98 results about "Perception model" patented technology

The present application relates to a kind of based on multi-source industrial data's car body welding quality intelligent perception, prediction and decision-making method and system, belong to the technical field of intelligent manufacturing and industrial data analysis.The present application obtains multi-source high-frequency synchronous original signal by sensor array and robot bus;Perform space-time alignment and multidimensional feature mining;Quality perception model is constructed based on multi-path parallel network and cross-modal attention;Quality evolution trend and electrode cap life are evaluated using time series prediction model;Integrate expert rules and reinforcement learning to make process compensation decision.The present application realizes the accurate perception of welding quality, forward warning and closed-loop real-time compensation, significantly improves the perception accuracy, production robustness and prolongs the service life of electrode.

This invention discloses a robot visual-tactile perception method and system based on transformer and contrastive learning, belonging to the field of robotics technology. The method introduces an enhanced Swing Transformer model as the perception model. This enhanced model uses the Swing Transformer as the backbone network and incorporates a coordinate attention mechanism. Global information is extracted through the sliding window attention mechanism of the backbone network, while local feature extraction is enhanced through the coordinate attention mechanism. This effectively extracts and fuses global and local features from visual and tactile data. In the pre-training stage, the enhanced Swing Transformer model undergoes multimodal contrastive learning training, specifically intra-modal and inter-modal contrastive learning, generating richer and more accurate discriminative features, thereby effectively improving the robot's intelligent perception and manipulation capabilities. This method can effectively extract features in multimodal data fusion and utilize the correlation and synergistic effects of features between different modalities to achieve deeper multimodal data fusion.

The application discloses a kind of intelligent unmanned vehicle environment perception systems based on multi-sensor information fusion, it is related to unmanned vehicle technical field.The system includes: sensor module, for collecting multi-source heterogeneous data, including GPS / IMU, laser radar, camera and millimeter wave radar;Data acquisition and fusion module, through the Kalman filter to multi-source data space-time synchronous fusion;Environment perception module, based on deep learning identification road boundary, obstacle and traffic sign, constructs dynamic perception model;High-precision map construction module, using SLAM technology and semantic information constructs and updates semantic three-dimensional map;Positioning algorithm module, fusion visual odometry, fusion data and high-precision map, calculates vehicle real-time pose;And control module, according to positioning and perception result carries out path planning and navigation control, the application provides a kind of intelligent unmanned vehicle environment perception systems based on multi-sensor information fusion, can be autonomously navigated and stably driven under complex and severe environment.

The present application relates to the technical field of intelligent auxiliary driving of automobiles, and particularly relates to a blind area monitoring system based on a low-computing-power processing unit of a commercial vehicle platform, a blind area range calibration module calibrates a blind area monitoring camera, and provides a preprocessing operation; an image processing module and a perception algorithm module are used for receiving blind area perception information of the camera, judging blind area perception results through an image processing algorithm and a deep learning algorithm, and customizing a blind area monitoring model; an information processing module provides computing power support for the blind area perception model, customizes the blind area monitoring model, and improves inference speed and inference accuracy of the model; a warning prompt module is used for feeding back potential danger warning information to a driver. Through combination of blind area range calibration and customized strategies such as image processing and perception algorithms, an efficient real-time blind area monitoring system based on a low-computing-power information processing unit is realized, which can send warning information to the driver in real time, has low system power consumption, and has strong robustness and scalability.

The application provides a virtual environment perception model self-evolution method, device, equipment, medium and product, and belongs to the intelligent interaction technical field of meta universe. The method comprises the following steps: acquiring perception flow generated by processing perception data of a virtual environment by a first perception model; acquiring a reference true value flow of the virtual environment from a bottom layer of the virtual environment; determining a perception difference between the perception flow and the reference true value flow; in the case that the perception difference satisfies a model update triggering condition, automatically labeling perception data corresponding to the perception difference in the perception flow by using the reference true value flow to generate incremental training data; and updating the first perception model based on the incremental training data to obtain a second perception model. The application realizes closed-loop self-evolution of the perception model by constructing a visual-semantic double-flow checking architecture, so that the recognition ability of an intelligent agent can continuously evolve with running time.

