73 results about "Projection space" patented technology

The invention provides a sheep body size measuring method and device based on three-dimensional reconstruction and point cloud segmentation, a medium and a program product, and relates to the technical field of computer vision. The method comprises the following steps: acquiring a sheep image of a to-be-detected sheep, and preprocessing the sheep image to obtain a multi-view sheep target point cloud; performing point cloud registration processing on the multi-view sheep target point cloud to obtain a sheep three-dimensional model; performing point cloud segmentation processing based on the body size measurement key points and the sheep three-dimensional model to obtain key point region point clouds; based on the key point region point cloud, determining a region point cloud projection space, performing attitude normalization on the interval point cloud, and determining an attitude normalized key point region point cloud; and calculating the body size data of the to-be-measured sheep based on the point cloud of the key point region subjected to attitude normalization. According to the embodiment of the invention, through three-dimensional reconstruction of the body surface point cloud of the sheep, non-contact measurement of the body size data of the sheep is realized, and the measurement efficiency and accuracy are improved.

The invention discloses an asymmetric cross-modal hash learning method for potential structure information, and relates to the technical field of cross-modal retrieval. Analyzing a sample clustering structure, sample similarity information and semantic similarity information among the multi-modal samples to obtain a projection matrix of different modals, a clustering center of a projection space, a clustering indication matrix, a mapping matrix, a discrete hash code and a calculation formula of an auxiliary variable; updating the projection matrix of each modal, the clustering center of the projection space, the clustering indication matrix, the mapping matrix, the discrete hash code and the auxiliary variable by using an iteration strategy; determining a hash function of each modal according to the discrete hash code obtained by the last iteration and the multi-modal training sample; the hash function is used for performing similarity retrieval on the query sample in the multi-modal data. According to the method, the Hash code with higher discrimination can be obtained, so that the retrieval performance of cross-modal retrieval is improved.

The invention belongs to the related technical field of power system optimization operation, and particularly relates to a non-iterative interconnected multi-microgrid system optimal power flow determination method based on a dimensionality reduction projection method, and the method comprises the steps: reconstructing a multi-microgrid centralized operation model, and obtaining a sub-problem of each sub-microgrid; taking a multi-dimensional variable space formed by all public variable sets in each sub-problem and a preset sub-microgrid operation cost variable as a projection space of the sub-problem; determining an initial vertex set of the projection of the convex hull corresponding to the original operation space constraint of each sub-problem in the projection space, and translating the coordinate origin to the interior of the convex hull enclosed by the vertex set; based on the convex hull and the outer normal vector of each boundary surface, circularly searching vertexes satisfying original operation space constraints outside each boundary surface in the projection space, and taking a space enclosed by a final vertex set as a projection space of the operation space constraints of the sub-problem; and reconstructing the sub-problem in the space and solving the sub-problem. According to the method, the optimal power flow can be solved under the condition that only public variables are disclosed.

The invention provides a boundary-guided rolling bearing semi-supervised fault diagnosis method and system, and relates to the technical field of bearing fault diagnosis, and the method comprises the steps: collecting a rolling bearing vibration signal data set which comprises a labeled sample and an unlabeled sample; constructing a diagnosis network; inputting the time domain and frequency domain signals into a feature extractor to extract features; performing cross-mask comparison learning on an output layer of the feature extractor, and calculating cross-mask comparison loss; carrying out asymmetric predictive contrast learning in the projection space, and calculating asymmetric predictive contrast loss; constructing a boundary guiding mechanism by using the labeled samples, and guiding unlabeled sample aggregation by monitoring loss; introducing a consistency regularization mechanism, and constructing consistency loss by using time-frequency prediction distribution of unlabeled samples; integrating each loss training network until convergence; and finally, fault diagnosis of the label-free signal is realized. According to the method, the problems of fuzzy inter-class boundary and poor feature clustering quality under limited labeling are solved, and the diagnosis precision and robustness are improved.

