48 results about "Cloud particle" patented technology

The invention provides a physical response data analysis method and system based on aircraft catalytic operation, and the method comprises the steps: obtaining three-dimensional cloud particle distribution data sets of a catalytic operation region before and after the catalytic operation through airborne cloud particle detection equipment, the three-dimensional cloud particle distribution data set comprises particle characteristic data before catalysis and particle characteristic data after catalysis; performing morphological feature extraction processing on the pre-catalysis particle feature data and the post-catalysis particle feature data by adopting a cloud particle AI image shape recognition algorithm to generate a pre-catalysis particle morphological difference feature set and a post-catalysis particle morphological difference feature set; constructing a dynamic physical response evaluation model, inputting the particle morphological difference feature set into the dynamic physical response evaluation model to execute multi-dimensional response correlation analysis, and generating a dynamic response feature set of the catalytic operation; and generating a catalytic strategy optimization instruction set according to the dynamic response feature set, and transmitting the catalytic strategy optimization instruction set to the airborne catalytic control system for catalytic parameter adjustment operation.

The invention discloses a wing icing full-period sensing and early warning method, and belongs to the field of aircraft icing detection and early warning. The method comprises the following steps: firstly, acquiring cloud particle data, flight parameter sensing data and planar electrode icing detection data; then, the cloud particle data and the flight parameter data are preprocessed, and the impact efficiency of the cloud particles impacting the surface of the wing and the icing mass flow rate are obtained; and finally, constructing a mixed model based on a BP neural network and a recurrent neural network, and processing the time sequence data through a long-short-term memory unit to realize prediction of the icing condition. According to the invention, a multi-scale dynamic cloud field in real flight can be monitored, the ice thickness and the ice type of predicted icing can be output in real time, and at the same time, multi-stage early warning thresholds are set to realize gradient early warning output of potential risk prompts; the icing prediction result correction in the real-time flight process can be realized, and the prediction accuracy is improved; the method has the advantages of high-precision prediction, real-time monitoring, strong generalization and the like, and is suitable for icing safety protection under complex meteorological conditions.

The invention provides a dual-channel cloud particle image recognition method and system based on transfer learning and UNET. The method comprises the following steps: acquiring a cloud particle image set; performing first feature extraction processing and second feature extraction on each cloud particle image in the cloud particle image set to obtain a first channel feature map and a second channel feature map corresponding to each cloud particle image; inputting the first channel feature map and the second channel feature map into a pre-trained feature adaptation model for feature space alignment to obtain an aligned first channel feature map and an aligned second channel feature map; and inputting the aligned first channel feature map and the aligned second channel feature map into a UNET network to carry out dual-channel feature fusion and identification so as to obtain a cloud particle identification result. According to the invention, the accuracy of the cloud particle identification result is improved.

The invention relates to a deep learning cloud particle target detection method based on a dynamic detection head. According to the method, firstly, complexity evaluation is carried out on a cloud particle image data set, the weighted information entropy, the weighted texture complexity and the weighted local standard deviation of each image are calculated, and the comprehensive complexity is calculated based on the complexity indexes. Then, according to the overall complexity grade of the data set and the image data complexity, a detection head selection standard is formulated, and a proper dynamic detection head combination is selected according to the standard; and finally, performing target detection on the cloud particle image data set by using the selected dynamic detection head combination. Through the method, the most suitable detection head combination can be adaptively selected, the target detection precision and efficiency are effectively improved, and the method is particularly suitable for a large-scale cloud particle image data set with complexity difference.

The invention relates to a cloud particle data target detection method based on image complexity. The method comprises the following steps: firstly, calculating weighted information entropy, weighted texture complexity and weighted local standard deviation of data of each cloud particle image, and evaluating the complexity of the image data; and marking according to the complexity of the image data, further calculating the comprehensive complexity of the data group, and judging the uniformity degree of complexity distribution of the data group. On the basis, different splitting and combining strategies are adopted according to the overall complexity distribution uniformity and marks of the data sets, and it is ensured that the deep learning detection model can select the most suitable strategy according to the complexity. Through the method, the computing resources and the detection precision of the deep learning model can be optimized in the face of cloud particle image data with different complexities, the image processing efficiency is improved, and the method is suitable for large-scale cloud particle data sets with large complexity differences.

