272 results about "Local space" patented technology

The invention discloses an automatic welding device for a rotary alloy furnace tube, and belongs to the technical field of arc welding equipment.The automatic welding device for the rotary alloy furnace tube comprises a base and a U-shaped bin located at the top of the base, and the two ends of the interior of the U-shaped bin are jointly provided with a centering clamping mechanism used for clamping the outer side of the furnace tube; a driving mechanism for providing power for the centering clamping mechanism to clamp the furnace tube is arranged on the outer side of the centering clamping mechanism, and a welding mechanism is arranged on the inner top of the U-shaped bin. A local sealing design is adopted for the inflation mechanism, the annular air bag is attached to the inner wall of the furnace tube after being inflated, an independent cavity communicated with the outside only through a welding seam gap is formed near the welding seam, the protection requirement can be met only by filling the local space with argon, the whole furnace tube or a large-range external space does not need to be filled, and compared with traditional overall argon filling, the argon filling efficiency is greatly improved. The argon consumption is greatly reduced, the industrial gas purchase cost is remarkably reduced, and the production cost investment is reduced.

The invention discloses a dynamic inspection method and device based on an unmanned aerial vehicle, electronic equipment and a storage medium, and relates to the technical field of software and platforms or other related technical fields, and the method comprises the steps: carrying out the three-dimensional scene modeling of a to-be-inspected region through the space-time alignment data obtained through preprocessing; acquiring an inspection operation instruction, and performing inspection task distribution of multiple flight constraints on the dynamic environment model obtained by modeling to obtain an optimized task execution scheme; and performing global path planning and local space-time cooperative path optimization on the task execution scheme based on the high-precision three-dimensional scene model obtained by modeling to obtain a target inspection path, thereby remotely controlling the target unmanned aerial vehicle to perform optimized path inspection and anomaly analysis according to the target inspection path, and generating an anomaly inspection report of the target inspection area. According to the invention, the technical problems of poor adaptability and low accuracy of the inspection result of the inspection mode of planning the inspection path based on historical data in the prior art are solved.

The invention belongs to the technical field of computer vision, deep learning and time-space prediction, and discloses an end-to-end time-space prediction method based on improved three-dimensional rotation position coding, which is suitable for various time-space sequence prediction scenes such as weather, traffic flow and the like. According to the invention, through four key improvements, a position coding mechanism is optimized; three-dimensional coding proportions of time, height and width are dynamically adjusted so as to adapt to different scenes; fusing the absolute time and the relative space position, and strengthening local space-time correlation modeling; the position information directly guides attention calculation, and the fusion with an Attention module is deepened; and a rotation matrix cache mechanism is introduced to reduce redundant calculation. Meanwhile, the model is matched with a Patch embedding layer, an adaptive Transform encoder and an MLP de-wharf, a complete link of'feature embedding-position encoding-space-time fusion-prediction output 'is constructed, and the precision, generalization and reasoning efficiency of space-time prediction are effectively improved.

The invention discloses a remote photoplethysmography (rPPG) signal processing method and system based on a long and short term space-time convolution network, and belongs to the crossing field of biomedical signal processing and computer vision. In order to solve the problem of signal distortion caused by illumination fluctuation, motion artifacts and skin color differences, the method constructs a multi-scale space-time modeling framework: extracting local space-time features of transient changes of facial capillaries by adopting a 3D convolutional network, capturing long-range periodic features of heart rate rhythm in combination with a 1D expansion convolutional network, and establishing a multi-scale space-time modeling framework; the spatial-temporal characteristics are dynamically fused through the self-attention weight and the gating residual structure, and the anti-interference capability is improved. In the preprocessing stage, a face area is positioned through MTCNN, motion artifacts are compensated by using an optical flow equation, and signal purity is enhanced by combining a skin color mask and a YUV color space. The system adopts a deep separable convolution and parallel acceleration strategy to realize light weight, and optimizes the network through time domain MSE loss and frequency domain KL divergence. Experiments show that the phase error of the method is reduced by 40% in a dynamic scene, the signal amplitude of a deep skin color group is improved by 60%, the method is suitable for non-contact health monitoring equipment, and the robustness and the measurement precision of the rPPG technology in a complex environment are remarkably improved.