The invention provides an immersive projection man-machine interaction method and system, and the method comprises the steps: constructing a closed projection space which comprises a plurality of walls, a bottom surface and a top, deploying a main image collection device at the center of the top, achieving the positioning of the head space of a user, and arranging auxiliary image collection devices at a plurality of corners of the combination part of the adjacent walls. Based on the real-time image data acquired by the main device, the intelligent recognition technology is adopted to analyze the three-dimensional coordinates of the center of the head circle of the user, the face orientation angle is calculated in combination with the facial feature data acquired by the auxiliary device, and the dynamic mapping model of the user interaction intention is determined by fusing the space coordinates and the orientation angle. According to the method, a main-auxiliary cooperative visual positioning framework is adopted, space perception limitation under a single visual angle is avoided, two-dimensional real-time tracking of a user position and a sight line direction is realized, an accurately matched interactive animation is generated at a corresponding position of a projection space according to a dynamic analysis result, and the user experience is improved. And the real-time response capability and behavior intention capture accuracy of human-computer interaction in an immersive environment are improved.

The invention belongs to the technical field of mining engineering, discloses a mining roadway non-uniform three-dimensional deformation grading calculation and analysis method, and aims to solve the problems that a traditional point cloud comparison method is low in precision and large in error in a roadway high-deformation and non-uniform-deformation environment, and traditional monitoring wiring is complex and cannot achieve global coverage. According to the technical key points, in a GNSS-free underground environment, a portable three-dimensional laser scanning device integrating SLAM synchronous positioning and a map construction technology is adopted to collect two-stage roadway point cloud data, a centimeter-level precision three-dimensional model is constructed through a pose map optimization algorithm, a two-stage deformation hierarchical optimization algorithm is designed, and a two-stage deformation hierarchical optimization algorithm is designed. In the first stage, a deformation core point is selected based on super voxel segmentation, rough calculation is performed in a cylindrical projection space, and a key deformation area is rapidly identified, and in the second stage, a refined displacement amount is calculated for a point whose displacement exceeds a threshold value through geometric constraint and a centroid method, and the data quality is improved in combination with point cloud denoising, so that high-precision continuous monitoring of the full section of the roadway is realized.

Methods, systems, and computer-readable storage media for using cluster kinematics for enhanced data analysis of temporal data. A data sample from a temporal sequence of data samples associated with a node is received and projected into a projection space of a clustering model defining one or more clusters. A distance value associated with each of the one or more clusters in the clustering model is calculated for the data sample. One or kinematic metrics associated with a cluster in the clustering model are calculated for the node from the calculated distance values, a time value associated with the received data sample, and previously calculated distance values and kinematic metrics for the node calculated from previously received data samples, where the calculated kinematic metrics represent a trajectory of the node in relation to the cluster in the projection space of the clustering model.

The invention relates to the technical field of computer-aided analysis, in particular to a floating foundation structure dynamic response real-time analysis system and method.The system comprises a residual vector calculation module for generating a residual force vector based on a full-order external excitation force vector, an increment base generation module for calculating a modal amplitude and constructing an increment base vector, and an increment base generation module for generating an increment base vector; the reduced-order matrix updating module constructs an enhanced reduced-order model by using the updated projection matrix, and the transient response solving module solves and obtains a corrected generalized coordinate value based on the enhanced reduced-order model and maps the corrected generalized coordinate value into a dynamic displacement field. According to the method, a residual vector is generated by calculating a difference value between full-order external excitation and internal force, residual force is mapped into static deformation distribution by using a stiffness matrix inverse operator, an incremental base is constructed through orthogonalization processing, adaptive updating of a projection space of a reduced-order model is realized, the system dimension is reduced, and the precision is guaranteed. And the calculation efficiency and the real-time performance of dynamic response analysis of the floating structure are improved.

