient operation of the system, improvement of the resource recovery rate and reduction of environmental pollution are ensured.

The invention discloses a cutting fluid solid particle concentration measurement method and system, and relates to the technical field of particle concentration measurement, a time sliding window mode is used for carrying out fusion processing on an image sequence, synchronous acquisition of temperature and illumination intensity is combined, a dynamic temperature illumination normalized background image is constructed, and the concentration of solid particles in a cutting fluid is measured. The problems of brightness drift, image flicker and the like caused by environmental fluctuation are effectively eliminated, and the stability and adaptability of background modeling are remarkably improved. Through a disturbance identification and characteristic parameter extraction module, multi-dimensional characteristic extraction is carried out on non-particle disturbance such as bubbles, oil films and light reflection possibly occurring in the image, a disturbance characteristic set Rdr is formed, and the system can accurately identify and distinguish non-real particle signals. And through a dynamic difference denoising analysis module, a disturbance feature set Rdr is combined to carry out adaptive denoising on a difference image, a real particle structure is reserved, a non-particle interference region is eliminated, and a purified image with an enhanced edge is output, so that cleaner and more accurate basic data is provided for subsequent particle identification.

The invention relates to the technical field of intelligent sorting, in particular to a full-automatic rice impurity removing method and device based on machine recognition, and the method comprises the following steps: extracting rice particle reflection intensity and analyzing a boundary trajectory, screening a mutation region to generate a recognition section, judging the consistency of recognition quality and weight error directions, and generating an error sample set. And performing statistics on disturbance frequency to generate a deformation mark, comparing numbers, extracting coordinates to generate a list, calculating a response difference, and outputting a rejection identification sequence. According to the invention, through comparison of reflection boundary trajectory abrupt change, focusing of a spectrum abnormal section, improvement of particle difference positioning precision, combination of consistency judgment of image identification errors and weight deviation, enhancement of abnormal particle screening accuracy, tracking of contour disturbance frequency identification of deformation behaviors, and enhancement of dynamic feature identification, the accuracy of abnormal particle identification is improved. An operation target is positioned based on multi-source feature intersection, orbital offset and air injection response time sequence calibration are combined, it is guaranteed that removal actions are coordinated and consistent, and efficient recognition and accurate removal of the mixed rice particles are achieved.

The present invention discloses a basin flood risk assessment method and system based on a particle recognition algorithm, and belongs to the field of flood warning technology. Although the existing scheme can measure the real-time flow of the river, it does not involve how to deduce and evaluate the flood inundation situation, resulting in the inability to quickly know the flood inundation situation around the river bank. The basin flood risk assessment method based on a particle recognition algorithm of the present invention, by constructing a basin spatial discrete model, a flow velocity calculation model, a flow simulation model, a two-dimensional hydrodynamic model, and a flood risk assessment model, can not only calculate the cross-sectional flow of the basin, but also deduce and evaluate the flood inundation situation, so that the flood inundation risk and inundation data around the water area can be quickly known, so as to meet the needs of basin flood risk analysis, improve the accuracy of flood disaster warning, and provide more data supplement and management basis for basin flood control.

The invention discloses an abrasive particle monitoring method and device and electronic equipment, and belongs to the field of monitoring methods, and the method comprises the steps: collecting a to-be-detected signal based on each probe in an electrostatic sensor; performing variational mode decomposition processing on the to-be-detected signal to obtain a target eigenmode function corresponding to the to-be-detected signal; performing sparse representation based on the target eigenmode function to obtain a target signal corresponding to the to-be-detected signal; performing dynamic time warping on each group of target delay signals by using a sliding window to obtain similarity results respectively corresponding to each group of target delay signals; and if a wave crest meeting a preset condition exists in the sequence diagram presented by the similarity result, determining the number of the abrasive particles according to the wave crest, and obtaining an abrasive particle monitoring result. Through the method, the abrasive particle trigger signal is effectively extracted, the interference of abnormal pulses on abrasive particle identification is reduced, and the accuracy of abrasive particle identification is improved.

The invention belongs to the technical field of oil production engineering, and provides a method for identifying proppant particles based on ResNet-KAN. The identification method comprises the following steps: acquiring logging rock debris samples at different well depths in a drilling and coring process; screening and photographing the single particles of the logging rock debris sample; extracting an image of the single particle, measuring the particle size of the single particle, and storing; making a training set and a verification set of the model; a ResNet-KAN residual error neural network model is constructed; training the model by using a training set, and verifying by using a verification set; and identifying the image of the single particle by using the trained model so as to identify the proppant particles. According to the method, the KAN network is used for replacing a linear full connection layer of a traditional ResNet network, so that further optimization of the ResNet network is achieved, and the method has the advantage that the propping agent particles in the coring well logging rock debris in the hydraulic fracturing test field can be automatically, rapidly and accurately recognized.