The invention discloses a human body behavior recognition method and system based on space-time decoupling multi-scale feature fusion. The system comprises a local space-time multi-scale feature fusion network MDCNN and a space-time decoupling network DTST. Extracting different scale features of multichannel sensor data by using a multi-scale convolutional neural network, extracting shallow layer features and deep layer features of signals at the same time, and extracting features representing action information by fully utilizing complementary information among different level features; the method comprises the following steps of: decoupling a network into a time decoupling structure and a space decoupling structure through a time-space decoupling network based on Transform, so as to accurately extract time-space characteristics in an inertial signal; and finally, using a Softmax classifier to carry out action classification and recognition. According to the method, consistency and complementarity among multi-sensor data can be efficiently utilized, multi-scale spatial-temporal characteristics are fused, changes in time dimensions and relevant information in spatial positions are fully considered, different attention degrees are kept for signals of different sensing channels, and the recognition precision in human behavior recognition is improved.

The invention provides a laser, inertia and image recognition combined calibration method and system. Comprising the following steps: setting a parameter calibration checkerboard on a marble platform for detection, and leveling the parameter calibration checkerboard; carrying out inertial navigation IMU static parameter calibration and inertial navigation IMU dynamic parameter calibration on the inertial navigation IMU, and correcting an inertial navigation IMU error; irradiating laser spots of a laser into the parameter calibration checkerboard, shooting the parameter calibration checkerboard from different angles through a machine vision camera, and obtaining a parameter calibration checkerboard photo set; and joint calibration of parameters of a machine vision camera, an inertial navigation IMU and a laser is carried out, and automatic correction of external parameters of various precise positioning measuring instruments and high-precision measurement of local space are realized.

The invention discloses a VAGCN-based complex industrial process operation state evaluation method. The method comprises the following steps: collecting process data and carrying out standardization processing; performing data division on the data by using a sliding window to obtain a time sequence; the method comprises the following steps: converting time sequence data into cross-time-space variable interaction graph structure data through a variable interaction graph structure, and then capturing variable interaction characteristics and local space-time dependency in each time window by using a moving window; time pooling is used, advanced features of variable interaction are extracted and then input into the operation state evaluation model, posterior probabilities of the online data belonging to different operation state evaluation grades are obtained, and a final evaluation result is a state grade corresponding to the maximum posterior probability at the current moment; and performing similarity analysis on the variable interaction graph structure in the non-optimal state and the optimality graph structure in the optimal state so as to evaluate the variable dynamic offset degree in the non-optimal state. According to the method, the interaction among the variables at different time points can be captured, and more accurate process operation state evaluation can be realized.

The invention discloses an industrial multi-source heterogeneous data feature fusion and dynamic modeling method and system, and relates to the technical field of industrial multi-source heterogeneous data processing, and the system comprises a data collection module, an event triggering type local space-time alignment module, a time sequence data set generation module, a feature fusion module and an industrial equipment state model generation module. According to the method, an event-triggered local space-time alignment module is arranged, an event-driven dynamic space-time anchoring mechanism is adopted, and in a preset tolerant time window, an image feature time sequence is constructed through industrial camera superframe sampling to be matched with sensor time sequence data, so that feature dislocation caused by sampling frequency difference is avoided; compared with an existing interpolation method, the non-linear coupling relation between the process parameter data and the sensor time sequence data is obtained, and the problem of feature fusion distortion caused by the time granularity difference is solved by performing non-linear interpolation on the process parameter data through Gaussian process regression and generating a continuous proxy curve synchronous with the sensor time sequence data.