The invention relates to an intention recognition system based on a multi-mode agent and data annotation, in particular to the field of artificial intelligence, according to the scheme, a semantic gap between different modes is effectively bridged by constructing a unified semantic projection space, the reliability of information of each mode is quantitatively and dynamically evaluated by using uncertainty, weighted fusion is performed on the basis, and the intention recognition accuracy is improved. A cross-modal semantic memory network and an online learning mechanism are innovatively introduced, so that the scheme not only can make more accurate intention inference by integrating historical experience and real-time perception, but also can continuously optimize own confidence assessment ability and a memory bank in an interaction process; therefore, the method has higher adaptability and robustness in the environment with noise, missing information and dynamic change.

A rapid verification method for adaptive radiotherapy dose consistency in the TOMO system is disclosed, relating to the field of medical information technology, particularly TOMO radiotherapy information processing. The method includes: acquiring flux distribution data and the original region of interest (ROI) from the original radiotherapy plan; projecting the ROI onto the flux projection space based on the beam geometry parameters of the original radiotherapy plan, and establishing a mapping relationship as a reference index; acquiring the target ROI for the adaptive radiotherapy phase; updating the mapping relationship based on the flux distribution data and the target ROI, and using the mapping relationship as the target index; detecting the data consistency between the target index and the reference index, and determining the dose distribution change information of the target object. This application solves the technical problems in existing technologies where adaptive radiotherapy dose consistency verification relies on a high-load dose calculation engine for data reprocessing, resulting in large amounts of computational data, high system response latency, and inability to meet the timeliness requirements for online verification.

The invention discloses a method for identifying a rod-shaped object in a plateau mountain complex environment, and the method comprises the steps: carrying out the quick 3D point neighbor search, carrying out the spherical projection of an original three-dimensional point cloud, and carrying out the point neighbor search in a spherical projection space through a sliding window method; performing semantic segmentation on the point cloud data by using an improved 3D-MiniNet network; after semantic segmentation, searching by using a K-nearest neighbor method, and endowing each 3D point with a semantic tag; performing segmentation processing on the point cloud data of the rod pieces, and then performing filtering processing to obtain the point cloud data of each rod piece. Through a spherical projection and sliding window method, adjacent points of a 3D point cloud are quickly and accurately found, semantic segmentation is carried out on point cloud data by using an improved 3D-MiniNet network, information of a rod-shaped object can be accurately positioned, it is ensured that an unmanned vehicle can accurately judge a weeding position and a weeding range, and unnecessary damage to equipment is avoided.

While convolutional neural networks have been prized for image-processing tasks, they have limitations. Embodiments introduce a new Hough-to-Radon Transform (HRT) layer into an artificial neural network, such as a convolutional neural network, to address one or more of these limitations, without compromising on the accuracy of the artificial neural network for image-processing tasks. The HRT layer converts an input image from a first parameter space into a second parameter space of reduced complexity. Inner layers may operate on the image in this second parameter space, instead of in the first parameter space, to reduce the overall computational cost of the artificial neural network.

To provide a blasting device for a wall surface that requires less manpower for work.SOLUTION: A wall-surface blasting device for performing blast treatment on a steel portion of a wall surface of a steel bridge, comprises a body portion and a lifting portion that lifts the body portion and causes the body portion to face the steel portion. The body portion includes: a projection unit that projects an abrasive material for the blast treatment; a collection unit that collects the abrasive material projected by the projection unit; and a circulation unit that supplies the abrasive material, collected by the collection unit, to the projection unit. A closed projection space in which the abrasive material is projected toward the wall surface is defined by its being closed by the wall surface while the body portion is kept facing the steel portion. The abrasive material projected from the projection unit in the closed projection space is collected by the collection unit, and the abrasive material is supplied to the circulation unit. The abrasive material supplied to the circulation unit is supplied to the projection unit, and the abrasive material is projected again from the projection unit.SELECTED DRAWING: Figure 1

The invention relates to the technical field of mixed reality and immersive display, and discloses an immersive scene experience system and method based on mixed reality superposition projection, and the system comprises an experience module which is formed by a plurality of projection curtains in a surrounding manner, and an experience space is formed in the plurality of projection curtains; the projection module is used for constructing a holographic projection scene; the MR glasses device is used for displaying a virtual image and space interaction; the space positioning module is used for generating a space model and forming an NFC label identifier in the space model; the control module is used for receiving position information of the MR glasses in the projection space; and controlling a playing picture of the projection module and / or the MR glasses device according to the position information. Compared with a traditional mode, the integration degree between the virtual image and the holographic projection scene is higher; meanwhile, the user triggers a virtual event or operates a virtual object in a scene through a gesture, and dynamically adjusts the content of a virtual image according to real-time sensor data, so as to keep synchronization with the action of the user.