Systems and methods for spectroscopic analysis and particle identification using out-of-plane imaging of multimode-interferometer waveguide (MMI-WG) scattering are provided. A MMI-WG comprises an input end, a lateral surface, and an input port disposed on the input end, such that light in the MMI-WG propagates in a direction away from the input end. A sensor is configured to detect scattered light that scattered through the lateral surface of the MMI-WG, and to generate data based on the detected scattered light, wherein the data indicates an intensity of the scattered light. One or more processors are configured to determine, based on the intensity of the scattered light indicated by the data generated by the sensor, one or more wavelengths of the input light. The identity of one or more particles may be determined based on the one or more determined wavelengths.

The invention relates to an intelligent particle identification method for lunar soil particle screening, which comprises the following steps of: acquiring an optical signal, image data and a surface electrical signal of the same lunar soil sample particle to form a multi-modal feature set of each lunar soil sample particle; carrying out standardization processing on the multi-modal feature set; constructing a multi-modal deep learning neural network model based on a CNN + MLP + splicing fusion module structure; performing supervised learning training on the multi-modal deep learning neural network model to obtain a trained multi-modal deep learning neural network model; and inputting the standardized data into the trained multi-modal deep learning neural network model to realize particle classification. According to the method, an intelligent identification integrated deep learning model is adopted, the functions of particle shape distinguishing, component identification, particle size sorting and the like are carried out by utilizing an intelligent algorithm, automatic classification, clustering and attribute prediction can be realized, manual marking one by one is not needed, and the working efficiency is effectively improved.

The invention discloses a charged particle identification method and system based on a screen grid ionization chamber signal waveform, and belongs to the technical field of radiation detection. The charged particle identification method comprises the following steps: acquiring a current signal waveform generated by charged particles at an anode of a screen grid ionization chamber and extracting a characteristic quantity; analyzing the relationship between the obtained characteristic quantity and the particle energy and incident angle, and constructing particle identification parameters; according to a relation between particle energy and a range, drawing by using related identification parameters, and determining distribution characteristics of different particles; and identifying different particles according to different distribution characteristics. According to the method, the energy deposition characteristics of the particles are utilized, and the constructed identification parameters weaken the influence of the incident angle of the particles on particle identification, so that the five particles of p, D, T, 3He and 4He can be accurately identified in a relatively large angle range, and the section measurement efficiency is improved.

The invention provides a liquid particle counting method and device based on an image sensing chip, and mainly solves the problems that in the existing oil product particle detection technology, a laser scattering automatic particle counting method is low in detection precision, and a microscope detection method is high in cost and poor in reliability. According to the scheme, a microscopic image of a liquid sample is shot by a chip end through a light source and an image sensing chip, the collected microscopic image of the liquid sample is encoded and preprocessed by a computer end, and particulate matter recognition and particle size analysis are realized through a computer vision target detection module based on deep learning. Experiments prove that the method can accurately identify particles and effectively remove interferents such as bubbles and water drops, the measurement precision is improved by about 25%, and the particle detection range is expanded to 0.4-1000 microns. According to the overall design, collaboration and unification of miniaturization, high precision, high efficiency and low cost are achieved, and a new technical path is provided for real-time monitoring and field application of oil product particles.

The invention discloses a radiation particle identification method based on an improved ResNet-18 network and a CIS transient response technology, and the method comprises the steps: collecting a CIS dark field image in a radiation environment, and generating a multi-modal data set according to neutron, proton and heavy ion radiation experiment sample images; a transient response image data set is obtained through theoretical calculation based on radiation analog simulation software; preprocessing the transient response image data set, and marking particle type, energy and angle information of each sample image as a label of multi-task learning; a pre-trained ResNet-18 model is used, an input layer is modified to adapt to the image size, and a channel attention module is added; and carrying out compression quantification on the model by adopting an optimizer in combination with gradient cutting through a self-adaptive category weight adjustment strategy. By constructing the intelligent feature extraction network, the system realizes automatic identification of transient response geometric features, and breaks through the bottleneck that a traditional method depends on artificially defined features.

