599 results about "Fog" patented technology

The invention relates to a pipe network station inspection task execution method and system based on multi-modal fusion, and belongs to the technical field of industrial facility detection. According to the method, multiple types of sensors are carried through an unmanned aerial vehicle and a ground robot, illumination, temperature and humidity, wind speed and rain and fog data are collected in real time, and an optimal sensor combination is dynamically activated; a feature level fusion strategy is adopted, the weight of each modal is adjusted in combination with environmental parameters, and the defect detection accuracy is improved; the system performs digital twinborn simulation verification on an abnormal result, so that the false alarm rate is reduced; an inspection path is adaptively adjusted according to a detection result, and a high-risk area is emphatically scanned; the edge computing nodes realize real-time data processing, and upload key information after compression; the maintainer rechecks the result through the AR glasses and marks a misinformation case; according to the method, the problems of poor environmental adaptability and insufficient utilization of multi-modal data of traditional inspection are solved, and the detection efficiency and reliability are improved.

The invention relates to a sea fog quantitative forecasting method based on a period enhanced Transform, and the method comprises the steps: firstly constructing a Transform model as a forecasting model, enabling the Transform model and a self-attention mechanism to better capture the complex nonlinear relation and long-distance time dependence related to the formation of sea fog, and being superior to a conventional statistical method and an early-stage RNN / LSTM model, a data set for model training is constructed based on a time period enhancement technology and visibility standardization mapping, and the explicit time period enhancement technology enables the model to more accurately learn and forecast seasonal and daily change rules of sea fog; and a sample amplification and training set balancing strategy is executed on the data set, so that the problem of sparse fog samples is effectively solved, and the forecasting capability of the model on key low-visibility events is improved. According to the sea fog quantitative forecasting method based on the period enhanced Transform, the precision, timeliness (forecasting per hour) and spatial resolution of sea fog forecasting of a target area can be improved, and sea fog data characteristics can be effectively processed.

The invention discloses a target identification method for multi-sensor data fusion, particularly relates to the technical field of data fusion, and comprises a dynamic weight fusion module, a double-branch neural network and an abnormal sensing compensation system. Data are synchronously collected through a visible light camera, a thermal infrared imager and a millimeter wave radar, and a standardized feature map is generated through sensor specificity preprocessing; the dynamic weight fusion module generates an adaptive weight matrix based on the real-time confidence score and the environmental parameters, emphasizes infrared data when illumination suddenly changes, and improves intelligent distribution of radar weights in a rain and fog environment; the combined feature map after weight fusion is input into a double-branch neural network, a channel self-calibration branch suppresses interference noise, and a target category and coordinates are output after residual connection optimization; when the recognition confidence is insufficient, the abnormal compensation system starts visible light Wiener filtering restoration, generative adversarial network infrared compensation and radar multi-frame accumulation algorithms, and the system reliability is ensured when a single sensor fails.

The invention discloses a transformer substation intelligent patrol recognition algorithm optimization method based on multi-modal fusion, and belongs to the technical field of transformer substation patrol, and the method comprises the following steps: synchronously obtaining a visible light image, an infrared thermal imaging spectrum, an ultrasonic partial discharge signal and environmental parameters, and building a four-dimensional space-time calibration matrix; a dual-channel attention mechanism is adopted to realize multi-source data fusion, a spatial alignment channel and a feature enhancement channel are included, the spatial alignment channel realizes pixel-level registration of visible light and infrared images through affine transformation, and the feature enhancement channel utilizes a graph neural network to construct topological correlation mapping of acoustic, thermal and optical features; constructing a three-stage training architecture containing noise injection; and adding an adversarial generative network in a preprocessing layer to simulate rain fog and electromagnetic interference environments. According to the method, the identification threshold is updated online by deploying a sliding time window mechanism in an output layer, and the problems of poor feature extraction robustness and high false alarm rate of a traditional algorithm in complex environments such as strong electromagnetic fields and rain and fog are solved.