The invention provides a 66kV main transformer load prediction method based on self-coding and multi-model fusion, and relates to the technical field of load prediction. The method comprises the following steps: adaptively determining an optimal cluster number by using an elbow rule, and reasonably grouping load data; by constructing a self-encoding network based on data correlation, compression and feature reconstruction are performed on data in each cluster. The invention provides a load prediction method fusing a BiLSTM (Bidirectional Long Short Term Memory) model, an Informer model and an improved STSGCN model. The BiLSTM is used for capturing local time sequence features of data, the Informer processes long sequence data to mine long-term dependency and global information in the data, and the improved STSGCN module is used for capturing local space-time correlation of the data. And splicing the output features of the three models, adaptively adjusting the contribution of each module through a gating unit so as to carry out feature fusion, and finally carrying out output prediction through a full connection layer.

The invention discloses a precipitation forecasting method based on Shearlet and L-PWC, the precipitation forecasting method is decomposed into a global space, a local space and a time angle, the time feature extraction effect is improved through the consistency evaluation of a reverse optical flow graph and a forward optical flow graph, the decomposition of a picture in a frequency domain is guided by using an optical flow with higher precision, and the time feature extraction efficiency is improved. Feature extraction is carried out on different frequency domains to obtain information to guide directional learning of multi-head attention, and more comprehensive global features are generated through assistance of other frequency band features. Extraction of local space information is guided by time information, feature enhancement of key information is completed, future weather is predicted through gradient decomposition of optical flow and time dependence before and after, and a litsea rotundifolia optimization algorithm is optimized through an ant colony algorithm, so that the algorithm can search for an optimal solution in a wider space, and the optimal solution is obtained. And the possibility that the algorithm falls into local optimum is greatly reduced, so that the ability of the model to deal with complex weather is improved.

The invention discloses a lightweight target detection method based on a local space mixed attention mechanism, and the method combines local average pooling, local average pooling removal and deep convolution, carries out the fine processing of a local region of an input feature map, achieves the self-adaption focusing of a key object and region in an image, and improves the detection precision. Therefore, the capturing capability of the model on important space information is obviously enhanced. According to the method, channel attention and space attention are adaptively learned under the condition that model parameters and calculation complexity are not remarkably increased, and the detection precision and the convergence speed are improved. The method is integrated in a YOLOv8 framework, and the accuracy and robustness of target detection are effectively improved while high frame rate real-time performance is kept.

The invention provides a trajectory clustering method based on space-time weighting and a density peak value, and relates to the technical field of trajectory clustering. The method provided by the invention comprises the following steps: based on a minimum description length criterion, fusing an adaptive space-time weight and a space-time geometric distance to segment trajectory data to obtain sub-trajectory segments; calculating a space-time local density, a relative distance and a decision value of each sub-track segment based on a density peak clustering algorithm; performing noise identification on the sub-track segments based on the average space-time geometric distance and the average local space-time density, iteratively selecting a class cluster center based on the decision values of the non-noise sub-track segments and the time-space factor decision values, and generating a class cluster set; and determining a candidate representative trajectory set from the class cluster set based on the spatio-temporal local density and the maximum density, extracting continuous sub-trajectory segments from the candidate representative trajectory set based on spatio-temporal continuity constraints, and sequentially connecting end points of the continuous sub-trajectory segments to generate a representative trajectory. According to the method, the trajectory clustering effect is improved through space-time weighted segmentation and density peak trajectory clustering.

The invention discloses a short-term power load prediction method and system based on a CNN-Transform hybrid model, and the method comprises the steps: carrying out the data collection of historical load data and meteorological data of a power system, carrying out the data preprocessing of the collected data, and carrying out the coding of a periodic time feature, and obtaining periodic time coding information; local space-time features of the load data and the meteorological data are extracted by using a convolutional neural network, and hierarchical expression of the features is realized through a multi-layer convolutional structure during extraction; the local spatiotemporal features and the periodic time coding information are fused to obtain fusion features containing a load sequence, the long-period dependency relationship of the load sequence is modeled through a Transform network, global modeling of the fusion features is achieved through a multi-head self-attention mechanism, and a CNN-Transform hybrid model is obtained; a CNN-Transform hybrid model is used for prediction, and a load prediction result is output; according to the invention, the precision of load prediction and the generalization ability of the model are significantly improved.