The invention discloses a debris flow surface SPH particle adaptive processing method and a debris flow real-time simulation method. The invention provides a debris flow surface SPH particle adaptive processing method for overcoming the defect that surface particles are easily misjudged in the prior art. A particle density dynamic threshold is used as a surface particle identification standard, and the optimized dynamic threshold dynamically balances and controls debris flow characteristics to make a judgment contribution to the surface particles by using a real-time velocity gradient, a density gradient and a weight. Neighborhood virtual particles are randomly generated around the mass center of the surface particles to support kernel calculation, and the density and speed attributes of the virtual particles are limited. The debris flow real-time simulation method is realized by utilizing a self-adaptive processing technology. According to the optimization scheme, SPH-DEM coupling is adopted, heterogeneous units form a unified SPH framework, hydrodynamic force is calculated through an SPH method, contact force is calculated through a DEM method, and bidirectional coupling force calculation is completed. An SPH method is realized by adopting a Niagara particle system, and a digital twinborn platform foundation of numerical simulation and visual integrated simulation is constructed.

Methods and apparatus are disclosed for concurrent digital autoradiography of alpha-particle events and positron events using a spatially resolving charged particle detector and a gamma-ray detector. Alpha particles and positrons are detected by the charged particle detector. Positrons are identified based on coincidence with gamma-ray events. A positron-emission autoradiograph is formed based on positron positions. Alpha particles, both coincident and anticoincident with positrons, are identified based on energy deposition patterns. An alpha-emission autoradiograph is formed based on alpha-particle positions and energies. Separation of coincident alpha and positron events recorded by the charged particle detector improves position detection accuracy. Validation of positron imaging for alpha-emitter distribution or dosage, based on correlation of respective autoradiographs, is described. Variations for particle identification based on isotropy, energy, coincidence, or anticoincidence are presented. Disclosed techniques are applicable to alpha-emitting radionuclides with complex decay chains.

The present invention discloses a particle statistics method, device, equipment and storage medium. The method comprises: obtaining a bright field image corresponding to a current particle identification and a dark field image corresponding to the bright field image, the dark field image comprising at least two droplets, at least one of the at least two droplets comprising at least one particle corresponding to the current particle identification; determining parameter data of each droplet in the bright field image, determining parameter data of particles corresponding to the current particle identification in the dark field image, the parameter data comprising position range data; determining the statistical results of particles corresponding to the current particle identification according to the position range data of each droplet and the position range data of particles corresponding to the current particle identification. The technical effect of quickly and accurately determining the particle statistics results is achieved.

Provided are a particle detection device and a particle detection method that have high particle identification properties, can ensure sterility, and have automation suitability.A particle detection device includes a holding part that holds a suspension containing particles, a first flow channel that is connected to the holding part, an observation window that is connected to the first flow channel, a liquid feeding part that transfers the suspension held in the holding part to the observation window, and a fluorescence detection part that detects fluorescence emitted from the particles contained in the suspension through the observation window.

The application discloses a kind of desulfurization slurry gypsum particle impurity rate analysis methods, comprising: using holographic reconstruction method to the holographic image of desulfurization slurry is locally reconstructed;The cross section of local reconstruction is axially positioned, adopts depth of field expansion method to focus image fusion in the same cross section in particle, obtain depth of field expansion chart;With neural network model identification particle image in depth of field expansion chart, output binary image and particle contour coordinates of depth of field expansion chart;With the neural network model with self-attention mechanism module, according to particle contour coordinates, single particle gray scale chart is intercepted from depth of field expansion chart, then according to gray distribution feature is divided into gypsum particle and impurity particle;The geometric parameter information of gypsum particle and impurity particle is extracted, combined with particle category, calculates the impurity rate of desulfurization slurry gypsum particle.The analysis method provided by the application improves the accuracy of particle identification and classification, and improves the accuracy of the calculation of the impurity rate of desulfurization slurry gypsum particle.

This invention discloses a machine learning-based method and system for identifying exhaust gas particles, relating to the field of image processing technology. By acquiring exhaust gas particle images, feature extraction is performed to obtain color cast features. Based on these color cast features, the exhaust gas particle images are color-corrected to obtain a color-corrected image. Particle distribution features are extracted from the color-corrected image using a target model. These particle distribution features are then fused with the color-corrected image to obtain an adaptive transmission map. The backscattering coefficient is calculated based on the adaptive transmission map, and the adaptive transmission map is enhanced based on the backscattering coefficient to obtain a final enhanced image. Exhaust gas particles are identified based on the final enhanced image. Through color correction, adaptive transmission map construction, and backscattering enhancement, the color accuracy and particle distribution visualization of exhaust gas particle images are improved, along with image clarity and detail, thereby enhancing the accuracy and reliability of particle identification.