The invention relates to the field of photovoltaic module detection, and discloses a module hot spot defect infrared and visible light fusion detection method, which comprises the following steps: multi-mode acquisition: multi-mode data of a photovoltaic module are synchronously acquired, and the multi-mode data comprise infrared temperature original data, visible light polarization images and environmental parameters; data preprocessing: preprocessing the multi-modal data based on a radiation transmission model and a polarization optical principle to obtain preprocessed data including an infrared temperature field and a visible light image; and dynamic feature extraction: extracting a space-time gradient tensor and a Riemannian curvature tensor of the infrared temperature field from the preprocessed data based on differential geometry and a computer vision algorithm. By synchronously acquiring multi-modal data, based on a radiation transmission model and a polarization optical principle, interference of atmospheric attenuation and environment temperature and humidity on infrared temperature measurement and influence of uneven illumination and fog effect on a visible light image are eliminated, and dynamic characteristics of an infrared temperature field and structural characteristics of visible light are mined.

According to the severe weather adaptive road violation real-time monitoring method and system provided by the invention, the image quality is dynamically judged by using the visibility index, and accurate traffic target identification and violation behavior detection are realized by combining image enhancement and a gating feature fusion mechanism. The system comprises a video acquisition module, a visibility evaluation module, an image enhancement module, a feature extraction and fusion module, a target detection module and a violation recognition module, can maintain high-accuracy recognition performance in various severe weathers (such as rain and fog), and is suitable for the field of intelligent traffic and urban safety management.

The invention discloses an unmanned aerial vehicle intelligent obstacle avoidance decision-making method and system for a complex environment, and belongs to the technical field of unmanned aerial vehicle obstacle avoidance. The method comprises the steps of S1, environment perception and data acquisition, S2, multi-source data hierarchical fusion processing, S3, complex environment obstacle recognition and risk assessment, S4, dynamic obstacle avoidance path decision and optimization, and S5, obstacle avoidance execution and real-time adjustment. Omnibearing acquisition of obstacle distance, environment images, unmanned aerial vehicle attitude and position data is realized; the method combines an improved YOLOv8 model (adding a CBAM attention mechanism), strengthens the obstacle feature extraction capability under complex weathers such as rain and fog, strong light and dust, can accurately distinguish static obstacles such as buildings and trees from dynamic obstacles such as pedestrians, vehicles and other aircrafts, and solves the problems that a single sensor perceives blind areas and is insufficient in precision in a complex environment.

The invention discloses a visibility inversion method based on millimeter wave cloud radar and laser radar data fusion, and relates to the technical field of meteorology, and the method comprises the steps: guiding a millimeter wave cloud radar and a laser radar to carry out data correction and equivalent relation construction through employing a fog drop particle size distribution function measured by a fog drop spectrometer in a sample scene in advance; and then applying the correction coefficient and the equivalent function relationship to a target scene to perform multi-source data correction and fusion on the millimeter wave cloud radar and the laser radar and invert visibility. The method can fuse complementary advantages of the laser radar and the millimeter wave cloud radar; the introduction of the fogdrop particle size distribution function effectively avoids the distribution hypothesis of fogdrop shapes in the traditional model, can weaken the seasonal drift of the traditional empirical coefficient, significantly improves the accuracy and reliability of visibility inversion, and achieves the high-precision visibility monitoring under various weather conditions from clear sky to dense fog.

The invention belongs to the technical field of sea fog monitoring, and particularly relates to a sea fog monitoring method and system based on multi-source satellite remote sensing data, and the method comprises the steps: collecting satellite radiation and microwave observation data, discriminating the cloud top phase state in combination with infrared ground microwave channel response differences, dividing clear sky and low cloud shielding scenes, and outputting identification data. And sea fog is detected in combination with the single-time infrared brightness temperature difference and the empty pixel parameters, optical parameters are inverted, and a sub-scene sea fog physical parameter set is formed. Carrying out normalized coupling reconstruction on the parameter set, constructing a sea fog exclusive feature space, separating a fog region from a non-fog region through phase state constraint to obtain an initial judgment feature field, correcting the edge of the fog region through neighborhood correlation and brightness temperature gradient constraint, strengthening spatial continuity to generate a spatial distribution field, locking a credible region, and outputting a final sea fog detection result. According to the invention, through layer-by-layer progressive design of scene division, feature optimization and multi-source verification, the precision and integrity of sea fog identification are significantly improved.

