77 results about "Visualization algorithms" patented technology

The invention discloses a real-time large-scale terrain visualization implementation method based on a kd tree. The real-time large-scale terrain visualization implementation method comprises the following steps: partitioning terrain according to a point cloud space; off line establishing an LOD (level of detail) model by utilizing kd tree layering, wherein the established LOD model is responsible for obtaining multi-resolution model data by the terrain point cloud data, and switching each resolution ratio of the model; loading the off line constructed LOD model data into a memory by utilizing an external storage algorithm when on line, wherein the external storage algorithm is responsible for scheduling the externally stored LOD model data to the memory; utilizing a simplified terrain eliminating technology to the LOD model data in the memory, wherein the simplified terrain establishing technology is responsible for eliminating invisible LOD model data, so as to reduce the amount of the data transmitted to graphical hardware; and further reducing the data mount and calculation amount of the eliminated data transmitted to the graphical hardware by utilizing a three-dimensional engine optimization technique. The real-time large-scale terrain visualization implementation method based on the kd tree is simple to realize and high in display efficiency and has excellent visualization effect based on a triangular irregular network visualization algorithm, so that the visualization speed is improved.

The invention relates to a three-dimensional visualized algorithm for maritime oil spill, which is realized in a way that: on the basis of drawing a large-scale sea level scene, dividing an oil spill process within a short time scale into an oil film self-spreading stage and an oil film drift diffusion stage, and calculating the oil spill loci of the two stages respectively by selecting an Lehr model and an oil particle model; simulating the oil particle model by using a particle system; and visualizing the oil spill loci on the simulated sea level by utilizing a planar refraction technology, accessing the veins in a fragment shader of a GPU by projecting vein coordinates, and simulating and reproducing the three-dimensional scene of maritime oil spill while drawing the sea level refraction effect. The algorithm solves the problems of simulation of the oil particle model, visualization of the oil spill loci on the simulated sea level, and the like, and realizes the vivid and visual simulation of the three-dimensional movement change process of oil spill within a short time scale.

The invention relates to the meteorological field. In order to improve the intuitiveness of meteorological data, the inside structure of a monomer is reproduced by aid of two-dimensional image information through certain three-dimensional visualization algorithm and enabled to be three-dimensional, and potential information of meteorological radar base data is excavated to the maximum so as to provide an initiative data display mode and relevant meteorological features for meteorological research activities. According to the technical scheme of a three-dimensional reconstruction method based on the meteorological radar base data, the method includes the following steps: (1) extraction and sampling of fault contour lines, which include binaryzation processing on images, performance on boundary smooth denoising through opening operation, small area removal through zone marking, and boundary tracking for boundary extraction and chord sampling for boundary sampling; (2) three-dimensional modeling of the monomer, which includes contour correspondence between faults, contour splice between the faults, interpolation and furcation processing between the faults; and (3) three-dimensional rendering and display based on an open graphics library (GL). The method is mainly applied to the meteorological field.

The invention belongs to the field of scientific visualization, and particularly relates to a GPU-based interactive ocean three-dimensional flow field dynamic visual algorithm. The GPU-based interactive ocean three-dimensional flow field dynamic visual algorithm is based on the mode data of a regular grid. A uniform seed point placement algorithm in a three-dimensional flow field space is adopted. Furthermore a density ball which is designed by means of a Gauss function is mapped to multiple patterns for realizing uniform distribution of the seed points. Finally a streamline is generated by means of GPU accelerated rendering technology. Then the streamline is expanded to a pipe line by means of Tessellation technology, thereby realizing dynamic visual drawing for the ocean three-dimensional flow field. For improving spatial perception for the three-dimensional streamline, a Phong/Blinn illumination model is used for performing illumination processing on the streamline. Furthermore the GPU-based interactive ocean three-dimensional flow field dynamic visual algorithm is realized based on the GPU and has higher streamline generating efficiency and higher control efficiency.

