720 results about "Multi resolution" patented technology

A system and method are disclosed for generating hyperspectral images from multi-modal sensor data including RGB, LiDAR, thermal, and near-infrared inputs. Training data includes hyperspectral images and corresponding multi-modal measurements. Spectral band grouping is performed based on correlation coefficients. A multi-modal decomposition network with cross-modal attention mechanisms generate reconstructed hyperspectral images by fusing complementary sensor information. A fine-tuning network creates reconstructed RGB images. A comprehensive quality assurance system analyzes spectral consistency, cross-modal coherence, and fusion artifacts to generate quality metrics. Missing data compensation strategies handle corrupted sensor inputs using information from other modalities. The system includes temporal integration for video sequences and multi-resolution processing for different sensor resolutions. Quality metrics guide network weight adjustments to improve reconstruction accuracy while maintaining robustness to sensor failures and environmental variations.

The invention relates to the technical field of remote sensing image recognition, in particular to a remote sensing sewage area recognition method and system based on a graph structure and multi-stage enhancement. The method comprises the steps of performing data preprocessing and representation enhancement on an acquired remote sensing image; performing sewage salient region preliminary screening on the enhanced remote sensing image, including abnormal enhancement mapping construction based on local statistical distribution; pollution candidate graph extraction based on spatial structure prior driving; enhancing the response of the stable region based on a structure consistency enhancing mechanism of the polluted region; high-precision segmentation and identification of the sewage area comprises the following steps: constructing a multi-resolution residual pyramid structure; carrying out fine-grained boundary structure modeling and uncertainty suppression; generating a sewage distribution probability graph and optimizing structural consistency; according to the method, the multi-resolution residual pyramid structure is constructed, image context information under different perception scales is fully mined, and the sensitivity and edge integrity of the model to a sewage area under a complex texture background are remarkably enhanced.

The invention proposes a land utilization monitoring method and system based on remote sensing and big data, and relates to the technical field of land monitoring, and the method comprises the steps: dividing sub-regions, and obtaining the multi-temporal and multi-resolution remote sensing data of the sub-regions; performing feature extraction and classification on the remote sensing data based on a deep learning model to generate a land utilization classification map; based on the dual-temporal difference attention network, identifying a change area and constructing a change driving factor library fusing meteorological data and human activity data; detecting an abnormal area based on the driving factor library, and generating an abnormal type label and an attribution analysis report in combination with a dynamic early warning threshold; performing visual rendering on the monitoring result, and outputting an abnormal region early warning map and a disposal suggestion; high-precision feature extraction and classification are realized, change areas and driving factors are deeply analyzed, abnormal areas are effectively detected and early warning is performed, and the accuracy, timeliness and practicability of land utilization monitoring are improved.

A system and method for neural network weight compression with intrusion detection capabilities that optimizes model storage and transmission while providing security. The system analyzes weight characteristics to identify statistical properties within different neural network layers, generates optimized encoding schemes based on the analysis, and creates reference distributions for security verification. The compression process employs a multi-resolution approach that produces a progressive representation with base and enhancement layers, enabling flexible deployment across diverse computing environments. Security markers and statistical fingerprints can be embedded throughout the encoded representation, allowing for detection of unauthorized modifications during transmission or deployment. The system monitors encoded weight streams, measures distribution divergence against reference baselines, and generates alerts when statistical anomalies indicate potential tampering. This approach achieves superior compression ratios while maintaining model performance and providing robust protection against increasingly sophisticated attacks targeting neural network weights.