The invention relates to the technical field of sea wave height prediction, and discloses a significant wave height prediction method and system based on spatio-temporal evolution multi-scale feature extraction. The method comprises the following steps: performing significant wave height prediction on input significant wave height grid data of a sea area to be predicted by applying a trained hybrid heterogeneous parallel double-convolution dynamic network; the method further comprises the steps of obtaining the effective wave height grid data, inputting the effective wave height grid data into the hybrid heterogeneous parallel double-convolution dynamic network for effective wave height prediction, and then outputting an effective wave height prediction sequence of a future time step. According to the method, multiple features of SWH spatio-temporal dynamic evolution can be extracted, and hierarchical spatio-temporal features from a local scale to a global scale and local spatio-temporal dynamic crossing from a coarse scale to a fine scale can be mined at the same time; by constructing a heterogeneous parallel double-convolution framework, the internal irregularity of the SWH field and the irregular relation between SWH field data are overcome, and meanwhile the capacity of capturing the invariant relation in the SWH field is kept.

The invention provides a factory three-dimensional digital twin model reconstruction and incremental updating method for industrial space intelligence, and the method comprises the steps: obtaining a factory multi-view basic image for a factory panoramic scene, so as to train a bifurcated structure neural implicit field network containing a radiation field branch and an SDF branch, and generating an initial three-dimensional digital twin base; acquiring factory real-time inspection images acquired by the inspection robot under different inspection poses, rendering the initial three-dimensional digital twin base by using a radiation field branch to obtain a virtual reference view, and performing difference analysis on the virtual reference view and the factory real-time inspection images to position a structural change area in a factory; determining an axis alignment bounding box for changing the physical range of the entity in the area, taking the axis alignment bounding box as an effective boundary of the local space patch, and determining matched radiation field network weight and SDF network weight according to the effective boundary; and adjusting the radiation field network weight and the SDF network weight, and updating the radiation field network weight and the SDF network weight to the network in an incremental updating manner to generate a factory three-dimensional digital twinning model based on the implicit three-dimensional geometric field.

The invention discloses an ultrasonic image denoising method based on variational mode decomposition and local space sparse fusion, which comprises the following steps of: firstly, adaptively optimizing key parameters of variational mode decomposition by using a grey wolf optimization algorithm to realize stable and efficient decomposition of an ultrasonic image; then, classifying the modal components according to the structural features of the modal components, and implementing differentiated denoising strategies for different types of modals to separate noise and reserve useful information; after modal reconstruction, a sparse expression method based on local space information is further introduced, according to the method, accurate boundary detection is carried out through gradient vector flow and gray scale proportion analysis, self-adaptive partitioning is carried out on an image according to boundary information, and finally sparse reconstruction is carried out through double dictionaries trained for different areas. According to the method, speckle noise in the ultrasonic image can be effectively suppressed, and meanwhile, the capability of keeping the edge and detail information of a tissue structure is remarkably improved, so that the ultrasonic image with higher quality is obtained.

The invention discloses a video saliency target detection method based on attention and graph collaborative enhancement. The method comprises the steps of video frame preprocessing, network model establishment and video saliency detection. A collaborative attention graph network is adopted, a collaborative attention graph module AGM and a space edge attention module SEM are respectively used for independently extracting space-time clues and space edge features, inter-frame attention is combined with a space-time graph convolutional network GCN, and then motion information between video frames is captured while a receptive field is expanded to capture a long-distance dependency relationship. The space-time diagram convolutional network GCN connects the features from time to space branches, can collaboratively fuse cross-modal features, and considers global and local space-time features at the same time. The SEM enhances high-level information by extracting spatial and edge features from low-level data using a sobel operator and a spatial attention module. According to the method, the most interested target content of human eyes in a video can be effectively extracted, more attention is paid to the salient region, and background interference is inhibited.