The invention relates to sensing and tracking of intelligent driving in rain, snow, fog, night and large shielding scenes. In order to solve the problems of target invisibility, track interruption and misconnection caused by low visibility, a shielded dynamic target identification and tracking method based on deep learning is provided; according to the method, under a unified spatial index, sparse point cloud and road topology, passable and traverse areas, static shielding volume coding, ray marking visibility boundary and shielding entrance and exit are carried out; generating a motion voting field with consistent visibility through multi-agent situation reasoning, performing normal directional enhancement according to a passable boundary, and extracting a risk corridor; voxelization is carried out on the multi-frame point cloud in the corridor, a dynamic sparse voxel map is constructed, a voting field is used for gating cross-frame edge connection, occupation changes and infinitesimal displacement are aggregated, and three-dimensional verification candidates, limited state estimation and a time continuous track are obtained; outputting the target and the corridor to which the target belongs, and forming a potential conflict zone according to the intersection of the target and the own vehicle path; the reproduction rate and the advance are improved, and cross-lane misconnection is reduced.

A density perception image defogging method in combination with physical prior comprises the following steps: S1, acquiring a fog-containing image, marking a fog area and a fog-affected area, constructing a data set D1 through size unification, normalization and multi-dimensional data enhancement, and dividing the data set D1 into a training set and a verification set according to a proportion; s2, a density perception defogging network is constructed, the density perception defogging network comprises an encoder sub-network, a middle layer sub-network and a decoder sub-network, core modules are a gridding atmospheric light attention module and a space detail enhancement module, and a GAAB integrates a multi-scale density perception convolution module, a grid perception atmospheric attention module and a feature fusion module; fog density features can be accurately captured, and physical prior can be fused; s3, inputting the data set into a defogging network to generate a multi-scale feature map; s4, carrying out training optimization on the network by adopting a combined loss function of L1 loss and perception loss; s5, evaluating the network performance through the verification set; according to the invention, adaptive defogging in different fog concentration scenes is realized, and the definition of the defogged image is effectively improved.

The invention belongs to the field of atmospheric boundary layer height prediction, and discloses a sea-land atmospheric boundary layer height prediction method and system under a sea fog condition. The method comprises the steps of sample screening and key meteorological element selection, multi-source data preprocessing, boundary layer height calculation, a Richardson number method after critical value optimization, a gas block method, a specific wet method, a potential temperature gradient method and laser radar wavelet covariance transformation calculation. And verifying the calculation method, screening and optimizing the optimal method, and evaluating the performance of the method. According to the method, multi-source observation data and a numerical simulation result are fused. Through a mode of combining physical process analysis and statistical verification, errors of different calculation methods can be quantified, and an algorithm combination most suitable for sea fog conditions can be identified. Calculation deviation caused by special conditions such as weak turbulence and strong temperature inversion in the sea fog environment can be improved.

The invention discloses an automobile laser radar point cloud semantic intelligent completion method in rainy and foggy weather, and particularly relates to the technical field of automatic driving perception. The method comprises the following steps: firstly, collecting time sequence point cloud data and inertial navigation information of a vehicle under rain and fog conditions, and constructing a sparse point cloud sequence; then constructing a sparse-semantic tensor graph based on point cloud density drift, echo energy anomaly and a semantic graph, extracting context structure features through a time sequence graph neural network, and generating an affine complement containing affine transformation parameters; further realizing consistent alignment of boundaries by using a semantic drift compensation mechanism, reconstructing a false point cloud complementation frame and fusing the false point cloud complementation frame with the original point cloud, and outputting a semantic complementation enhancement frame; according to the method, the point cloud integrity and semantic precision in severe weather can be improved, the stability of obstacle detection and path planning is enhanced, and the method is suitable for a complex environment sensing scene in an intelligent driving system.

