39 results about "Information projection" patented technology

The invention discloses an agent training data set construction method and system combined with cross-modal learning, and relates to the technical field of agent training, and the method comprises the steps: obtaining multi-modal intelligence data from a multi-source database, extracting feature information, and projecting the feature information to a preset semantic space to obtain cross-modal alignment features; constructing an agent function call syntax tree based on cross-modal alignment features, analyzing the syntax tree into an instruction sequence, constructing an analysis reasoning chain, and generating a sample label and an interaction track; constructing a task dependency graph by using the sample labels and the interaction tracks, decomposing the task dependency graph into a plurality of parallel decision branches, and dynamically adjusting a processing strategy to form a training data set; mapping the training data set to an agent target space, optimizing and analyzing an inference chain through a training feedback channel, and outputting a standardized sample library; and finally inputting to-be-analyzed data into the trained agent, and generating an intelligence analysis report according to the analysis reasoning chain.

The invention relates to the field of intelligent driving, and discloses a map updating method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining inertial measurement data, vehicle-mounted wheel speed and camera data on the vehicle; calculating according to the inertial measurement data and the vehicle-mounted wheel speed, and determining the vehicle pose; semantic segmentation is carried out on the camera data, a semantic image is determined, and the semantic image comprises ground semantic information; projecting ground semantic information on a road to a three-dimensional space on a world coordinate system based on a vehicle pose, and determining point cloud information of a target lane line; and the point cloud information of the target lane line is matched with the point cloud data on the high-precision map, the map is updated according to a matching result, and the point cloud data on the high-precision map is determined by the priori map. According to the method, the limitation of traditional high-precision map static updating is broken through, real-time dynamic correction of map data is realized through multi-sensor space-time synchronization and three-dimensional reconstruction, the cost is reduced, and the coverage use scene is expanded.

The invention discloses a video generation method and system based on a scene adaptive track and mirror moving control, and belongs to the technical field of computer vision. The method comprises the following steps: firstly, aligning a scene point cloud and a parameterized human motion sequence to a unified coordinate system, generating a ground adaptive three-dimensional track according to a two-dimensional track input by a user and scene geometry, and redirecting the motion; projecting the three-dimensional information into a two-dimensional condition graph based on camera parameters; and finally, in a conditional diffusion model generation process, selectively fusing scene geometric condition information after noise addition into the current denoising latent variable by utilizing a visibility mask generated by the depth map, so as to realize view angle self-adaptive mirror operation control. According to the method, the problems of conflict between character motion and scene, inflexible track control and inconsistent background under camera motion are solved, and a controllable video which is physically consistent and visually real can be generated.

The application relates to a rapid matching positioning optical projection system and method, wherein the optical projection system comprises a projector, a camera, an information editing module, a rapid positioning module and a projection operation module; the information editing module is used for compiling virtual process information required by each working procedure according to an assembly process flow, and generating a guide information set file; the rapid positioning module is used for determining two-dimensional coordinates of a preset target ball mark reference point in a camera image after movement according to projected workpiece images before and after movement of the optical projection system, and corresponding points of the target ball mark reference point in a projector image, and determining pose information of the projected workpiece relative to a projector space coordinate system in combination with measurement data of three-dimensional coordinates of the target ball obtained by a measuring device; the projection operation module is used for generating a working procedure guide image through a three-dimensional space projection algorithm according to the guide information set file and the pose information; and the projector is used for projecting the working procedure guide image on a surface of the projected workpiece.

This application provides a joint emotion recognition method and system based on action unit-driven attention, relating to the fields of computer vision and affective computing. The method includes: inputting a feature map into an action unit (AU)-driven attention branch, processing it through a sigmoid activation function to obtain an activation vector, and inputting this vector into a learnable mapping layer; projecting the AU information back into the spatial geometric space to generate a single-channel spatial attention map, which is then multiplied element-wise with the feature map to obtain a weighted feature map; determining continuous values ​​for discrete emotion category probabilities, valence, and arousal; constructing a total loss function; and training the model using the acquired AffectNet and Aff-wild2 datasets to determine the target model for emotion recognition. This application achieves end-to-end joint optimization of AU detection and emotion recognition, significantly enhancing the interpretability and generalization ability of the model while improving emotion recognition accuracy, enabling it to better adapt to the emotion recognition needs in complex scenarios.