Methods, systems, and computer-readable storage media for using cluster kinematics for enhanced data analysis of temporal data. A data sample from a temporal sequence of data samples associated with a node is received and projected into a projection space of a clustering model defining one or more clusters. A distance value associated with each of the one or more clusters in the clustering model is calculated for the data sample. One or kinematic metrics associated with a cluster in the clustering model are calculated for the node from the calculated distance values, a time value associated with the received data sample, and previously calculated distance values and kinematic metrics for the node calculated from previously received data samples, where the calculated kinematic metrics represent a trajectory of the node in relation to the cluster in the projection space of the clustering model.

The application proposes a polyhedral immersion projection system, and the system comprises a polyhedral projection space which comprises a plurality of wall surface bodies which are arranged in a surrounding manner, top surface bodies which are disposed at the top side of each wall surface body, and bottom surface bodies, and the top surface bodies are disposed in an inclined manner relative to the wall surface bodies, and a top opening is formed by the surrounding of the plurality of top surface bodies; the projection assembly is arranged at the top opening and used for projecting pictures to the wall surface body and the top surface body, and an inclined plane is designed at the top of a traditional pentahedron space to construct a polyhedral immersion space for obtaining immersion experience induction and projection area.

The present invention discloses a method and system for predicting film thickness based on self-supervision and representation learning, which belongs to the field of machine learning and artificial intelligence technology. A labeled first training set and an unlabeled second training set are obtained, the encoder is pre-trained with the second training set, and then fine-tuned with the first training set, while training the dynamic feature selector and the regression head. The dynamic feature selector reduces the dimension of the first training set data, and then uses the encoder to initialize the global prototype projection space. The encoder generates an embedded representation of the reduced-dimensional data, the feature representation head calculates the projection coordinates, and the projection loss, regression loss, diversity loss and global prototype orthogonal loss are combined to update the model; finally, the actual data is used as input to predict the film thickness through the dynamic feature selector, encoder and regression head. The present invention does not need to rely on a large amount of labeled data, solves the feature selection problem, representation aliasing and localization problem of highly complex semiconductor manufacturing data, and has high prediction accuracy and stability.

The present invention discloses a method for measuring the spatial distance between a hazard source and a target using monocular vision, and relates to the field of distance measurement. The method comprises: deploying front-end acquisition equipment; controlling a laser projection device to perform a full-area scan of a target scene, transmitting laser point cloud data back, controlling a visual camera to perform full-field scene acquisition, and transmitting image data back; constructing a scene space model using the laser point cloud data and image data; and determining the target spatial distance by using the scene space model, by matching the image space with the point cloud two-dimensional projection space, and by mapping the point cloud two-dimensional projection space with the point cloud three-dimensional space, to determine the target point cloud position and perform spatial distance calculation. This method solves the technical problem of insufficient accuracy in existing methods for measuring the spatial distance between hazard sources and targets using monocular vision, and achieves the technical effect of improving the accuracy of spatial distance measurement.

This invention proposes a structurally enhanced discriminative width learning system for image classification. First, it introduces graph regularization based on class relationships to ensure that similar samples remain closely distributed in the projection space. Second, it adds intra-class divergence regularization to further enhance the aggregation of samples within the same class. Furthermore, it introduces inter-class divergence regularization to ensure that the centers of different classes are far apart in the feature space. Finally, in the output layer regression stage, it replaces the Frobenius norm regularization in ridge regression with sparse norm regularization to effectively suppress noise and redundant information. This image classification scheme, while maintaining the simplicity and efficiency of the optimization process, organically integrates three different types of structured information—intra-class structure, inter-class structure, and local manifold information—into a unified model, enabling the learned feature representations to possess stronger structure preservation and discriminative capabilities.