A high-fidelity three-dimensional reconstruction method for a specular reflection plane comprises the steps of decomposing pixel colors into diffuse reflection and specular reflection components based on a 3D Gaussian sphere, introducing a dynamic reflection ratio parameter and a spherical harmonic function coefficient to respectively represent two reflection characteristics, and simulating light multi-reflection behaviors through weight fusion of cumulative projection and a reflection ratio map. Secondly, in combination with monocular inverse depth calibration, depth smoothing constraint of color gradient weighting and edge mutual exclusion loss, geometric consistency is enhanced, and artifacts are suppressed; a progressive multi-resolution training strategy is further adopted, low resolution is gradually optimized to complete resolution, reflection parameters are activated in stages, 3D Gaussian sphere overgrowth and floating artifacts are inhibited, and efficiency and precision are balanced. According to the method, the reconstruction fidelity of the specular reflection scene is remarkably improved while the real-time rendering advantage of the 3DGS is reserved, and the method is suitable for the high-precision modeling fields of virtual reality, augmented reality and the like.

The invention provides a mutual inductor test data cloud edge cooperative processing method and device, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time domain and frequency domain combined feature extraction of the original measurement data of a mutual inductor test through a multi-mode decoupling preprocessing model, and forming a data feature vector set, further generating a confidence label flow through state recognition and Bayesian inference, evaluating a prediction error, and realizing data classification screening and priority queue construction; the data are uploaded to a cloud end in a semantic compression and multi-resolution representation vector mode, so that the transmission efficiency is improved; carrying out deep modeling and model performance monitoring at the cloud, and if degradation is detected, returning edge original data to update the model; and finally, generating a scheduling weight factor according to the prediction error distribution graph, and dynamically optimizing an edge cloud task allocation proportion. According to the invention, the problem of low resource allocation efficiency caused by lack of a dynamic scheduling mechanism based on prediction errors and confidence driving in existing mutual inductor test data edge cloud cooperative processing can be solved.

The invention relates to the technical field of equipment anomaly detection, and discloses an anomaly detection method and system for an intelligent motor, and the method comprises the steps: collecting multi-source heterogeneous data when the intelligent motor works, carrying out the feature-level fusion, and carrying out the potential local learning, and obtaining optimized data features; performing data fragment segmentation on the optimized data features to obtain segmentation features, and determining a time-frequency feature spectrum of the segmentation features under multiple resolutions; mining a feature association relationship of the multi-dimensional operation features of the intelligent motor to construct a reinforcement learning strategy, and performing target feature screening on the multi-dimensional operation features of the intelligent motor to obtain a screened feature subset; and carrying out confrontation generation processing on the screening feature subset to obtain a simulation sample, carrying out feature ensemble learning processing on the simulation sample and the screening feature subset to obtain a judgment feature, and carrying out anomaly analysis on the intelligent motor to obtain an anomaly detection report. According to the invention, the detection precision of non-obvious or weak-feature anomalies in the intelligent motor can be improved.

The invention discloses a micro-grid dynamic scheduling method based on deep learning, and the method comprises the steps: fusing industrial Internet of Things collection and GIS positioning, and constructing a multivariable original spatio-temporal data set covering multiple nodes; extracting multi-scale features through multi-resolution wavelets and Fourier transform, combining the multi-scale features with a dynamic adjacency matrix, and realizing feature adaptive distribution and nonlinear dynamic modeling by using multi-scale attention gating, graph convolution and a time sequence neural network model; the micro-grid load and state prediction accuracy, the system generalization ability and the abnormal response level can be effectively improved, and powerful support is provided for intelligent scheduling and abnormal analysis.

The invention relates to the technical field of three-dimensional image reconstruction, in particular to a VR-based three-dimensional reconstruction method and system for textile culture heritage, and the method comprises the steps: obtaining a multi-view image collection sequence of the textile culture heritage, carrying out the geometric topological feature extraction of the multi-view image collection sequence, generating a three-dimensional point cloud geometric model, and carrying out the reconstruction of the three-dimensional point cloud geometric model. Performing multi-resolution texture feature mapping processing on the three-dimensional point cloud geometric model to generate an initial three-dimensional model fusing geometric topological features and multi-resolution texture features; performing dynamic geometric optimization processing on the initial three-dimensional model based on preset VR display parameters to generate a target three-dimensional model after surface continuity correction; and performing dynamic matching processing on the target three-dimensional model and a preset VR interaction algorithm to generate a VR visual model with a multi-dimensional interaction attribute, and transmitting the VR visual model to a VR display terminal to perform three-dimensional space rendering, so that the three-dimensional reconstruction and interaction intelligence degree of the textile culture heritage can be improved in combination with a VR technology.