An automated portable Raman microscope combines the simplicity, ruggedness, and selectivity of Raman spectroscopy with the sensitivity of single-particle identification. It includes a visible camera, controllable illumination, and a moving optical head that together acquire images of the sample at different focal planes and under different lighting conditions. An internal processor automatically identifies particles or other targets in these images for Raman analysis, e.g., shifted excitation difference Raman spectroscopy. The moving optical head scans the Raman pump beam across the sample to different targets. A built-in user interface displays the results of the Raman analysis superimposed on visual images of the sample. The Raman microscope's simplicity, reliability, and low maintenance make it easy to use for a wide variety of applications, including analyzing and identifying particles and residues, linking chemical and biometric evidence, and forensics analysis.

The invention relates to the technical field of oil particle recognition, in particular to an oil particle image recognition analysis method and system based on reinforcement learning. When to-be-monitored oil liquid is monitored in real time according to the data of the reinforcement learning model, after the trained reinforcement learning model receives a to-be-detected infrared image, single-particle and large-particle substances in the image are automatically recognized, the large-particle substances are combined with the projection area of the large-particle substances and the average particle size threshold value in the S3, and the to-be-monitored oil liquid is monitored in real time. The calculation method for calculating the number of the single particles comprises the step of dividing the area of the large particles by the area of the small particles to obtain the number of the large particles composed of the small particles so as to avoid counting errors of the small particle sizes; after calculation, a result is output, the output content is the actual count (the number of single particles + the sum of the number of single particles after aggregate splitting) of the particles in the output oil, and data support is provided for equipment loss diagnosis; therefore, the problem of low counting caused by misjudgment of aggregates as single particles due to particle aggregation in the high-viscosity oil liquid is avoided.

This invention belongs to the field of digital holographic imaging and artificial intelligence technology, specifically a holographic particle recognition method based on EFPFNet. While maintaining the traditional holographic optical path structure, this invention addresses the problems of low target contrast, blurred edges, and strong background interference in micro-nano particle holograms. Unlike the traditional approach of directly transferring general target detection models, it proposes a neural network architecture specifically for holographic image features: employing an edge feature convolution (EFC) module and a particle focusing mechanism (PFM) module to enhance the perception of weak particle diffraction features and suppress complex background noise, respectively; and using simulation data generated based on a physical model as the training set, effectively solving the problem of scarce real labeled data. This method possesses high detection accuracy, high robustness, and excellent generalization ability, and can be applied to target detection scenarios such as biomedicine and industrial inspection without significantly increasing optical hardware costs or detection time.

The invention provides a ceramic slurry production monitoring method for a multilayer ceramic capacitor, and relates to the technical field of production monitoring. The ceramic slurry production monitoring method comprises the steps from S01 to S07. S01, obtaining a ceramic slurry sample. And S02, detecting bubble information of the ceramic slurry sample by using an ultrasonic detector. And S03, measuring the viscosity of the ceramic slurry sample at a preset temperature and a preset rotating speed by using a rotary viscometer. And S04, drying the ceramic slurry sample with a preset mass, and calculating the mass percentage of solid residues after drying to obtain the solid content of the ceramic slurry sample. And S05, preparing the ceramic slurry sample into a dry smear, and obtaining a microscope image of the smear. And S06, identifying particle information in the microscope image through the particle identification model. Wherein the particle information comprises the particle type and the particle size of each particle. And S07, according to the bubble information, the viscosity, the solid content, the particle type and the particle size, judging whether the ceramic slurry is qualified or not by referring to a preset judgment standard.

The application discloses a kind of recognition methods of row ampoule plastic particles, it is related to ampoule plastic particle identification technical field, to solve the technical problem of low accuracy rate of plastic particle identification in the process of existing ampoule filling, comprising the following steps: S1, image acquisition, in the filling process after row plastic ampoule blow molding, the image acquisition equipment integrated on filling head is used to carry out multiple image acquisition to ampoule inside;S2, abnormal feature identification, the image collected is handled using image recognition algorithm, and the abnormal feature on the inner wall of ampoule is identified;S2a, boundary distinguishes, to the abnormal feature identified, by analyzing its and surrounding area's gray level co-occurrence matrix and texture feature, distinguish abnormal feature itself edge and liquid film edge, retain abnormal feature independent contour data;S3, based on the judgment of edge feature, abnormal feature is edge detection and analysis.The application has the advantage of improving the accuracy rate of plastic particle identification.