The invention relates to the field of meteorological prediction, and discloses a sea fog prediction method and device, and the method comprises the following steps: S1, obtaining multi-source heterogeneous data; s2, generating a scene representation and confidence coefficient weight matrix for representing the macroscopic physical background; s3, calculating compound loss; s4, updating parameters of the neural network model based on the composite loss. The device comprises a data acquisition module; a scene representation generation module; a confidence coefficient matrix generation module; a space-time prediction module; a composite loss calculation module; a parameter optimization module; a computer program for executing the scheme is stored in the storage medium. Through a dynamic adjustment mechanism of a macroscopic scene and a mesoscale process, the model can deeply understand how a long-term climate background affects the generation and elimination process of local sea fog, and the accuracy of sea fog prediction and the generalization ability of the model are improved.

The invention relates to the field of automobile electronic testing technology and equipment, and discloses a vehicle-mounted camera imaging quality testing method and device suitable for a rain and fog scene, and the method comprises the steps: installing a to-be-tested camera module in a rain and fog environment cabin, setting the spectrum and illumination of a top natural illumination simulation system, and simulating the illumination at different time periods; selecting a test target plate in the darkroom; starting an automatic centering module to align the optical axis with the standard plate reference; rain and fog parameters are set through the environment control host, the rain and fog supply system is started to simulate different levels of rain and fog environments, and the raindrop particle size, concentration, dripping speed, fog visibility and fogdrop particle size are dynamically regulated and controlled; shooting a target plate graph card, extracting characteristic parameters by an image analysis module, performing quantitative analysis, and generating an evaluation report, so as to solve the technical problem of singleness of rain and fog scene environment simulation in the prior art.

The invention discloses a thunder-vision fusion multi-sensor collaborative sensing system along a railway and an installation and calibration method, and belongs to the technical field of intelligent traffic and environment sensing. Aiming at the problems of low target detection precision, poor multi-source data fusion timeliness, dependence on manpower on sensor installation and calibration and the like in a complex scene along a railway, the system provides an innovative architecture of'layered sensing-dynamic fusion-autonomous calibration '. The multi-modal sensor array of a laser radar, a millimeter-wave radar, a visible light camera and a thermal infrared imager is deployed; autonomous optimization of sensor pose parameters is realized by using track geometric constraint and a deep learning model; an edge computing node and a cloud collaboration platform are integrated, and real-time target tracking, intrusion early warning and equipment state diagnosis are supported. Experiments show that the target detection accuracy rate of the system is greater than or equal to 98.5%, the false alarm rate is less than or equal to 0.3% and the self-calibration error of sensor installation parameters is less than 0.05 degree in the scenes of rain and fog, night, high-speed movement and the like, which are improved by more than 40% compared with the traditional scheme.

The invention provides an industrial visual imaging method and system suitable for a high-humidity water vapor environment and a medium, and the method comprises the steps: obtaining temperature fluctuation data and humidity fluctuation data in the water vapor environment at different time nodes based on a temperature and humidity sensor, and analyzing the environment parameters of a shooting environment; acquiring a shot image, analyzing brightness values of pixel points of the shot image, and performing equalization processing on the brightness values to obtain a brightness average value; analyzing fog concentration information of the shooting environment based on the environment parameters of the shooting environment, and analyzing interference information on the brightness mean value based on the fog concentration information; switching a shooting mode based on the interference information, and carrying out defogging processing on the shot image; carrying out enhancement processing, carrying out number domain stretching on the enhanced image, and then executing histogram equalization processing to obtain a visual imaging image; the shooting mode is adjusted by analyzing the interference information of the fog concentration on the image brightness, the image is defogged, the image is enhanced to recover image details, and the recognition capability is improved.