The application relates to a model iteration method and device, electronic equipment and a storage medium, comprising: obtaining original sensor data collected by a vehicle terminal in a preset time window before and after a trigger time in response to a preset event trigger; clustering the obtained multiple groups of original sensor data using an unsupervised clustering algorithm, identifying an outlier cluster from the clustering result, and marking the original sensor data corresponding to the outlier cluster as long-tail candidate data; extracting images from the long-tail candidate data, using a cloud perception model and a vehicle terminal perception model to respectively infer the images, screening out images with inconsistent inference results of the two perception models as difficult example images, and labeling the difficult example images according to the inference result of the cloud perception model; optimizing and training the vehicle terminal perception model using the labeled difficult example images, and deploying the optimized perception model to the vehicle terminal. Thus, the problem of insufficient performance of the perception model in a long-tail scene is effectively solved.

This invention relates to a method and apparatus for four-dimensional occupancy grid panoramic perception based on visual information. The method acquires multi-view images, inputs these images into a pre-constructed four-dimensional panoramic perception model based on visual information, and outputs a four-dimensional panoramic occupancy grid prediction result. The four-dimensional panoramic perception model based on visual information includes an encoder module, a decoder module, and a query vector propagation module. The encoder module extracts three-dimensional volume features, the query vector propagation module updates the four-dimensional query vector, and the decoder enables the interaction between the three-dimensional volume features and the four-dimensional query vector. Compared with existing technologies, this invention has advantages such as generating the final four-dimensional panoramic occupancy grid prediction result in an end-to-end, streaming manner, eliminating the need for extensive post-processing.

This invention discloses a method for quantifying and classifying suspected electricity theft based on multi-dimensional weighted fusion. The method includes: collecting multi-source heterogeneous data through a graded suspected electricity theft early warning analysis platform; extracting abnormal gradient features through grid load anomaly gradient identification; constructing personalized behavioral benchmarks based on a user electricity consumption behavior profile perception model; quantifying the degree of electricity consumption deviation using a dynamic baseline electricity theft deviation assessment model; generating a comprehensive quantitative value for suspected electricity theft through multi-dimensional weighted fusion; classifying the suspected electricity theft according to preset standards; and utilizing dedicated hardware acceleration and parameter optimization. Six functional units work together to achieve the entire process of data collection, feature extraction, deviation assessment, and quantitative classification. This method overcomes the limitations of traditional single-dimensional detection, dynamically adapts to grid operating conditions and user characteristics, improves the accuracy of suspected electricity theft identification and the rationality of classification, provides efficient technical support for power regulation, and ensures the safe and stable operation of the power grid.

The application discloses a kind of postoperative debilitation risk dynamic prediction methods based on deep learning, it is related to deep learning technical field, including, based on key feature subset, using key feature discriminant layer, trajectory evolution prediction layer and trajectory probability calculation layer, construct dynamic trajectory perception model, output debilitation risk comprehensive state quantity;Gradient back propagation and parameter contribution degree analysis are carried out to debilitation risk comprehensive state quantity, generate feature influence intensity and evaluate result stability, obtain core risk factor;Based on core risk factor, construct feature adjustment strategy set and carry out analog deduction, obtain training sample, feedback to dynamic trajectory perception model executes online learning update.The application carries out gradient back propagation and parameter contribution degree analysis to debilitation risk comprehensive state quantity under deep learning framework, improves the dynamic adaptation capability of postoperative debilitation risk prediction, prediction timeliness and clinical interpretability.

The application provides a perception method based on a light-weight geometric information decoupling graph convolution point cloud perception model, and belongs to the technical fields of point cloud perception and deep learning. The method comprises the following steps: obtaining candidate region retrieval by discretizing the original point cloud data space coordinates through 2D voxel down-sampling based on the light-weight geometric information decoupling graph convolution point cloud perception model; obtaining adjacent edge index by using a scale factor-based expansion KNN algorithm based on the candidate region. The encoder performs multiple graph convolution and down-sampling processes based on the adjacent edge index and the point cloud subset, and the decoder performs multiple up-sampling and graph convolution processes based on the point cloud subset and the adjacent edge index. The application can complete point cloud classification by extracting multi-layer graph convolution features through the encoder. For point cloud segmentation and target identification tasks, the decoder gradually reconstructs spatial details in the up-sampling stage, effectively compensating for the loss of geometric information caused by down-sampling in the encoding process.