The invention relates to a multi-form central point fused cloud particle detection method, which mainly comprises the following steps of: firstly, performing morphological processing of different levels on a cloud particle image to generate a plurality of morphological processing results; then, for each piece of morphologically processed image data, extracting different types of center points, including a geometric center point, a minimum circumscribed rectangle center point, a maximum connected region center point, a contour centroid and a minimum circumscribed circle center point; then abnormal center point filtering and fusion are carried out on center points in each maximum connected area, and a final target center point position is obtained; and projecting the central points onto the original image data to generate a predetermined number of anchor frames, establishing a deep learning detection model to predict the anchor frames, and screening the anchor frames to obtain a detection result. According to the method, the cloud particle region detection positioning precision is improved, and a reliable basis is provided for cloud microphysical parameter inversion.

The invention relates to a weighted adaptive anchor frame cloud particle detection method, which mainly comprises the following steps of: firstly, carrying out different morphological processing on a cloud particle image, extracting multiple types of central points and carrying out first fusion; then, according to the expansion corrosion degree of morphological processing, different weights are given to each fused center point, the weight given to the center point with the larger expansion degree is lower, and the weight given to the center point with the larger corrosion degree is higher; projecting the weighted central points back to the original image, and performing secondary central point fusion and weight updating; and finally, adjusting the size of an anchor frame according to the weight of the central point, generating the anchor frame adaptive to the size of cloud particles, and establishing a deep learning model for detection. According to the method, through adaptive weight and anchor frame size adjustment, the precision of cloud particle target detection is remarkably improved, and the use efficiency of the anchor frame in the detection process is optimized.

The disclosed embodiment is about an adaptive strategy for the optimal photon number of a quantum interferometer radar under a stratocumulus cloud background. It includes: a QIR system generates detection photons and reference photons with preset photon numbers, and directs the detection photons toward the detection target; the QIR system receives the reflected detection photons; a mathematical model is established based on the influence of the scattering, absorption, and other characteristics of the liquid water content of the stratocumulus clouds in the atmospheric environment on the reflected detection photons, and the relationship between the liquid water content of the stratocumulus clouds, the number of radar transmitter signal photons, and the transmission distance of the detection photons is obtained; based on the relationship between the liquid water content of the stratocumulus clouds, the number of transmitter signal photons, and the transmission distance of the detection photons, the number of radar transmitter photons is adaptively adjusted so that the radar system can achieve the optimal average photon number. The disclosed embodiment effectively improves the robustness of the QIR system under the water content of stratocumulus cloud particles by adjusting the average photon number of the QIR system in real time.

The application discloses a cloud particle spectrum optimization method and system based on a mask autoencoder, a medium and equipment, and the method comprises the following steps: obtaining a cloud particle shadow image to be completed, performing pretreatment, and obtaining a pretreated image; based on a trained cloud particle image completion model, a reconstruction patch is performed to obtain a completed image; the completed image is evaluated and screened, the cloud particle geometric scale statistics and particle size grading of the completed image passing the evaluation are performed, the flight sweeping distance and the effective sampling width corresponding to each particle size grading are combined, the grading sampling volume and the grading concentration are calculated, and finally the cloud particle spectrum is obtained. The overall process of the method does not depend on cloud particle level artificial labeling, and only the original image is needed to complete the training and application, thereby reducing the data preparation cost, being suitable for popularization and application in large-scale airborne observation data, and increasing the number of available large cloud particle samples entering the cloud particle spectrum statistical link without increasing the hardware complexity.