The invention discloses a video abnormal event detection method combined with a space-time attention mechanism, and belongs to the technical field of video content understanding. In order to solve the problems that an existing detection method is low in accuracy and difficult to adapt to complex scenes, depth separable convolution is introduced into an I3D architecture, and therefore a local space-time feature extraction network based on lightweight I3D is created. A global spatiotemporal feature extraction network is formed by combining a time attention unit and a space attention unit and stacking the units in a crossed mode, and all-around and deep feature mining is carried out on video data from the two dimensions of time and space. And finally, exploring the internal relation and correlation between the anomaly detection and the classification task by means of a multi-task joint learning network. Through joint optimization of a plurality of related tasks, detection of various types of abnormal events is realized.

The invention relates to the technical field of computer vision, in particular to a behavior recognition method based on a space-time mixed attention module, which comprises the following steps of: S1, dividing joint points into regions: dividing each joint point into regions, and respectively performing feature enhancement on the sub-regions to obtain corresponding feature weights; and S3, space-time stream fusion: combining the attention scores of the space-time streams to obtain a space-time attention score statt, and endowing the obtained space-time attention score statt to an input feature map to realize enhancement of time domain and space domain features. According to the method, the perception ability of the model to the local space structure and the time dynamic change is enhanced, the interaction and information exchange between the time domain and the space domain characteristics are enhanced, and the understanding ability and prediction precision of the model to the complex space-time mode are improved.

The invention discloses a multi-modal emotion recognition system based on global-local space-time semantic alignment, and the system comprises the steps: dividing a multi-modal physiological signal into a plurality of single-modal physiological signals, and then dividing all the single-modal physiological signals into a plurality of batches; obtaining the maximum sequence length of the feature sequence of the single-mode physiological signal fragment in each batch, and by taking the maximum sequence length in each batch as a reference value, supplementing the sequence length of the single-mode physiological signal fragment in each batch to the maximum sequence length, therefore, the lengths of the feature sequences of the single-mode physiological signal fragments in each batch are the same, and the problems that the lengths of the physiological signal data are inconsistent and key information is lost are solved; according to the method, the local time semantic alignment module is provided to align each mode in the time dimension, so that the interpretability and robustness of the model are enhanced; according to the method, the semantic information of each mode is aligned in the spatial dimension, so that the emotion recognition process is optimized.

The invention discloses a risk map search-based heterogeneous agent collaborative underground space exploration method, which comprises the following steps of: acquiring peripheral information of a heterogeneous agent group at a current position to construct a local map, and constructing a first local risk grid map by aerial agents in the heterogeneous agent group according to the local map, calculating a risk value of each cell in the first local risk grid map; searching candidate points, starting from the current position, of the heterogeneous agent group, selecting a next search point from the candidate points by the ground agent according to a search gain combined with a risk value, and selecting a next search point from the candidate points by the aerial agent according to the search gain; and after the aerial intelligent agent and the ground intelligent agent complete the exploration of the local space according to the local planning path, performing global path planning of the global space on the heterogeneous intelligent agent group according to the global exploration gain function until the exploration of the unknown area is completed. And guiding the UGV to avoid potential dangerous areas through a terrain risk map constructed by the UAV to balance heterogeneous cluster search efficiency and safety.

The invention belongs to the field of self-organizing networks, and relates to a self-organizing node space reference construction method, which comprises the following steps of: constructing a local space reference topology comprising a distance and an orientation angle by utilizing mutual measurement among distributed self-organizing nodes, and realizing global space reference construction through distributed fusion. And the self-organizing multi-node network distributed space reference construction requirement is met. According to the method, external information support such as satellites is not needed, and only distance and angle measurement between self-organizing nodes is needed; the space reference parameters comprise position and orientation angle information at the same time, and more applications can be supported; a distributed fusion processing method is adopted, a center node is not needed, and the complexity is low.