The invention provides a model construction and display method for a fusion projection space and electronic equipment. The method comprises the following steps: acquiring an initial model corresponding to the fusion projection space to be projected; obtaining an area needing independent display control in the fusion projection space, and constructing a shape object in the initial model according to the position of the area; setting projection textures of the initial model and the shape object to be different, and performing projection; and adjusting the shape object according to the overlap ratio of the projected shape object and the area until the overlap ratio reaches a threshold value, and storing the shape object and the initial model in an associated manner to obtain a final model. According to the method, the area is covered by constructing the shape object in the initial model, so that the modeling of the area is realized, the modeling efficiency is improved, the projection effect of the shape object can be visually seen, and the modeling accuracy is ensured.

The invention discloses a projector setting method, which comprises the following steps of: setting a group of projection space parameters and a group of projector erection parameters of at least one projector by utilizing a computer system; executing a projection simulation program, performing calculation based on the set of projection space parameters and the set of projector erection parameters by using the computer system, and determining a set of setting parameters related to the projectors; setting a group of configuration parameters related to the projector by using the computer system, wherein the group of configuration parameters comprises an ID (Identity) number and an IP (Internet Protocol) address; and storing the set of configuration parameters and the set of setting parameters about the projector to the mobile device.

The invention discloses a Grasshopper-based parameterized modeling method for a polygon projection space distorted surface reticulated shell, which comprises the following steps of: S1, setting geometric parameters of a reticulated shell model in a Grasshopper platform through an input plug-in, and obtaining a corresponding projection surface of a space curved surface reticulated shell surface; s2, connecting vertexes of the projection surface to generate ridge lines forming a reticulated shell curved surface, and constructing a polygonal projection space distorted surface through curved surface generation operation based on the ridge lines; s3, performing grid division on the spatial warping surface by using an isoparametric method or a mapping method to generate grid points; and S4, based on the grid points, generating a single-layer polygonal projection space distorted surface latticed shell structure through a polyline connection plug-in. The method is used for solving the technical problems that an existing spatial reticulated shell structure with a complex shape is difficult to model, the modeling efficiency is low, and the repeated utilization rate is low.

An electronic device is disclosed. The present electronic device comprises: an output device that emits infrared rays from the front of the electronic device; a distance sensor that measures the distance to an object; a camera; a memory storing at least one instruction; and at least one processor that executes the at least one instruction, wherein the at least one processor controls the output device to emit infrared rays in the direction in which an image is projected, acquires information of an object approaching the projection space between a source and the projection area in which the image is projected, on the basis of distance information acquired by the distance sensor and the infrared distribution in a captured image acquired by the camera, and controls an image projection operation on the basis of the acquired object information.

The present invention discloses a robust principal component analysis method for application scenarios with a large amount of unsupervised noise data. By adaptively assigning a weight of 0 or 1 to each sample to discard outliers and calculating the average value through the selected normal samples, the resulting projection space is more reasonable and representative for most samples. In order to maintain the local smoothness of the sample distribution, the present invention learns the affinity relationship of the samples in the subspace, maintains the local smoothness of the sample distribution, and finally proposes an iterative algorithm to solve the model. Based on robust principal component analysis and combined with adaptive outlier detection, the present invention can be applied to multiple fields such as image recognition, data compression, pattern recognition and classification, machine learning, statistics and data analysis.

The invention discloses a multi-modal BEV perception consistency alignment checking method based on language structure prior, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining multi-modal sensor data, and mapping the multi-modal sensor data to a virtual projection space; and based on the large language model, generating language structure priori in a structured text form, and mapping the language structure priori to the virtual projection space to generate a structure constraint field. In the training stage, based on a structure constraint field, weak supervision consistency loss is constructed, an evidence gating mechanism is introduced, loss weight is dynamically adjusted according to an evidence intensity score, and joint optimization is carried out on the network. In the reasoning stage, consistency checking is carried out on BEV sensing results by utilizing a structure constraint field, and confidence reestimation, hallucination suppression and open set rejection are carried out in combination with a structure conflict score and an evidence intensity score. The problems of structure default and high-confidence illusion under the weak evidence working condition are solved, and the physical consistency and reliability of the sensing result are improved.