The application provides a kind of complex environment under image enhancement and passage cost chart generation method and system, applied to image processing technical field, the method obtains original environment image and standardization processing, based on local gray variance self-adapting determination cutting limit threshold, combined with brightness protection weight generation enhanced image;Extract boundary mutation, surface roughness, local contrast and shadow risk and other multi-response characteristics, determine candidate high-cost area and generate boundary uncertain response;Fusion generates pixel-level passage cost chart above-mentioned response, and combined with camera calibration parameter and posture information projection to ground coordinate system, output spatial passage cost chart and passage area result.The application can reduce the influence of complex illumination, shadow and texture change on passage area extraction, improve the continuity and stability of passage risk representation.

Provided is an information projection method in which a projection device mounted on a vehicle displays information, wherein if it is determined that the vehicle is in an automatic driving state during automatic driving, the project unit displays information related to automatic driving on the basis of a display condition during the automatic driving.

This application provides an intelligent control method and device for a multi-functional projection device, belonging to the field of projection control. The method includes: acquiring personnel flow information, personnel distribution information, and personnel trajectory information of a target area; dividing the personnel flow information into multiple target personnel flow information based on the personnel flow information; each target personnel flow information corresponds one-to-one with a target sub-area; each target sub-area is an area within the target area that is at a different distance from the target multi-functional projection device; dividing the personnel trajectory information into multiple target personnel trajectory information based on the personnel flow information; each target personnel trajectory information corresponds one-to-one with a target sub-area; determining a target projection control strategy based on the multiple target personnel flow information, multiple target personnel trajectory information, and personnel distribution information; and controlling the target multi-functional projection device to perform projection display based on the target projection control strategy. This application can improve the adaptability of information projection by the projection device.

This invention discloses a visual recognition-based method and system for detecting OCA bonding defects, belonging to the field of defect detection. The method includes: acquiring the original grayscale image of the OCA bonding area; automatically filtering a candidate set of normal regions based on local texture flatness, and estimating the global background joint grayscale distribution based on the candidate set of normal regions; extracting the empirical joint grayscale distribution of the local neighborhood window for each pixel position in the original grayscale image; calculating the minimum information projection distribution of the empirical joint grayscale distribution onto the global background joint grayscale distribution under exponential family constraints, and calculating the cross-entropy between the global background joint grayscale distribution and the minimum information projection distribution to obtain a cross-entropy map; performing threshold segmentation on the cross-entropy map and outputting a defect location mask. This application solves the technical problem that existing OCA bonding defect detection technologies are unable to distinguish between grayscale changes caused by normal process fluctuations and real defects, resulting in insufficient detection efficiency and production line adaptability.

Systems and methods for tracking a non-self-reporting vessel ("dark vessel") with an unknown identity are provided. A method includes: providing a system configured to receive and store a plurality of ship detections; comparing a timestamp T2 of a new ship detection having a geolocation of P2 to a timestamp T1 of a first ship detection having a geolocation of P1 and determining that the T2 is later than T1; using kinematical information of the first ship detection, projecting the first ship detection to a projected second geolocation at T2; determining that the projected second geolocation is within a threshold proximity of P2; forming a vessel track having a geolocation of P1 at T1 and a geolocation of P2 at T2; and storing the vessel track in the system as a dark vessel track.

The invention relates to a projection lamp and method for an underground coal mine, and belongs to the technical field of intelligent illumination and information projection for the underground coal mine. The technical problems that traditional underground lighting equipment can only provide environment floodlight and cannot dynamically project graphical information, and a universal projector is poor in adaptability, low in safety and prone to causing image blurring in the underground high-dust, humid and vibration environment are solved. According to the technical scheme, the projection lamp is provided, the projection lamp is provided with an anti-explosion shell, an optical projection module, an image generation and control module, an environment sensing module and a power supply and communication module, image parameters are dynamically optimized through environment sensing data, and clear projection is achieved. The technical effects are that the underground information transmission mode is innovated, the environmental adaptability and intrinsic safety are improved, the intelligent integration capability is enhanced, and the long-term operation and maintenance cost is reduced.

A hand piece for dermal therapy is provided that includes a projector configured to project digital information onto the skin of the patient. A clinician may thus monitor the digital information to confirm that the proper therapy parameters are being used or to adjust the dermal therapy based upon the therapy parameters including in the digital information. In addition, the hand piece includes a camera so that image processing may be used to identify features such as a lesion within a treatment area on the skin of the patient. Based on this image processing, the projector may then selectively illuminate the lesion.

A work information projection system for estimating a position in a work site. The work information projection system includes a camera (13), a stereo camera (14), and a control device. The camera (13) is disposed toward a first direction, and acquires first environment information that is information of an object disposed in the first direction and around. The stereo camera (14) is disposed toward a second direction, and acquires second environment information that is information of an object disposed in the second direction and around. In a case where it is determined that the second environment information cannot be acquired, the control device estimates a position in a first coordinate system based on the first environment information.