The present application discloses a training method, application method and related system for a waveform data processing model, which relates to the fields of environmental radioactivity detection and particle physics technology. The training method includes obtaining raw waveform data and its corresponding real particle label data through a TPC detector; performing downsampling processing and data compression processing on the raw waveform data in sequence; inputting the pre-processed raw waveform data into each channel of the waveform data processing model according to the spatial dimension of the channel in which it is located, determining the network loss based on the predicted probability values ​​of each category of the output particles and the real labels of the particles; updating the network parameters of the waveform data processing model through back propagation based on the network loss, and obtaining a trained waveform data processing model after multiple iterations. Using the trained waveform data processing model for particle identification not only saves data storage and reduces the amount of calculation, but also improves the accuracy of the multi-dimensional information particle identification results.

This invention proposes a multi-frequency electromagnetic feature fusion method for abrasive particle identification, involving signal processing, sensor information fusion, and intelligent monitoring of mechanical conditions. It solves the problems of measurement distortion caused by hardware parasitic parameter coupling, the inability to directly solve the highly nonlinear time-harmonic field model analytically, and the poor convergence of conventional optimization algorithms leading to misjudgment of abrasive particle parameters in existing abrasive particle identification methods. This invention constructs a forward analytical model of the time-harmonic field and a nonlinear objective function, and utilizes the Levenberg-Marquardt (LM) optimization algorithm to jointly invert and extract the equivalent diameter, conductivity, and relative permeability of unknown metallic abrasive particles in a multi-dimensional parameter space. This invention can accurately decouple the equivalent diameter, conductivity, and permeability of abrasive particles, accurately distinguish materials, and the LM algorithm converges quickly and does not diverge, achieving millisecond-level inversion, meeting the high precision and real-time requirements of online monitoring of industrial oil.

In some examples, an apparatus may include an environmental sample analyzer that is executed by at least one hardware processor to analyze an environmental sample at a plurality of wavenumbers. A particle type identifier that is executed by the at least one hardware processor may isolate, based on the analysis of the environmental sample at the plurality of wavenumbers, particle types for a plurality of particles in the environmental sample.

The present invention relates to the field of intelligent detection of space environment, and particularly to a particle discrimination method and system based on a one-dimensional convolutional neural network. The method of the present invention includes: preprocessing the waveform data of various particles in space collected to obtain the time-domain waveform data and frequency-domain waveform data of various particles; simultaneously inputting the time-domain waveform data and frequency-domain waveform data of various particles into a pre-established and trained particle discrimination model to obtain a particle discrimination result; the particle discrimination model is established based on a one-dimensional convolutional neural network. The system of the present invention includes a data preprocessing module and a particle discrimination module; the data preprocessing module preprocesses the waveform data of various particles in space to obtain time-domain waveform normalized data and frequency-domain waveform normalized data; the particle discrimination module simultaneously inputs the time-domain waveform data and frequency-domain waveform data into a pre-established and trained particle discrimination model to obtain a particle discrimination result.

The present invention belongs to the field of fluid measurement technology and proposes an integrated measurement method for water depth distribution and three-dimensional surface velocity field in urban flooded areas. The method comprises: using two cameras to capture the dynamic water surface of flood propagation in the flooded area from different perspectives, extracting, matching and removing errors from images captured by the two cameras at time t and time t+1; identifying tracer particles in the images at time t and time t+1, matching the tracer particles in the images at time t and time t+1 respectively by using coordinate fitting and a matching probability method, obtaining the three-dimensional coordinates of the feature points and the tracer particles at time t and time t+1 respectively, and jointly constituting the water depth distribution at time t and time t+1; and calculating the three-dimensional velocity distribution of the tracer particles at time t by combining the three-dimensional coordinates of the tracer particles at time t and time t+1 and the identification and matching of the tracer particles at time t and time t+1 by the same camera, thereby realizing the integrated measurement of water depth distribution and three-dimensional surface velocity field in urban flooded areas.

The invention discloses a micro-plastic particle identification method based on deep learning and application. The method comprises the following steps: receiving an image to be analyzed; inputting a to-be-analyzed image into the micro-plastic particle recognition model to carry out feature extraction on the to-be-analyzed image, and generating a result map for marking the positions of the micro-plastic particles; and extracting the coordinates of the positions of the marked micro-plastic particles in the result map, drawing a bounding box on the image to be analyzed, obtaining the image marked with the micro-plastic particles, and obtaining the information of the micro-plastic particles based on an image processing technology. Through the technology, the micro plastic particles in the image can be automatically identified, the micro plastic particles existing in the image can be accurately identified and analyzed, detailed data of the micro plastic particles are provided, the detailed data comprise shape, size, number and other information, a user is helped to better understand distribution of micro plastic pollution, and the user experience is improved. Therefore, the distribution and development trend of micro-plastic pollution can be tracked more comprehensively, and an effective micro-plastic pollution prevention and control strategy can be formulated.