The application discloses an ice wind tunnel non-uniform particle diameter multi-particle cloud simulation method, device, equipment and medium, and the method comprises the following steps: S1, grid division is performed on a calculation region in the ice wind tunnel; S2, air flow field calculation is performed by using the grid and convergence is calculated; S3, on the basis of the flow field calculation convergence, a discrete phase model is introduced, combined with a multi-nozzle incident source position, boundary characteristics of the non-uniform particle diameter multi-particle incidence, non-uniform particle diameter cloud multi-particle non-steady state motion heat transfer simulation is completed; S4, cloud particles are updated once when the flow field calculation in the iteration process is set to a step number; and S5, when the number of incident particles and the number of particles leaving the calculation domain reach a dynamic balance, if the evaluation of the particle diameter distribution and the cloud uniformity at the outlet section does not meet the requirements, the nozzle incident source position and the cloud particle boundary setting are redesigned, and the coupling iteration calculation of the cloud non-steady state is performed. The application can obtain more accurate distribution results of the cloud at each section position.

The present application relates to the technical field of liquid water content detection in air, and particularly relates to a method and device for detecting liquid water content in air by using a hot-wire water content meter. The method comprises obtaining first data related to the hot-wire water content meter and second data related to a cloud particle probe, wherein the hot-wire water content meter and the cloud particle probe are on the same detection platform; establishing a data set of physical parameters changing with time according to the first data and the second data; setting a cloud entering standard and calculating liquid water content according to a liquid water content calculation formula; calculating the liquid water content of each sub-class according to the sub-class of the cloud particle probe, and selecting the optimal value of the liquid water content. The present application solves the problems of error overlap caused by multi-parameter combination calculation in the laboratory empirical formula of the prior art, and the error caused by human operation.

The invention discloses a spherical particle size measurement method and device based on deep learning numerical prediction, and belongs to the field of particle measurement and image processing. The method comprises the following steps: acquiring interference fringe image data of spherical particles with different sizes, marking a corresponding particle size label for each image, and dividing the images into a training set, a verification set and a test set; the training set is used to train a pre-constructed neural network model, the verification set monitors the training process and stores the optimal model weight, and the test set tests the prediction precision of the model; the pre-constructed neural network model is based on an original UNet + + network and comprises a plurality of residual blocks and a space attention module, and a full-connection output head is added to realize end-to-end mapping from an image to a size value; and finally, inputting an interference fringe image of a to-be-measured particle into the trained network, and directly outputting a particle size prediction value. According to the method, high-precision, real-time and end-to-end prediction of the spherical particle size is realized, and the method is suitable for scenes such as cloud particle field on-line monitoring.

1. The name of the design product: Holographic cloud droplet spectrometer (LA24605). 2. The use of the design product: The design product is used for observing and analyzing the particle size distribution of cloud particles. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

This invention relates to the field of airborne cloud particle image processing technology, and provides a method and system for intelligent assessment and screening of airborne cloud particle image quality. The method extracts comprehensive quality features through multi-scale spatial and frequency domain joint analysis, employs a multi-task learning architecture to calculate image quality scores and defect types, determines a quality screening threshold based on score distribution characteristics and environmental parameters, retains valid images above the threshold and removes invalid images, and updates the learning architecture parameters based on the distribution of quality features of valid images and the clustering results of defect features of invalid images. This invention improves the accuracy and adaptability of cloud particle image screening.

The invention relates to the technical field of cloud particle measurement and analysis, in particular to an airborne cloud particle measurement data analysis system based on multi-source data fusion, and the system can analyze the cloud particle measurement data according to a vector real-time key index which is calculated in real time through a dynamic adjustment mechanism of dynamic reasoning key vectors. The most critical vectors for judging the playable area in different scenes are adaptively screened out, and the limitation of traditional fixed vector input is broken through, so that the model can accurately adapt to complex changes of diversified cloud scenes; through mutual cooperation of the core key vector and the dynamic reasoning key vector, on the premise that the core key vector guarantees the basic precision of playable area analysis, the dynamic reasoning key vector only brings the key vector in a real-time scene into calculation, redundant vectors do not need to be processed, the analysis precision is guaranteed, the efficiency of playable potential analysis is improved, and the accuracy of playable area analysis is improved. And good balance between precision and efficiency is realized.