The invention discloses a multi-target tracking method for an offshore rain, fog and jitter environment, and the method comprises the steps: obtaining multi-target image data in the offshore rain, fog and jitter environment, and carrying out the target cutting and fusion of the multi-target image data based on a target enhancement segmentation strategy, and obtaining an enhanced sample image; mapping the enhanced sample image and target features in the multi-target image data into a candidate frame set of a pixel scale, and obtaining a sample data set containing adaptive Anchors according to the candidate frame set based on a clustering algorithm; performing model training on the multi-target detection network through the sample data set to obtain an optimal detection model so as to realize target detection; and defining a target state vector and a target observation vector according to a detection result, and realizing multi-target tracking under the marine rain and fog and jitter environment based on an improved Kalman filtering algorithm. The problems that in the prior art, the precision of multi-target detection under the marine rain and fog and jittering environment is insufficient, and a systematic solution for multi-target tracking under the marine rain and fog environment and the jittering scene is lacked are solved.

The invention discloses a marine frequency converter intelligent cooling system based on integrated modularization, relates to the technical field of ship power electronic equipment heat dissipation, and discloses the marine frequency converter intelligent cooling system based on integrated modularization. Comprising five core modules, wherein a data acquisition module is responsible for acquiring and preprocessing temperature, humidity and salt mist concentration data; the route analysis and model establishment module establishes and optimizes a model for predicting future environmental parameters by analyzing a route and applying a random forest algorithm; the threshold value calculation and optimization module calculates a dynamic threshold value in real time and continuously adjusts according to the ship speed and environmental parameters to continuously optimize threshold value parameters; the mode processing and conversion module identifies an extreme environment combination, triggers corresponding processing measures, and smoothly adjusts processing parameters and modes; and the route adjustment and safety protection module sets an intermediate value and an adjustment curve, continuously monitors the state, and automatically gives an alarm and takes measures when a fault or an abnormality occurs.

The invention relates to the technical field of information processing, in particular to an anti-interference fusion sensing system for a multi-source heterogeneous sensor, and aims to solve the problem that the sensing performance of the multi-source heterogeneous sensor is reduced in severe weather such as rain and fog. The system comprises a data acquisition and synchronization module, a multi-modal feature extraction and weather interference suppression module, an adaptive multi-modal fusion module, a target detection and tracking module and a sensing result output module. According to the method, robust features are extracted from interfered laser radar and camera data through resistance robust feature learning, and millimeter wave radar information is combined; dynamically adjusting the weight of each modal according to weather through a self-adaptive fusion module; the sensing precision and reliability of the vehicle in complex environments such as rain and fog are effectively improved, and driving safety is guaranteed.

The invention discloses a power transmission line mountain fire smoke detection method and system, and the method comprises the steps: segmenting a mountain fire smoke candidate region in a to-be-detected power transmission channel picture through a pre-trained semantic segmentation model, and calculating a semantic segmentation confidence score; for each mountain fire smoke candidate area, calculating a shape similarity score through a shape matching method, calculating a position confidence score through a position function, and calculating a color confidence score through color distribution comparison; and carrying out weighted summation on the semantic segmentation confidence score, the shape similarity score, the position confidence score and the color confidence score to obtain a final smoke confidence score, and comparing the final smoke confidence score with a preset threshold to judge whether the corresponding mountain fire smoke candidate region is mountain fire smoke. According to the technical scheme provided by the invention, the forest fire smoke condition in the power transmission line channel picture can be effectively identified, the misidentification conditions of pond smoke, chimney smoke, cloud and mist and the like are obviously reduced, and the method has a wide application scene.

The invention discloses a traffic accident video evidence analysis system and method based on multi-modal deep learning, and relates to the technical field of deep learning, and the method comprises the steps: carrying out the target recognition and tracking through radar point cloud, extracting a target number component and a motion variance component, carrying out the calculation through normalization and linear combination, and obtaining a dynamic fusion weight of each modal; and quantization and adaptive adjustment of scene complexity are realized. Then extracting each modal feature vector in a preset time sequence window, and performing weighted fusion based on the weight to form a unified feature vector; and inputting the uniform feature vector into a pre-trained accident classification model, and outputting whether an accident occurs or not and a type result. The method has the advantages that through complementary enhancement of multi-modal data, the accident identification accuracy in complex scenes such as low illumination and rain and fog is improved; adaptive adaptation to different scenes is realized through dynamic weight distribution; the whole scheme forms a closed loop, and has high robustness and practical value.