The invention relates to the field of machine vision, and discloses a printing and dyeing defect detection method and system, and the method comprises the steps: carrying out the image collection of the surface of a printed and dyed object through a multi-angle imaging device disposed on a production line, and obtaining an original image set containing multiple visual angles; performing alignment processing on the original image set to obtain a registration image set under a unified coordinate system; a three-dimensional point cloud model is generated according to the registration image set, and geometric structure features of the surface of the printed and dyed object are obtained; projecting color information in the registration image set to the three-dimensional point cloud model to obtain a three-dimensional color distribution model; pixel points in the three-dimensional color distribution model are grouped to obtain boundary features of a color difference area; and according to the boundary features of the color difference region, analyzing the coordinate range of the color difference region in the three-dimensional color distribution model, and determining the spatial position coordinates of the defect. According to the technical scheme, color information projection distortion caused by geometric fluctuation is avoided, and the identification accuracy of the color difference area is improved.

The invention discloses a black-start overcurrent suppression method and device, electronic equipment and a storage medium, and relates to the field of artificial intelligence or other related technical fields, and the method comprises the steps: collecting electric power data of an optical storage grid-connected system, carrying out the preprocessing of the electric power data, and extracting an electric power feature vector; the electric power feature vector is input into a virtual impedance coefficient optimization model, a target virtual impedance coefficient is output, and the virtual impedance coefficient optimization model integrates an attractor trend strategy, a coupling interference strategy and an information projection strategy for dynamic optimization; and the power data and the target virtual impedance coefficient are substituted into a virtual impedance calculation formula, a virtual impedance value is calculated, a reference signal of an inverter control loop is adjusted based on the virtual impedance value, and overcurrent caused by black start is suppressed. According to the invention, the technical problem that the overcurrent suppression stability is low when the overcurrent is suppressed based on a fixed virtual impedance control strategy in the prior art is solved.

The invention provides a model resource processing method and device, a computer program product and electronic equipment. The model resource processing method comprises the following steps: determining a geometric grid corresponding to a target neural field model based on the target neural field model and original images under a plurality of observation view angles; performing UV expansion on a geometric grid corresponding to the target neural field model to obtain a UV layout of the geometric grid, and performing reverse solution based on the UV layout of the geometric grid to obtain a PBR material map of the geometric grid; semantic instance segmentation information of each original image is determined through semantic instance segmentation, and the semantic instance segmentation information of each original image is projected to a surface corresponding to the geometric grid to obtain a semantic instance label corresponding to the geometric grid; and packaging the geometric grid, the PBR material map of the geometric grid and the semantic instance label corresponding to the geometric grid into a standardized PBR asset package based on a preset packaging format so as to improve the reusability of model resources.

The application relates to a laser radar and camera external parameter online calibration method and device based on a deep neural network and a storage medium, wherein the method comprises the following steps: building a laser radar and camera joint calibration platform; calibrating camera internal parameters; synchronously time stamping; acquiring a gray image and a point cloud and completing pretreatment; initializing external parameters; converting the point cloud to a camera coordinate system; based on a random mask mode, respectively covering the point cloud and the gray image according to a preconfigured covering ratio; projecting the covered point cloud to an image plane to obtain depth information projection and reflectivity information projection; constructing an external parameter online calibration network, which is composed of an input branch, a prediction branch, a fusion branch and a result output module, and realizing external parameter calibration by fusing the features of the depth information projection, a depth map, the reflectivity information projection and a covered gray image; training the external parameter online calibration network; and completing external parameter online calibration. Compared with the prior art, the application has the advantages of good calibration effect, wide application range and the like.

A projection system (3) is provided with a projection device (31) that projects information (I) onto or near to a production facility (2) that includes a conveying device (21) for conveying a vehicle body (V), wherein the information (I) includes facility information (IA) relating to the production facility (2).

The application discloses a kind of multi-feature regularization rendering methods and systems for sparse view angle.It is easy to appear appearance overfitting, detail loss, geometric drift and floating artifact and other problems for traditional three-dimensional Gaussian splashing (3DGS) new view synthesis under sparse view angle input, the joint optimization mechanism of pixel reconstruction constraint, multi-source feature consistency regularization, depth information projection consistency, scale-insensitive depth geometry regularization is used to realize stable learning and high-quality new view rendering of three-dimensional scene representation.The application does not need large-scale pre-training and additional artificial labeling, is compatible with 3DGS, training cost is controllable and inference keeps real-time rendering ability, significantly improves the quality and stability of new view synthesis under sparse view angle input, and can be widely applied to few-shot three-dimensional reconstruction, real-time new view synthesis and high-quality scene visualization and other fields.