The application discloses a kind of multi-modal forest fire target detection method and system based on thought chain, belong to forestry intelligent monitoring technical field.It includes the following steps: collecting and pre-processing paired visible light and thermal infrared forestry scene image, after data labeling, pre-training large visual language model is trained using supervised fine-tuning, and the basic perception model is constructed;Composite reward function is constructed, and the reasoning process of the basic perception model is optimized using group strategy optimization reinforcement learning algorithm;By inputting the preset thought chain prompt to the optimized basic perception model, the model synchronously generates structured natural language text.Compared with the prior art, the application has the advantages that the "black box" model is converted into an interpretable and traceable decision report by the thought chain, greatly enhancing the credibility of the system.

This application discloses an adaptive intelligent control method for ventilator parameters, comprising: constructing a disease perception model and inputting a sequence of state basis vectors into it, outputting a probability vector representing the probability of the patient being in the acute, stable, and recovery phases; determining a dynamic weight vector for state modulation through a weighted generative neural network and combining it with the state basis vectors to determine a dynamic state vector; designing a hierarchical reward function, including basic safety sub-item rewards, iatrogenic risk sub-item rewards, and long-term prognosis sub-item rewards, determining a dynamic weight vector for reward synthesis based on the disease phase probability vector, and weighted summing it with the reward values ​​of each sub-item to obtain a total reward value; based on the dynamic state vector, determining a ventilator parameter control scheme through a reinforcement learning agent, performing safety verification, and if the verification fails, selecting a safety fallback scheme as the control scheme for parameter control, and feeding the verification results back to the agent for strategy optimization. This application can improve the accuracy and safety of parameter control.

This invention provides an intelligent UAV patrol method and system based on edge-cloud collaboration in the field of UAV perception and artificial intelligence. The method includes: Step S1, the UAV collects visible light video and infrared video and inputs them into a multimodal fusion perception model to obtain fusion perception results; Step S2, an edge-cloud collaboration mechanism is automatically triggered based on uncertainty scoring, or control commands are generated; Step S3, the ground station gathers the basic target data carrying temporal information transmitted back by the UAV and inputs it into a spatiotemporal attention model to obtain atomic event recognition results, which are then uploaded to the server; Step S4, the server inputs the atomic event recognition results and image slices uploaded by the edge-cloud collaboration mechanism into an anomaly event detection model to obtain anomaly event detection results. The advantages of this invention are: it greatly improves the UAV's accurate perception capability throughout the entire time in wide-area complex scenarios, the interpretability and understanding capability of long-term events, and the dynamic collaborative efficiency of edge-cloud resources.

This invention discloses a collaborative confidence fusion perception method for autonomous driving, belonging to the field of autonomous driving. This method eliminates the differences in perception models between different intelligent agents, makes full use of the advantages of sensor data from different modalities, and avoids the operation of forcibly converting visual images that are not good at extracting BEV features into BEV features in order to achieve feature unification. Based on fully leveraging the advantages of different modal data features, learning to adapt to the features that specific models are good at, and the inherent collaborative advantages of collaborative perception, it achieves better collaborative perception results.

The application discloses a kind of robot gas source positioning method, belong to intelligent mobile robot technical field.For the intermittent, broken and strong turbulent disturbance of gas plume in complex unknown environment, the method first constructs a relative time-varying perception model, and converts the historical observation data to the local coordinate system of the robot;By introducing motion uncertainty weight and turbulent perception time decay mechanism, adaptive filtering and dynamic updating of historical observation information are realized;Further design dynamic window mechanism and multi-modal feature fusion strategy to improve the modeling accuracy of local gas concentration field in complex environment;At the same time, combined with the improved dynamic window path planning method driven by environment, the high concentration area and the direction of adverse wind are cooperatively guided.The method can run in unknown environment, improve the accuracy, robustness, real-time and efficiency of source search, and is suitable for industrial leakage detection, disaster rescue and dangerous gas search and other application scenarios.

This application discloses a method, apparatus, device, and medium for transferring a perception model, which facilitates the transfer from the source domain to the target domain in point cloud perception tasks, enables the reuse of point cloud data from the original perception model, and improves the generalization ability of the perception model for different application scenarios. The method for transferring a perception model provided in this application includes: transferring the point cloud data of the original perception model to the target perception model according to the application scenario information of the target perception model obtained, thereby obtaining a truth database of the target perception model, wherein the truth database includes point cloud and label information of the perceived target; and training the model based on the truth database of the target perception model to obtain the target perception model.