The invention discloses a cascade center point fusion cloud particle detection method. The invention relates to a cloud particle detection method based on cascade center point fusion, and the method mainly comprises the steps: carrying out the morphological processing of different levels on a cloud particle image, and extracting a plurality of types of center points; then, carrying out first-stage central point screening and fusion to obtain a fusion central point in each processed image; projecting the central point of the first-stage fusion to the original image, and performing second-stage screening and fusion to further improve the positioning precision of the central point; and finally, generating an anchor box based on the fused central points, and realizing target detection of cloud particles in combination with a deep learning target detection model. According to the method, through a cascade fusion strategy, the defect of positioning errors under processing of different scales and forms in a traditional method is overcome, and the precision and stability of cloud particle target detection are improved.

The ice wind tunnel cloud and mist particle measuring device is suitable for testing an incoming flow cloud and mist field in a wind tunnel test section, a first hole and a second hole are formed in corresponding positions of two side faces in the length direction of the wind tunnel test section, and the ice wind tunnel cloud and mist particle measuring device comprises a first shell, a second shell and a controller, a first cylinder is arranged in the center area of the first shell, a second cylinder is arranged in the center area of the second shell, the first cylinder and the second cylinder are each of a hollow structure with the two ends open, laser beams are scattered after irradiating an incoming flow cloud field, water drops of the irradiated part in the cloud field shield laser to form projection, and a projection image is shot by the camera and fed back to the controller; and the detection is completed by the controller, so that the detection or test efficiency is improved.

The invention discloses a virtual cloud generation method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out cloud modeling through employing a three-dimensional substrate, and obtaining a cloud particle model; coloring the cloud particle model by using a preset shader to obtain a colored cloud particle model; performing cloud generation according to the basic fusion ball to obtain fusion ball clouds; and filling the fused cloud with the colored cloud particle model by using preset particle parameters to obtain a target virtual cloud. Thus, through the technical scheme, the coloring particle model and the fusion ball cloud can be automatically generated, then the target virtual cloud is obtained according to the coloring particle model and the fusion ball cloud, manual manufacturing is not needed, then consumption of manpower resources and time resources can be effectively reduced, and the working efficiency of cartoon style cloud manufacturing is further improved.

The invention discloses an icing wind tunnel non-uniform particle size multi-particle cloud simulation method, device and equipment and a medium. The method comprises the following steps: S1, carrying out grid division on a calculation region in an icing wind tunnel; s2, performing air flow field calculation by using the grid and calculating convergence; s3, a discrete phase model is introduced on the basis of flow field calculation convergence, and non-uniform particle size cloud and mist multi-particle unsteady state motion heat transfer simulation is completed in combination with the position of a multi-nozzle incident source and the boundary characteristics of non-uniform particle size multi-particle incidence; s4, updating cloud and mist particles once when a set step number is calculated in an iteration process flow field; and S5, when the number of incident particles and the number of particles leaving the computational domain reach dynamic balance, if the particle size distribution and the cloud uniformity of the section particles at the outlet are evaluated not to meet the requirements, redesigning the position of the incident source of the nozzle and the cloud particle boundary setting, and carrying out unsteady-state coupling iterative calculation of the cloud. According to the invention, a more accurate distribution result of the cloud and mist at each section position can be obtained.