The invention belongs to the technical field of information processing, and discloses an in-vehicle environment monitoring and early warning method based on cloud service and a depth neural network. When an individual enters a vehicle, information is detected by a human body infrared detection module. The detected real-time data is transmitted to an intelligent interconnected cloud platform through a 5G wireless network, a database server of the intelligent cloud performs cloud storage and backup operations, and a content management server performs cloud computing and big data cluster analysisdata mining operations. According to the method, the intelligent cloud is utilized to monitor the indoor air of the automobile in real time, and a proposed alarm information level classification algorithm based on a logic regression optimization model is utilized at a cloud end to accurately classify the danger level of the automobile owner, and a decision is made by using a designed system self-adaptive deep reinforcement learning algorithm; at the cloud, a proposed Pyechars-Based visualization algorithm is used for statistical analysis of mass data and independent and safe visualization services are provided for all vehicle owner users.

The invention relates to the field of neuroelectricity physiological information data processing and provides a system for comprehensively processing massive multi-channel nerve information. The system comprises a script-driving control component, a data obtaining control component, a data analysis processing component and a data visualization processing component. Collection, processing and visualization processes of multi-channel nerve information undergo componentization and integration so as to establish a uniform service interface; and the processing process of the massive multi-channel nerve information can be automatically finished in terms of batches under the control of a script, and a visualization result can be provided. Simultaneously, the procedures for processing neuroelectricity physiological information data can be stored in a storage medium, thereby improving the repeatability and controllability of data processing. The invention also has the advantages that hardware control and the data processing are combined together; according to the processing result of neuroelectricity physiological information data, hardware operation is controlled, and an experiment proposal is dynamically revised; and due to the uniform interface, expansion processing and visualization algorithm can be dynamically carried out.

The invention relates to construction and application of a multi-dimensional superimposed color mixing model and a gradient chromatography matrix algorithm. A brand-new digital color matching logic design is adopted to construct a pigment discrete color mixing model and a visualization algorithm of a color mixing chromatography of the pigment discrete color mixing model, and digital virtual color matching of color yarns is realized. In actual application, standard color mixing chromatography can be constructed by combining and blending five primary color fibers such as yellow, cyan, fuchsin, black and white fibers, a reference basis can be provided for color matching of color yarns, and a digital model of pigment color matching is specifically constructed; an algorithm and chromatographic big data of serialized mixed chromatography based on pigment blending are constructed, and a complete hue based on pigment color blending and a full chromatographic system with different lightness and saturability are included; a color gradient mode based on pigment blending and gradient chromatography big data are constructed; and compared with a traditional color matching scheme, the obtained color matching is visualized, and the working efficiency of actual chromatography construction is improved.

The invention provides a malicious code homologous judgment method based on deep learning. The method comprises the following steps: disassembling the to-be-judged malicious code by using an IDA tool,obtaining the core binary content therein through regular expression matching, and removing useless information at the same time; receiving the binary content as the input, mapping the binary contentas the malicious code image by using the malicious code visualization algorithm; training a deep learning model convolutional neural network by adopting a sample set composed of the malicious code image and a label value, thereby obtaining a mature judgment model; and receiving the to-be-judged malicious code as the input to accomplish the homologous judgment. The homogeneous judgment task is converted into an image classification task through the malicious code visualization algorithm, and the available malicious code homogeneous judgment method is realized by combining the deep learning judgment model. And the homogeneous judgment technology with higher judgment accuracy rate in comparison with the existing system is realized.

The invention belongs to the field of scientific visualization, and particularly relates to a dynamic two-dimensional vector field streamline visualization algorithm based on a GPU (Graphics Processing Unit). The algorithm comprises the following steps: carrying out random distribution on particles in a vector field so that the particles move along with the vector field, and utilizing a geometry shader mechanism of the GPU to generate streamlines; generating the probability density graph of the vector field by calculating a radial basis function taking the particles as the center so as to control the generation, extinction and multiresolution distribution of the particles. According to the algorithm, the complete picture of the vector field is displayed in a streamline mode, so that the density, length, motion speed and smooth degree of the streamlines can be changed in real time; the streamlines are continuous and independent from one another, so that complicated characteristics of the vector field can be well dynamically displayed, and the cognitive purpose of visualization is achieved. Moreover, the algorithm is realized based on the GPU, and the generation and control efficiency of the streamlines is higher.