The application discloses a kind of lithium battery fault intelligent diagnosis method based on deep learning, comprising the following steps: collecting lithium battery operating data and carrying out quality correction, obtain operating data set;Input to operating condition perception model generates multi-scale feature vector representing different operating states;Initial fault diagnosis model is obtained by initial fault class label, initial fault probability value and diagnostic feature vector;Form uncertainty feature vector set by confidence evaluation;Perform bidirectional backtracking correlation analysis, extract associated time period operating data subset;Diagnosis is carried out again, generates review result and carries out consistency comparison with initial result;Conflict result generates conflict mode information and is stored to fault mode library, without conflict when directly output final diagnosis result;The diagnosis result of new data is corrected using fault mode library Pattern matching, output corrected final diagnosis result.The application can significantly improve fault detection rate, reduce misjudgment rate and shorten diagnosis duration.

This invention provides a motion environment perception method based on biomechanical features and gait adaptation, belonging to the field of human-computer interaction perception. The method includes: acquiring the subject's original motion feature sequence and biomechanical feature vector; scaling and aligning the physical dimensions using a biomechanical scaling matrix to obtain a motion feature tensor; inputting the motion feature tensor and biomechanical feature vector into a multi-task temporal convolutional network for feature-level linear modulation to obtain modulated deep features; processing these features through a backbone network, with the classification branch outputting terrain categories and the regression branch outputting initial values ​​of environmental geometric parameters; freezing the backbone network parameters; fine-tuning the regression branch based on an adaptive loss function driven by gait phase and foot arch features to obtain a motion environment perception model; inputting real-time data from the subject; and outputting terrain categories and environmental geometric parameters. This invention solves the technical problem of low accuracy in environmental geometric parameter estimation caused by biomechanical differences between individual subjects in existing technologies.

The application discloses a kind of multi-modal sporadic abnormal environment generalization perception methods and systems for data closed loop, it is related to automatic driving and intelligent transportation system technical field.The method comprises: using a variety of sensors real-time collection and pre-processing environmental data, obtains the high-dimensional feature representation of each mode coding through the feature extraction module of perception model;Utilize the sporadic abnormality in high-dimensional feature representation to be identified by exception detection model, when detecting abnormality, determine abnormal state by reconstruction difference and trigger adaptive adjustment mechanism;Through multi-modal data fusion technology, the fusion feature is obtained by fusing each mode feature;Through environmental generalization learning algorithm, introduce environment-related constraint loss, prompt the perception model to learn environment-independent general features to optimize perception algorithm.The application aims to realize the accurate detection and efficient processing of sporadic abnormality in complex environment through data closed loop feedback and environmental generalization learning, improve the accuracy and robustness of perception task.

Systems and methods include detecting obstacles and drivable areas by an autonomous vehicle by inputting image and map data into a neural network to extract feature vectors. A transformer encoder converts these vectors from camera space to Bird's Eye View (BEV) space. A detection head identifies objects, and a segmentation head generates a BEV map showing objects and drivable surfaces. Attributes from both heads are compared, and the segmentation head's weights are updated accordingly, resulting in an updated BEV segmentation map output by the updated segmentation head.

The application discloses a method and system for computing and resource collaborative scheduling of a wisdom calculation center facing energy efficiency optimization, belongs to the technical field of data centers and high-performance computing, and comprises the following steps: establishing a multi-level resource and energy efficiency perception model of the wisdom calculation center to acquire real-time perception data; constructing a full-stack energy efficiency evaluation and prediction model based on the real-time perception data; solving an optimal collaborative scheduling strategy through a joint optimization engine based on the full-stack energy efficiency evaluation and prediction model, and outputting a prediction value according to the collaborative scheduling strategy; executing the collaborative scheduling strategy, monitoring actual energy consumption and task performance data after execution of the collaborative scheduling strategy, taking the difference between the monitoring result and the prediction value as a feedback signal, and inputting the feedback signal into the full-stack energy efficiency evaluation and prediction model to realize online self-adaptation and closed-loop optimization of the model. The application realizes optimization of the overall energy efficiency ratio of the wisdom calculation center by constructing a unified energy efficiency model and intelligent collaborative decision-making.

The application discloses a social perception recommendation method under multi-category sensitive link relationship protection and relates to the technical field of big data analysis. The application generates vector representations of users and items by using a heterogeneous graph neural network, introduces a mask protection mechanism of multi-type edges, solves the privacy leakage problem of multiple types of sensitive links, and thus prevents reasoning attacks of attackers based on background knowledge. A heterogeneous social perception model based on a session is constructed based on a session and multi-type nodes, recommendation performance is improved, the problem that multi-type node information cannot effectively act on a recommendation model is solved, and the problem of limited data use in a real scene is solved.