The invention discloses a three-dimensional scene object identification method based on display precision, and the method comprises the following steps: obtaining a viewpoint parameter of a current three-dimensional scene, and determining a display precision state of current three-dimensional space data based on the viewpoint parameter and a preset precision grading strategy; calculating a model precision deviation parameter for quantifying the geometric simplification degree of the current model according to the display precision state; based on the model precision deviation parameter, dynamically determining a space determination parameter for three-dimensional space object identification; responding to a user interaction operation, obtaining a corresponding screen coordinate and mapping the screen coordinate to the three-dimensional projection space; taking the mapped projection position as a center, constructing an effective judgment area according to the dynamic space judgment parameters, judging the three-dimensional space object in the effective judgment area, and determining a target object; based on the identification information of the target object, triggering and executing an interaction processing operation associated with the target object; the method has stability and practicability in a complex large-scale scene.

A projector setting method includes: using a computer system to configure a set of projection space parameters and a set of projector installation parameters about at least one projector; executing a projection simulation program, which uses the computer system to perform calculations based on the set of projection space parameters and the set of projector installation parameters, and determines a set of setting parameters for the projector; using the computer system to configure a set of allocation parameters for the projector, in which the set of allocation parameters includes an ID number and an IP address; and storing the set of allocation parameters and the set of setting parameters about the projector to a mobile device.

This application provides a scanning image reconstruction method and scanning imaging system adapted to incomplete data, relating to the fields of image processing, artificial intelligence, and radiation imaging technology. The method includes: in response to a ray scanning of an object to be imaged, acquiring projection data of the object, wherein the projection data is incomplete projection data in a projection space; transforming the incomplete projection data into first template data according to the mapping relationship between the projection space and a template data space; pre-completing the first template data to obtain second template data; acquiring confidence information of the second template data; inputting the second template data and confidence information as input to a pre-trained data processing model; processing the input using a first sub-model to obtain third template data; and performing image reconstruction based on the third template data to obtain a reconstructed image. This method can improve the efficiency and accuracy of data completion while ensuring the fidelity of the original sampled projection, and is beneficial for obtaining high-quality reconstructed images.

A computer-implemented method for profile detection in a discrete 3D model. The discrete 3D model represents a mechanical part. The method includes, for each 3D surface of at least one 3D surface of the 3D model, providing a candidate parameterization of the 3D surface and a candidate projection of the 3D surface into . The method further includes computing a function. The function penalizes, for each couple of points of the 3D surface having neighboring parameter values, a distortion. The distortion is between a disparity between the projection of the points, and a disparity between the parameterization of the points. The method further includes determining that the candidate parameterization and the candidate projection form a valid profile of the 3D surface if the computed function is smaller than a predefined threshold.

The invention discloses a vector query method and device, and the method comprises the steps: mapping a received query vector to a low-dimensional projection space, and obtaining a projection query point; a root node of the pre-constructed index tree is provided with a plurality of sub-trees, leaf nodes of the pre-constructed index tree are used for storing data points in the low-dimensional projection space, each node of the index tree corresponds to an area in the low-dimensional projection space, and the area corresponding to the sub-nodes in the index tree is included in the area corresponding to the father node; distributing the plurality of sub-trees to a plurality of processing units, and executing range query on the distributed sub-trees by each processing unit: traversing nodes of the sub-trees, and for non-leaf nodes in the sub-trees, if the lower bound of the distance between a corresponding region of the non-leaf nodes and a projection query point is greater than a preset search radius, pruning the sub-trees of the non-leaf nodes; and for a first leaf node in the sub-tree, if the lower bound of the distance between the corresponding region and the projection query point is greater than the search radius, adding data points contained in the first leaf node into a local candidate set for determining a query result.