The invention discloses a figure portraying method based on a combined space-time attention mechanism and a related device, and relates to the technical field of user portray.The method comprises the steps that user behavior attribute data of a target user are acquired, and in order to conveniently capture complex relations and features in the user behavior attribute data, the user behavior attribute data are acquired; the method comprises the following steps of: firstly, converting discrete user behavior attribute data into embedded vector representation by adopting an embedded layer to obtain a comprehensive embedded vector, and then respectively extracting first and second space-time relationship characteristics from the comprehensive embedded vector by adopting a space-time attention and convolution fusion module; according to the method, a global time dependency relationship, a local space dependency relationship, a global space dependency relationship and a local time dependency relationship of user behaviors are extracted, classification and clustering are carried out according to the global time dependency relationship, the local space dependency relationship and the global space dependency relationship, a label of a target user is determined, and an accurate figure portrait is generated. According to the scheme, the local and global spatial-temporal features in the data can be accurately and simultaneously extracted, and a solid foundation is provided for further analysis and decision making.

The invention discloses a two-way traffic flow dynamic modeling method based on a deep space-time diagram neural network, and belongs to the field of intelligent traffic and data mining, and the method comprises the following steps: S1, constructing a traffic flow dynamic diagram, and capturing the local space-time dependence of a traffic flow; s2, after the local space-time dependency is obtained, extracting time features by using a time convolutional layer; s3, after the time features are obtained, a selection gate is used for screening the effective information obtained in the step S2, and final output is obtained; s4, after a result in S3 is obtained, using a composite graph convolution module to extract spatial features; s5, fusing the two-way traffic, and obtaining a final prediction result through a prediction layer; s6, the loss of the model is calculated, and whether the model meets the requirement is determined, the problems that in the prior art, in traffic flow prediction, bidirectional flow modeling is insufficient, dynamic dependency capture is insufficient, and adaptability to complex traffic scenes is poor are solved, and therefore the traffic flow prediction precision and the model applicability are improved.

The invention relates to a binding material single particle local spatial evolution analysis method, and belongs to the technical field of civil engineering binding material microstructure characterization. The method comprises the following steps: acquiring back scattering (BSE) and characteristic X-ray energy spectrum (EDS) images of a sample; selecting a target gel particle based on the BSE image and intercepting a local area; identifying the reaction edge of the target gel particle; performing equal-width stepped strip division inwards and outwards by taking the reaction boundary as a starting point, and storing the mask; carrying out image operation on adjacent masks to extract strips, and carrying out statistical analysis on gray feature and element feature changes of each strip; according to the method, surrounding phase interference is avoided through local analysis, whole-process analysis is realized by adopting unified image processing software, a programming basis is not needed, the technical threshold is reduced, the operation process is simplified, the uniformity of strip division and the reliability of an analysis result are ensured, and the method is suitable for large-scale popularization and application. And a simple, convenient and effective technical means is provided for revealing the reaction mechanism and quantitative reaction activity of the cementing material.

The invention discloses a space-time elastic weight consolidated edge federal continuous learning method, which is applied to a system comprising a server and a plurality of edge devices, is used for processing space-time heterogeneity time sequence data, and comprises the following steps: initializing training, broadcasting a previous time sequence global time Fisher diagonal matrix (the first time sequence is not broadcasted, the second time sequence is not broadcasted, and the third time sequence is not broadcasted) by the server; however, the global model needs to be randomly initialized and broadcasted); in the model training stage, based on local data, a global time Fisher diagonal matrix and the like, an edge device updates a local model through a loss function containing a time / space regular term, calculates a local space Fisher diagonal matrix, uploads the local space Fisher diagonal matrix, and then a server weights and aggregates the global model according to the data volume and issues the global model, and circulates until convergence; and in the global time Fisher diagonal matrix calculation stage, the equipment calculates a local time Fisher diagonal matrix based on a convergence model, and uploads and aggregates the local time Fisher diagonal matrix for the next time sequence. Historical data does not need to be stored, original data does not need to be transmitted, storage calculation / communication overhead is reduced, privacy is protected, and the model convergence speed and precision are improved.