A method for enabling a resilient lane-level navigation in harsh weather conditions is provided. The method includes collecting a dataset associated with real-world driving conditions during harsh weather conditions using artificial intelligence (AI). The dataset includes at least one of a plurality of images, a plurality of sensor readings, and corresponding metadata. The method also includes setting a generator network and a discriminator network to create a plurality of images resembling real driving scenes under the harsh weather conditions. The method also includes generating synthetic data by employing a pre-trained AI model by varying one or more weather conditions fed into the AI model. The method also includes incorporating the generated synthetic data into a training dataset for training perception models used in autonomous driving systems. The method also includes optimizing a navigation solution using trained perception models, weather information, and sensor data for enabling the resilient lane-level navigation.

The application discloses a kind of based on image recognition's road surface state intelligent perception method and system.The system includes: image pre-processing module, fusion ESCA attention mechanism and LCAhead's road surface state intelligent perception module, model training module, image classification identification module and risk classification and driving guide module.The method includes the following steps: data enhancement is carried out to data image;Based on EfficientNetB0 model, fusion ESCA attention mechanism and LCAhead module;Configuration training hyperparameter and based on road environment perception dataset updates model weight;Data are input into ESCA-LCA Net road environment perception model training, realize the real-time classification identification of road environment state, and obtain corresponding risk classification and driving guide according to identification result.The application can improve the identification ability of multi-dimensional fine-grained road surface features, realize the accurate distinction and comprehensive judgment of dry and wet state, snow, water, unevenness and other types of road surface features.

A scene layout anomaly perception text-to-image large model hallucination detection method, system and device, the method is: acquiring scene layout anomaly detection data and preprocessing, obtaining object instance set; Construct a scene layout anomaly perception model for text-to-image hallucination detection; Including double flow graph module, cross-modal structure alignment module and anomaly sorting module; Train the cross-modal structure alignment module and the anomaly sorting module to obtain the model weight file of the training iteration; Read the model weight file and perform hallucination detection on the text-to-image large model; The invention decouples the detection of hallucination into two dimensions of semantic context perception and geometric structure modeling of object instance through the innovative semantic-geometric double flow interaction and cross-modal alignment framework, the consistency between the two is inferred through the cross-modal interaction Transformer network, the mismatch signal of semantic and geometry is accurately captured, and the fine-grained hallucination such as unreasonable object attribute and mismatched object relationship in the generated image is accurately positioned.

This invention discloses a three-dimensional measurement method for tomato internode length. Addressing the geometric errors easily caused by greenhouse shading, stem bending, and direct two-dimensional pixel ranging, the method acquires a single-frame color image and depth data of the plant. The color image is input into a feature-enhanced multi-task perception model, which outputs an internode instance segmentation mask and two-dimensional coordinates of key points at both ends. Candidate three-dimensional point clouds are extracted based on pixel correspondence, and the three-dimensional coordinates of the internode endpoints are generated from the neighborhood depth of the key points. The candidate point clouds are subjected to voxel downsampling, statistical outlier filtering, depth bandpass filtering, and spatial density clustering. The target main stem point cloud is obtained by combining spatial distance and depth consistency evaluation. Principal component analysis is used to estimate the principal axis, and axial slicing is performed. The centerline is extracted and fitted using a cubic B-spline curve. The three-dimensional coordinates of the endpoints are projected onto the fitted curve, and the outer endpoints of the curve are linearly extrapolated and compensated along the corresponding tangential direction. The arc length of the spatial curve is calculated to obtain the three-dimensional measurement result of the internode length.

The present invention discloses a method for establishing a correlation between LiDAR sensors, so that a perception model created by means of a first LiDAR sensor is used with a second LiDAR sensor having a different interpretation of point intensity than the first LiDAR sensor. Said method comprises the steps of: placing the first LiDAR sensor at a LiDAR sensor position (1) to perform a sensing operation by means of the first LiDAR sensor (101); placing the second LiDAR sensor at the LiDAR sensor position (1) to perform a sensing operation by means of the second LiDAR sensor (102); comparing the point intensity values of a first point cloud and a second point cloud at obstacle points (104), wherein the first point cloud is obtained from a sensing operation performed by the first LiDAR sensor and the second point cloud is obtained from a sensing operation performed by the second LiDAR sensor; and deriving a point intensity correlation function between the first and second LiDAR sensors based on the comparison of the point intensity values (105).