The utility model provides an airborne holographic cloud particle measuring instrument, which comprises a main control part and a particle monitoring and collecting part, and is characterized in that the main control part comprises a cylindrical pod, a conical pod rear cover fixed at the end part of the rear end of the cylindrical pod and a spherical front cover fixed at the end part of the front end of the cylindrical pod; an optical terminal unit and a main control circuit board are assembled in the rear end of the cylindrical pod; the particle monitoring and collecting part comprises a strip-shaped ellipsoidal shell; the strip-shaped ellipsoidal shell is suspended at the front end of the cylindrical pod through a transition connecting rod. The particle screening device is simple in structure and convenient to use and maintain, particularly, the overall modeling is streamline, the phenomena of wind resistance and particle breaking are greatly reduced, and the particle screening device has high timeliness, accuracy and reliability.

This invention provides a method and apparatus for hail cloud attenuation based on lidar detection feedback, belonging to the field of meteorological control technology. The method includes: registering and correcting joint detection data from weather radar and optical lidar; identifying the spatial distribution characteristics of hail-containing layers based on the registered and corrected joint detection data to generate spatial positioning data for laser control; acquiring the charge density distribution of the target cloud, establishing a safety window constraint matrix corresponding to the charge density gradient, and defining the laser path and laser power parameters; performing laser irradiation operation according to the laser path and laser power parameters, and updating the echo time difference compensation model and the safety window constraint matrix based on the lidar feedback signal. This invention achieves adaptive adjustment of laser irradiation parameters according to changes in cloud particle size and charge, thereby safely and accurately attenuating hail layers.

The present invention provides a cloud physics parameter inversion method based on a convolutional neural network, comprising the following steps: S1, calculating the scattering characteristics of cloud particles in the terahertz frequency band, combining the particle spectrum distribution, simulating and calculating the functional relationship between the terahertz radar equivalent reflectivity factor and the cloud physics characteristic parameters, and establishing a forward physical model; S2, preprocessing the measured data and microphysical parameter products of the terahertz cloud radar to obtain a training database; S3, establishing the structure of the convolutional neural network, using the forward physical model as a constraint, and adaptively adjusting the learning rate based on the deviation between the predicted value and the actual value of the radar echo; S4, repeatedly training the convolutional neural network using the training database, and obtaining the cloud physics characteristic parameters by inverting the terahertz radar equivalent reflectivity factor. This method improves the inversion accuracy of cloud physics parameters, eliminates the reliance on empirical values, and is an algorithm with stronger generalization and adaptive learning.

The invention discloses a cloud particle spectrum optimization method and system based on a mask auto-encoder, a medium and equipment, and the method comprises the steps: obtaining a to-be-complemented cloud particle shadow image, and carrying out the preprocessing, and obtaining a preprocessed image; performing patch reconstruction based on the trained cloud particle image completion model to obtain a completed image; and evaluating and screening the complemented image, carrying out cloud particle geometric scale statistics and particle size grading on the evaluated complemented image, and calculating grading sampling volume and grading concentration in combination with the flight scanning distance and the effective sampling width corresponding to each particle size grading to obtain a final cloud particle spectrum. The whole process of the method does not depend on cloud particle level manual annotation, training and application can be completed only through an original image, the data preparation cost is reduced, the method is suitable for application and popularization in large-scale airborne observation data, and the number of available large cloud particle samples entering a cloud particle spectrum statistical link is increased on the premise that the hardware complexity is not increased.

The invention relates to a cloud particle image data complexity judgment method, which comprises the following steps of: firstly calculating weighted information entropy, weighted texture complexity and weighted local standard deviation of data of each cloud particle image, then evaluating the comprehensive complexity of the cloud particle image data by synthesizing the weighted information entropy, the texture complexity and the local standard deviation, and finally judging the complexity of the cloud particle image data. And the complexity is marked as low, medium or high. According to the method, a complexity voting mechanism is further designed, different complexities are comprehensively evaluated through multiple rounds of weighted voting, and finally the complexity level of data of each image is determined. The method can effectively process the complex scene in the cloud particle image, improves the processing precision and efficiency of the complex image, does not depend on target detection, and is suitable for rapid evaluation of large-scale cloud particle image data.