The application discloses a vehicle information projection method, device, equipment and medium. The method comprises the following steps: acquiring road condition satellite information and traffic satellite information on a target navigation route of a vehicle, and a real-time driving position of the vehicle; acquiring a pre-display distance input by a user and an information collection radius; calculating a data collection area and a data pre-collection area of the vehicle, querying the road condition satellite information and the traffic satellite information corresponding to the data collection area of the vehicle, and determining vehicle environment information of the vehicle; querying the road condition satellite information and the traffic satellite information corresponding to the data pre-collection area located at the pre-display distance, and determining vehicle pre-display information of the vehicle; screening objects in the vehicle environment information and the vehicle pre-display information, obtaining object space positions and object sizes of to-be-displayed objects, generating to-be-projected image information, and projecting the to-be-projected image information through a holographic projection vehicle lamp. The technical scheme of the embodiment of the application can improve the accuracy and efficiency of vehicle information projection.

The application provides a robot explainable intelligent computing method and system, which comprises the following steps: S1, collecting action behavior data and tactile representation data of a target object operation task of a robot in a working state; S2, extracting feature information of semantic description of the action behavior data and the tactile representation data, projecting feature information with strong correlation into a shared semantic subspace, and establishing a feature information vector matrix corresponding to the data and the task; S3, combining a symbolic planner with a specific tactile model based on a GEP algorithm, so as to achieve the goal of combining a long-term task structure of the robot with an operation strategy learned from tactile signals; and S4, based on a Bayesian trust model, cognitively fusing and calculating the robot action behavior data and the tactile representation data by using a probability method, so as to predict the next action of the robot. The problems of lack of explainability and transparency of a machine learning model are effectively solved.

The present invention discloses a method, system, and device for the dynamic fusion of multimodal perception data of an embodied intelligent body, including: S1. The embodied intelligent body collects image data in real time during motion; S2. Uses a multimodal large model to determine the point cloud filling density of the image data, and uses a mapping algorithm to fill the 3D point cloud of the camera image during data collection; S3. Generates a lidar feature map through point cloud data projection conversion; S4. Initializes and aligns the feature data of the multimodal large model; S5. Constructs a set of multi-scale, multi-channel, differentiable mutual information fine optimization methods to process the feature data; S6. Performs online monitoring and fusion self-correction operations of the multimodal large model; S7. Projects camera information onto the laser 3D point cloud, and continuously constructs a fused RGB-3D scene during the motion of the embodied intelligent body. The present invention can achieve accurate, robust, automatic calibration and fusion of external parameters between the lidar and the camera.

The application relates to a multi-level 3D target expression method, system and medium applied to an automatic driving scene, the method comprising the following steps: B1. A vehicle travels on a road, real-time road image data information is acquired based on a vehicle-mounted camera, real-time road point cloud data information is acquired based on a vehicle-mounted laser radar, and distance data information of an obstacle is acquired in real time based on a vehicle-mounted ultrasonic radar; B2. The road image data information is input into a convolutional neural network for identification, and is classified according to highway signs, and highway scene classification data information and non-highway scene classification data information are output; and B3. The road point cloud data information is projected to the road image data information based on the highway scene classification data information. The application not only facilitates selection of an optimal expression scheme for decision planning under suitable working conditions, but also solves the problem of perception and planning cooperation in different scenes.

The application provides a kind of engineering structure multidimensional variable non-directional cooperative measurement method and system, it is related to data determination technical field, comprising: obtaining the original observation data about the to-be-measured multidimensional variable of engineering structure;Based on the original observation data of mute working condition, estimate sensor noise covariance matrix;Combined with sensor noise covariance matrix, direct information matrix describing non-coupling information is sequentially constructed and composite information matrix describing cooperative information;Combined with load working condition original observation data, construct information projection regularization functional about composite information matrix, to suppress cooperative estimation divergence caused by cooperative information quantity condition number too large;According to the information non-directional filling rate in composite information matrix, determine the inhibition intensity of cooperative estimation divergence in information projection regularization functional;According to the inhibition intensity, update information projection regularization functional;Solve the updated information projection regularization functional, obtain the optimal estimated value of to-be-measured multidimensional variable.