The invention belongs to the crossing field of physical oceanography, computer graph and image processing and oceanography visualization, and specifically relates to a mesoscale vortex visualization algorithm based on a Morphing technique. According to the method, on the basis of track data obtained by tracking mesoscale vortexes, specific to each track, at the adjacent vortex positions, firstly, the original boundaries and target boundaries of the vortexes are resampled by using the Morphing technique, thus obtaining feature points; then corresponding positions among the feature points are determined by using Hausdorff distance matching; transform relationship among the feature points of the vortex boundaries is established by using a least square method; the feature points of the vortexes at middle moments are obtained according to the transform relationship; the line segments among the feature points at the middle moments are obtained through linear interpolation; finally the original information of the vortex boundaries and the middle information of the vortexes obtained through Morphing are organized and stored; and further visualization expression is carried out on the shape evolution in the vortex moving process. The technical support is provided to track the vortexes and know the growth, decline and fall shape evolution of the vortexes well.

The invention discloses a reactor design method based on a visualization algorithm. According to the method, the reactor iron core magnetic path length l, the reactor iron core cross section S and the reactor coil turn number N are used as three variables, the value of the minimum volume V of a reactor is obtained under the limitation conditions that the inductance value L of the reactor and the volume V of the reactor meet the value of the inductance value L and simultaneously meet the coil turn number N and the iron core cross section S, and then, the reactor iron core magnetic path length l, the reactor iron core cross section S and the reactor coil turn number are obtained through combining an inductance value change trend figure and a volume change trend figure. The reactor design method has the advantages that the simplification problem is solved through the visualization algorithm, the design process is relatively visual, simple and convenient, and in addition, the function change condition can be conveniently observed.

The invention provides a big data visualized analysis and display framework and a construction method thereof. The visualized analysis and display framework comprises a visualized core and a data conversion access module, an algorithm structure access module, a parameter setting module and a UI display module which are coupled with the visualized core. The construction method includes: input data is processed by distributed parallel computation clusters and then saved in different data source sets in different formats; data in data sources is converted into target formats needed by a selected visualized algorithm; the algorithm selected from a visualized algorithm library is compiled according to algorithm parameters; compiled data is used for processing the data in the target format to obtain algorithm output data; the data change of the algorithm output data and appearance attribute parameters is monitored; corresponding functions are called to obtain new data, and the new data is used to redraw a visualized area pattern.

Texture data are important constituents of a three-dimensional model and greatly affect the visualization effect on the three-dimensional model; but the texture data volume of the three-dimensional model is large and is generally a bottleneck for increasing the model visualization speed. The invention provides a three-dimensional model texture coloring method based on image partitioning. According to the method, the texture data of the three-dimensional model are processed, and the surface of the three-dimensional model is partitioned into parts (blocks with the same colors) with similar colors, thus eliminating image textures and greatly reducing the data volume. Furthermore, the adjacent blocks with the same colors are subjected to combination operation to generate texture trees of the model, and a texture coloring scheme with different detail levels is provided for different application scenes. By the three-dimensional model texture coloring method, the data volume of the three-dimensional model is greatly reduced, and the three-dimensional model visualization speed is increased.

The invention discloses a visualization management system based on distributed data mining algorithms. The system includes a data access analysis module, a data processing module, a visualization algorithm library module and a data output module. The data access analysis module is a data input source of the management system, and accesses data from different sources. The data processing module carries out data cleaning and processing. For processed data, calling of the distributed data mining algorithms is carried out through the visualization algorithm library module, and a service application model is established. Data output is carried out on a data result, which passes through the visualization algorithm library module, through the data output module. According to the system, the use problem of a power network service big-data platform is solved, massive investment of manpower, material resources and financial resources of a power company is saved, repeated development of a programis avoided, work efficiency is improved, and a scene need of quickly building a big-data service application is met.