The invention belongs to the technical field of atmospheric environment monitoring and cloud and mist micro-physical observation, and particularly relates to a light-weight cloud and mist vertical structure comprehensive sounding device which comprises a device body. The temperature and humidity measuring module is arranged on the outer side of the device main body; the water vapor measuring module is arranged on the outer side of the device main body; the holographic particle measurement module is arranged on the outer side of the device main body; a navigation positioning module; and the data processing system is arranged in the device main body and is used for receiving and storing temperature, humidity, air pressure, water vapor content and cloud and mist particle holographic image data. The device realizes synchronous measurement of cloud temperature, humidity, air pressure, water vapor and cloud particle microphysical parameters, is simple in structure, and can be carried on platforms such as a sounding balloon and an unmanned aerial vehicle to realize flexible and refined measurement of a cloud vertical structure; the method has a wide application prospect in the fields of cloud rainfall physical scientific observation, artificial cloud and mist reduction effect evaluation, cloud parameter remote sensing product quantitative inspection and the like.

The embodiment of the invention discloses a customer management method, device and equipment. The method comprises the following steps: receiving a customer management request for a target customer; in response to the client management request, determining a candidate operation strategy corresponding to the target client based on the client value contribution, the client value change and the business stickiness of the target client; determining a model by using a pre-trained strategy, and selecting a target operation strategy corresponding to the target customer from the candidate operation strategies; wherein the strategy determination model is constructed based on a preset cloud drift algorithm, and the cloud drift algorithm is used for searching a global optimal solution and a local optimal solution in a preset search space by using cloud particles related to the candidate operation strategy based on a preset fitness function, and determining the candidate operation strategy based on the global optimal solution and the local optimal solution. Determining the target operation strategy; and performing customer management on the target customer based on the target operation strategy.

The present invention provides a dual-channel cloud particle image recognition method and system based on transfer learning and UNET. The method comprises obtaining a cloud particle image set; performing first feature extraction and second feature extraction on each cloud particle image in the cloud particle image set to obtain a first channel feature map and a second channel feature map corresponding to each cloud particle image; inputting the first channel feature map and the second channel feature map into a pre-trained feature adaptation model for feature space alignment to obtain an aligned first channel feature map and an aligned second channel feature map; and inputting the aligned first channel feature map and the aligned second channel feature map into a UNET network for dual-channel feature fusion and recognition to obtain a cloud particle recognition result. The present invention improves the accuracy of cloud particle recognition results.

1. The name of this design product: Small-particle cloud droplet spectrometer (24606). 2. Purpose of this design product: used for measuring airborne holographic cloud particles. 3. The key point of the design of this product lies in its shape. 4. The picture or photo that best illustrates the design points: three-dimensional picture.

The embodiment of the invention provides a hail cloud weakening method and device based on laser radar detection feedback, and belongs to the technical field of meteorological regulation and control. The method comprises the steps of performing registration correction on joint detection data based on the joint detection data of a weather radar and an optical laser radar; based on the combined detection data after registration and correction, identifying spatial distribution characteristics containing hail layers so as to generate spatial positioning data for laser regulation and control; acquiring charge density distribution of the target cloud body, establishing a safety action window constraint matrix corresponding to a charge density gradient, and limiting a laser action path and a laser power parameter; and executing laser irradiation operation according to the laser action path and the laser power parameter, and updating the echo time difference compensation model and the safety action window constraint matrix based on a laser radar feedback signal. According to the scheme of the invention, self-adaptive adjustment of laser irradiation parameters along with changes of cloud particle sizes and charges is realized, so that the hail layer is weakened safely and accurately.

The invention relates to a system and a method for SLD particle phase state identification. In the technical scheme of the invention, a feature matrix of cloud particle image data can be firstly constructed based on CIP detection data, image binarization is performed on the constructed feature matrix, and a binarized image of cloud particles is extracted. A circle proximity calculation may then be performed on the binarized image to determine whether the cloud particle is an SLD particle.