The invention discloses a deep learning-based satellite remote sensing collapsible loess foundation evaluation method, which comprises the following steps of: acquiring an optical remote sensing image, a radar remote sensing image and an infrared remote sensing image of a target area, and preprocessing; collapsibility feature weighted data are screened, and feature weights are distributed; feature extraction is carried out through a local space exhibition structure and a multi-layer Transform structure of the collapsibility foundation evaluation network; collapsibility area self-supervised collaborative learning is carried out, and self-supervised training is carried out based on spatial correlation, pseudo labels and spatial consistency regular terms; and carrying out risk grade division on the target area, and outputting a collapsibility risk result of each spatial position. According to the method, multi-mode remote sensing and deep learning are fused, high-precision intelligent partitioning of the collapsible loess foundation is achieved, and the method has the advantages of being self-adaptive, low in manpower and high in spatial resolution.

The invention relates to the field of GRP pipeline joint testing, in particular to a GRP pipeline joint water pressure testing method, which comprises the following steps of S1, checking whether joints of two adjacent pipelines to be subjected to pressure testing are well connected or not; s2, sealing the joint of the two adjacent pipelines to be subjected to pressure test by using an air bag, wherein the joint of the two adjacent pipelines to be subjected to pressure test is surrounded by the pressure test space; s3, injecting water into the pressure test space until the pressure in the pressure test space reaches the test pressure, and then stabilizing the pressure; s4, observing whether the pressure drop in the pressure test space is in an allowable range or not, and if so, entering the next step; if the pressure drop exceeds the allowable range, the steps S1 to S3 are repeated after the joint is repaired until the pressure drop is within the allowable range; and S5, moving to the next test point, and repeating the steps S1 to S4 until the joint water pressure test of all the pipelines to be subjected to pressure test is completed. According to the method, only the local space at the joint of the two adjacent pipelines to be subjected to pressure test needs to be subjected to water filling and pressing, the water consumption is greatly reduced, the working efficiency is improved, and the progress of the whole project is accelerated.

The present invention makes it possible to effectively increase the density of a three-dimensional point cloud.　A plurality of clusters are generated by clustering three-dimensional point clouds acquired on the basis of the phase difference, or the time difference between the transmission time and the reception time, of a transmission signal and a reflection signal corresponding to each of a plurality of discretely arranged basic transmission points among a plurality of transmission points arranged in the form of a two-dimensional array. For each of the plurality of clusters, geometric information in a local space formed by the three-dimensional point clouds included in the cluster is calculated. For each of the plurality of clusters, a determination is made, on the basis of the geometric information, regarding whether to perform a density increase involving increasing the density of the three-dimensional point clouds included in the cluster by using another transmission point in addition to the basic transmission points corresponding to each point of the three-dimensional point clouds.

The invention discloses a high-definition ultrasonic blood flow imaging method and a high-definition ultrasonic blood flow imaging system based on space-time multi-scale flexible registration, which are used for removing adverse effects of motion artifacts and realizing high-definition ultrasonic blood flow imaging, and remarkably improving the ultrasonic blood flow imaging quality of moving organs by utilizing a registration method based on a multi-scale space-time similarity matrix. The method specifically comprises the following steps: a preprocessing module M1 is responsible for mapping acquired N frames of ultrasonic echo IQ complex signals X1 frame by frame from a complex field to a real field to construct a matrix Y1; the similarity calculation module M2 is responsible for calculating a multi-scale space-time similarity matrix # imgabs0 # (i = 1, 2,..., W-C + 1, j = 1, 2,..., H-C + 1, k = 1, 2,..., N-1) of Y1 according to different sliding window sizes C; the clustering module M3 is responsible for performing multi-scale adaptive local space-time clustering on the # imgabs 1 # and labeling the IQ complex signal at each position in the X1 as # imgabs 2 # (p = 1, 2, 3,..., P) based on a clustering result; the registration module M4 is responsible for registering the IQ * complex signal corresponding to each label according to a multi-scale time direction similarity distribution feature and a multi-scale intra-class spatial structure feature; and the result output module M5 is responsible for carrying out wall filtering on the IQ complex signal after multi-scale registration and obtaining a high-definition ultrasonic blood flow imaging result W.