The invention relates to a permanent magnet motor rotor strength optimization design method based on multi-dimensional visualization. The method comprises the following steps: solving an equivalent Mises stress and permanent magnet pressure equation of a non-magnetic metal sheath by using an analytical method; constructing a four-dimensional feasible region of the equivalent Mises stress of the non-magnetic metal sheath and the pressure intensity of the permanent magnet, and taking an intersection to obtain a four-dimensional feasible region intersection; and performing dimensionality reduction processing on the four-dimensional feasible region intersection, after a three-dimensional static feasible region and a three-dimensional high-speed feasible region are obtained, taking the intersection, obtaining a three-dimensional feasible region intersection, and solving the sheath thickness and the interference magnitude. The influence of the sheath thickness, the interference magnitude and the rotor speed on the rotor strength is researched by applying a multi-dimensional visualization algorithm, intersection operation is performed on each multi-dimensional visualization graph, the feasible region of the sheath thickness and the interference magnitude meeting the strength requirement in the static state and the high-speed operation state is obtained, robust optimization is conveniently achieved, and the method has the characteristics of intuition and convenience.

Methods and systems of displaying flow-volume loops of a patient and variability of the flow-volume loops across measured breaths are disclosed. The methods comprise obtaining a physiological dataset of the patient on a data acquisition device, applying a smoothing and curve fitting algorithm to the physiological dataset on a processing device to obtain real-time volume and flow data at a plurality of time instances, applying a visualization algorithm on the processing device to the volume and flow data to create a series of flow-volume loops based on the volume and flow data, and outputting a plot of the flow-volume loops on a display device to aid evaluation or diagnosis of the patient.

The invention provides a network layout method and system based on a segmentation drawing method and a medium. The network layout method comprises the following steps: a graph segmentation step: segmenting a whole graph by using a heuristic algorithm according to original data to obtain a plurality of sub-graphs; an inter-block layout step: generating an inter-block layout of the sub-graph based on the force guidance model of the graph layout according to the segmentation result; a sub-graph layout step: respectively laying out the sub-graphs by using the force guidance model to obtain an optimal sub-graph layout; a splicing step: splicing the sub-images according to the inter-block layout of the sub-images to obtain an initial finished image; and an adjustment step: adjusting the layout of initial mapping according to the force guidance model to obtain an optimal full-graph layout. According to the method, the super-large-scale network is laid out based on the segmentation drawing method, so that the data volume capable of being processed by a network layout algorithm is greatly improved, and the million-order barrier of a network visualization algorithm is broken through.

The invention discloses a method for calculating temperature rise of a reactor based on a visualization algorithm. The length l of a magnetic path of an iron core of the reactor, the cross section S of the iron core of the reactor and the turn number N of a coil of the reactor are used as three variables, and a to-be-solved temperature rise value is obtained by a visualization figure at the premise of meeting the induction quantity and optimizing the volume. The method has the beneficial effects that the length l of the magnetic path of the iron core of the reactor, the cross section S of the iron core of the reactor and the turn number N of the coil of the reactor are used as the three variables, the temperature rise value of the reactor is obtained under the condition of meeting the requirements of the turn number N of the coil and the cross section S of the iron core, and the length l of the magnetic path of the iron core of the reactor, the cross section S of the iron core of the reactor and the turn number N of the coil of the reactor are obtained by utilizing a change tendency chart; the problem of unification is solved by the visualization algorithm, the design process is relatively visual, simple and convenient, and the change of a function is conveniently observed.

The invention provides a data resource arrangement method and device, computer equipment and a readable storage medium. The data resource arrangement method comprises the following steps: acquiring at least one corresponding data resource and all data structure models of each data resource from a resource library according to a data resource adding instruction; establishing at least one output object according to the output object adding instruction, and configuring an output structure model corresponding to each output object; processing all the data structure models and the output structure models through a preset visualization algorithm to generate corresponding data structure graphs and output structure graphs, and displaying the data structure graphs and the output structure graphs on a configuration page; acquiring a configuration relationship between a data structure graph and an output structure graph set by a user in the configuration page; and configuring a relationship between the corresponding data structure model and the output structure model according to the configuration relationship, and generating a corresponding data resource arrangement model. According to the invention, the threshold of data resource arrangement developers can be reduced, and the efficiency of data resource arrangement is improved.