The invention discloses a dynamic weighing control method in a long-strip-shaped bare seedling transplanting process, and relates to the technical field of agricultural intelligent equipment, and the method comprises the steps: S1, obtaining environment humidity data and fog concentration data, judging whether the data exceed a normal range or not through a preset threshold value, obtaining a sensor surface condensation risk assessment result in a high-humidity fog state, and determining whether the data exceed a normal range or not; s2, according to the condensation risk assessment result, activating a sensor surface temperature control mechanism by adopting a heating module, and determining a temperature rise value to prevent formation of water drops and maintain a dry state of the sensor surface; according to the dynamic weighing control method in the long-strip-shaped bare seedling transplanting process, closed-loop control is achieved. The precision and the survival rate of seedling transplanting are remarkably improved, and the agricultural production efficiency is improved.

The invention provides a power transmission line bird damage risk assessment and early warning method and system, and relates to the technical field of intelligent power grid environment perception and risk early warning, and the method comprises the steps: data collection and multi-modal fusion: collecting image data, voiceprint data, meteorological parameters and ecological environment information along a power transmission line to form multi-modal data, the multi-modal data is uniformly coded by synchronizing timestamps and geographic coordinates, an original bird activity sample set is generated, and for the problems that the monitoring means is single and the recognition precision is insufficient, image, voiceprint, weather and ecological information are fused through a multi-modal data acquisition module, so that the recognition precision is improved. The acousto-optic image fusion algorithm and the two-channel convolutional neural network are adopted, all-weather bird type and behavior accurate recognition is achieved, the recognition accuracy is improved, key behaviors such as nesting and dung pulling can be stably captured even at night or in a foggy environment, and potential risk missing detection is remarkably reduced.

The invention belongs to the field of land surface temperature reconstruction, and relates to a high-spatial-temporal-resolution land surface temperature reconstruction method used in a cloud environment, which comprises the following steps: calculating a first reference land surface temperature and a first land surface temperature residual error of a clear sky pixel under a low spatial resolution according to a first land surface temperature annual change model; seamless surface reflectance data under high spatial resolution are obtained; defining a relative solar radiation index based on the digital elevation model data with high spatial resolution; screening clear sky pixels participating in modeling of the XGBoost model; an XGBoost model is adopted, and a plurality of model relations between a low-spatial-resolution land surface temperature annual change model coefficient and a land surface attribute and between a day-by-day land surface temperature residual error and the land surface attribute are constructed; obtaining a high-spatial-resolution surface temperature annual change model coefficient and a day-by-day surface temperature residual error based on the constructed multiple model relationships, and obtaining a high-spatial-resolution daily surface temperature; the accuracy and the stability of a surface temperature reconstruction result are improved.

The invention relates to the technical field of unmanned aerial vehicle control, in particular to an unmanned aerial vehicle adaptive shooting method and system based on multiple sensors and AI, and the method comprises the steps: multi-sensor data collection, deep learning fusion processing, multi-modal exception processing, decision-making layer path planning reinforcement, adaptive shooting precision improvement, and a re-planning mechanism. Compared with the defects that environmental perception is incomplete and the shooting quality is suddenly reduced under complex illumination or severe weather due to dependence on a single sensor in the prior art, according to the scheme of the invention, a multi-mode sensor array (laser radar / millimeter wave radar / illumination sensor and the like) and Transform-GNN deep learning fusion architecture is adopted; joint modeling of obstacle positions, dynamic tracks and illumination distribution is realized; therefore, the dimensionality and precision of environmental perception are remarkably improved, and the unmanned aerial vehicle can still output clear and stable high-quality pictures under extreme conditions of strong light, rain and fog and the like.