The application discloses a phonon transport property calculation model, system and method based on wave-particle duality, which comprises a normal mode analysis module, a phonon wavelet transform module and a thermal conductivity calculation module, and unifies the particle nature and wave nature of phonons in a thermal conductivity prediction framework, wherein the normal mode analysis module is used for projecting atomic scale dynamic evolution information to a phonon mode space to obtain mode-resolved space-time evolution data; the phonon wavelet transform module is used for wavelet transforming the mode-resolved space-time evolution data to quantitatively characterize the coherent time and coherent length of phonons; and the thermal conductivity calculation module is used for simultaneously calculating the contribution of phonon particle nature lifetime and wave nature coherent time to thermal conductivity based on the mode-resolved space-time evolution data, the coherent time and the coherent length, thereby providing a phonon transport image closer to physical reality and greatly improving the research and development efficiency.

This invention discloses a federated transfer fine-tuning system using visual cues and contrastive learning, relating to the field of intelligent transportation. This invention addresses the contradiction between data constraints and the high data requirements of large-scale traffic scene model training. It introduces non-public or private data from multiple parties for large-scale model training through a federated learning approach that eliminates the need for direct data transfer. Simultaneously, by using a visual cue learning module and a contrastive learning loss module, the overall model training effect is effectively improved, further enhancing the performance of generative large models in traffic scene image question answering tasks. By utilizing the cross-modal projection mechanism of linear multimodal large models, and employing a process of projecting visual modal information onto the text modality for language model processing, the text generation capability of the large language model is preserved, avoiding the problem of traditional training methods based on training data answer labeling limiting model response performance due to answer limitations imposed by data content.

To provide a technology that enables more appropriate display of images for the driver and images for people around the vehicle. According to this invention, it contributes to Sustainable Development Goal 3, "Good Health and Well-being." [Solution] The information projection device comprises an acquisition unit that acquires vehicle information relating to a vehicle, and a projection unit that projects images. Based on the vehicle information acquired by the acquisition unit, the projection unit projects a rectangular image, which is projected linearly to the left and right sides of the vehicle, and when detecting the turning direction of the vehicle, it projects the image to display and transmit the turning direction.

The application relates to the field of machine vision, and discloses a printing and dyeing defect detection method and system. The method comprises the following steps: an image of a printing and dyeing object surface is collected by a multi-angle imaging device arranged on a production line to obtain an original image set containing multiple visual angles; alignment processing is performed on the original image set to obtain a registration image set in a unified coordinate system; a three-dimensional point cloud model is generated according to the registration image set to obtain the geometric structure features of the printing and dyeing object surface; color information in the registration image set is projected to the three-dimensional point cloud model to obtain a three-dimensional color distribution model; pixel points in the three-dimensional color distribution model are grouped to obtain boundary features of a color difference area; according to the boundary features of the color difference area, the coordinate range of the color difference area in the three-dimensional color distribution model is analyzed to determine the spatial position coordinates of the defect. Through the technical scheme, color information projection distortion caused by geometric undulations is avoided, and the recognition accuracy of the color difference area is improved.

The invention discloses a traffic emergency dispersion system based on unmanned aerial vehicle ground vehicle window information projection, which realizes accurate identification of spectral data of different glass materials and protective layers of a front vehicle window through a multispectral imaging and polarized light detection technology, and dynamically adjusts projection parameters in combination with a reflected light spot acquisition and quality evaluation technology. Projection information is ensured to be clear and visible; wherein the information comprises road congestion information, traffic management prompt information and the like. The system is composed of a multispectral camera, a polarization camera, a miniature projector, an edge calculation module and a communication module, can detect different glass materials and protective layers of a front car window in real time, assesses projection quality, and adjusts projection brightness, contrast and focal length according to needs. The system has the advantages of high precision, strong real-time performance, adaptability to complex environments and the like, can be widely applied to the fields of traffic emergency guidance and smart city management, and effectively overcomes the defects of the existing traffic dispersion means in the aspects of accuracy, timeliness and flexibility of information transmission.

The present disclosure relates to the technical field of data processing, and in particular, to a multi-target tracking method, device, storage medium and electronic equipment. The method comprises: determining image data sequence and point cloud data information of a current time in a current scene, wherein the elements in the image data sequence are moving targets; projecting the point cloud data information onto the corresponding elements in the image data sequence to generate a target sequence; generating a correlation matrix according to the feature information of each element in the target sequence; and matching the latest point cloud data information and the correlation matrix to obtain a tracking result. In this way, the point cloud data information corresponding to the three-dimensional space can be projected into the image data sequence corresponding to the two-dimensional space, the elements in the image data sequence are used to filter part of the point cloud, the calculation amount when generating the correlation matrix is reduced, the target tracking efficiency is improved, and meanwhile, the uniformity of the image data sequence and the point cloud data information in space and time is ensured, and the accuracy of the target tracking is improved.