The invention discloses a heavy fog visibility prediction system and method based on deep learning, and relates to the technical field of data processing, and the method comprises the steps: determining a particle size modal category representing a heavy fog particle size distribution type; screening a historical particle size modal sequence matched with the particle size modal category of the current region based on an internal corresponding relation between the meteorological environment state and the particle size modal category; based on the category transfer complexity reflected by the transfer path, performing sequence reconstruction on typical particle size modal categories in the historical particle size modal sequence; inputting the reconstructed historical particle size modal sequence and the current particle size modal distribution characteristics into a long short-term memory network model optimized by a particle size modal attention mechanism, and outputting a predicted heavy fog visibility grade of the current region at a future moment; according to the method, the fog particle size modal physical characteristics and the deep learning model are fused, so that the fog visibility grade prediction accuracy and reliability are remarkably improved.

The invention provides a highway heavy fog weather dynamic risk assessment method, and relates to the technical field of traffic safety, and the method comprises the steps: collecting highway historical traffic data, and extracting and preprocessing a historical risk factor data set to construct a heavy fog weather risk analysis database; analyzing the incidence relation between the risk factor and the accident, extracting the space-time and dynamic characteristics of the risk factor, identifying a combined action mode, and building a four-layer risk fusion assessment framework after obtaining a risk evolution rule; and inputting current risk factor data, carrying out framework processing, and outputting a quantitative dynamic risk assessment value with a confidence interval. According to the method, the staged adjustment of risk modeling is realized by extracting the multi-dimensional dynamic features, the weight is dynamically corrected, the factor interaction is quantified, the evaluation value is iteratively optimized, and the confidence interval is output, so that the limitation that the traditional static modeling cannot adapt to the dynamic change of the heavy fog risk is broken through, and the real-time performance, the accuracy and the scene adaptability of risk evaluation are greatly improved.

The invention relates to the technical field of intelligent walking assisting equipment, and discloses a blind person intelligent guiding walking assisting method and system based on deep learning. The intelligent guiding power-assisted walking method for the blind is applied to guiding power-assisted electronic equipment, and specifically comprises the following steps: S101, receiving a starting signal sent by a user terminal, starting a multi-source sensor array to collect environment data in parallel, executing a Kalman-Transform space-time alignment algorithm, establishing a multi-modal data space-time unified coordinate system, and establishing a multi-modal data space-time unified coordinate system; when gait phase changes are collected and detected, a multi-source sensor array is activated and dynamically triggered for high-precision scanning, and a threat evaluation model is generated to calculate an obstacle collision probability point cloud matrix; and S102, processing the obstacle collision probability point cloud matrix through Point VoxelNet to generate a three-dimensional environment skeleton. The cross-modal attention mechanism of the millimeter wave radar and the vision improves the rain and fog weather recognition accuracy, and in addition, the dynamic threat evaluation module is combined with the double-threshold early warning strategy, so that the collision false alarm rate is effectively reduced.

The invention discloses a foggy day self-adaptive pedestrian trajectory prediction method for an old-age care robot. The method comprises the following steps: S1, extracting foggy day physical parameters and fog invariant features; s2, based on a multi-scale Mama space-time sequence module, extracting fusion features in the foggy weather; s3, constructing a fog sensing dynamic heterogeneous network; and S4, trajectory modeling is carried out by using the fusion features. The invention also discloses a foggy day self-adaptive pedestrian trajectory prediction system for the old-age care robot, which is applied to the foggy day self-adaptive pedestrian trajectory prediction method for the old-age care robot. Comprising a physical prior fusion fog sign decoupling unit used for estimating foggy day physical parameters and extracting fog invariant features, and a multi-scale Mamba space-time sequence module used for extracting space-time fusion features in a foggy day environment. The fog perception dynamic heterogeneous network is used for modeling the social relation of pedestrians in foggy days and fusing the spatio-temporal features; and the trajectory generation module is used for generating a prediction trajectory based on the fused features.