The invention discloses a ship desulfurization system reliability evaluation method and system based on multi-source information fusion, and the method comprises the steps: collecting multi-source information data of a hybrid desulfurization system, and carrying out the self-adaptive preprocessing; generating a fusion feature vector; constructing a dynamic Bayesian network based on multi-source fusion features, performing real-time reasoning by adopting data-driven transition probability learning and particle filtering, describing transient behaviors of system state evolution and mode switching, and performing dynamic multi-state reliability modeling; a fault mode is automatically extracted, and data-driven systematic risks are identified and quantitatively analyzed; a multi-resolution digital twinborn architecture is constructed, dynamic simulation prediction is carried out, a self-adaptive updating mechanism is adopted to keep the model synchronous with a physical system, and a virtual verification environment for reliability evaluation is provided; according to the method, an intelligent decision optimization system is constructed, self-adaptive generation and dynamic adjustment of a maintenance strategy are realized, closed-loop feedback is carried out, and the accuracy of reliability evaluation of the hybrid desulfurization system is improved.

The invention discloses a bridge structure surface crack multi-scale image segmentation and damage assessment method, and belongs to the technical field of bridge structure health monitoring, and the method comprises the steps: a multi-scale pyramid feature preprocessing step: carrying out the multi-resolution feature extraction of a bridge surface image; in the adaptive attention-guided crack segmentation step, crack region response is enhanced through a channel and space attention mechanism; the crack geometric parameter accurate quantification step is used for calculating the length, width, depth and direction of the crack; in the time sequence comparison crack development trend prediction step, the crack propagation rate is calculated according to the parameter difference value between the current detection data and the historical detection data divided by the time interval, and the development trend is predicted; in the multi-dimensional damage comprehensive evaluation step, damage scores are calculated, damage grades are determined, segmentation parameters are fed back and adjusted, and scientific data support is provided for bridge safety evaluation and maintenance decision making.

The invention relates to the technical field of image analysis, in particular to an automatic marking robot calibration method based on visual guidance, which comprises the following steps of: establishing an image set comprising different resolution levels based on operation site image data acquired by a marking robot, and performing corner detection and straight line segment detection on each level image in parallel. According to the method, through a parallel detection mechanism of a multi-resolution hierarchical image set, angular point and straight line segment features under different scales are synchronously extracted, and a multi-scale feature point set with high robustness is constructed in combination with cross-hierarchical coordinate stability measurement and response intensity quantification. And performing dynamic screening and grouping association on the feature points based on a preset geometric constraint condition, eliminating noise interference and false detection features, and generating a candidate calibration structure set with spatial consistency. Weighted contribution value fitting is adopted, cross-scale stability and detection confidence of feature points are integrated, and geometric accuracy and anti-interference capability of calibration reference point coordinates are improved.

The invention relates to the technical field of remote sensing image segmentation, and discloses a remote sensing image segmentation method based on a foreground sensing network, and the method comprises the steps: constructing the foreground sensing network based on an encoder and decoder structure, and constructing a combined loss function; training the foreground sensing network; segmenting by using the trained foreground sensing network to obtain a category probability graph of each pixel; and based on the dynamic multi-resolution attention, performing multi-scale grading on the category probability graph in combination with the environmental characteristics, dynamically adapting a post-processing strategy, and performing post-processing according to the adapted post-processing strategy to optimize a segmentation result. According to the method, an end-to-end intelligent segmentation framework is constructed through multi-scale feature fusion, environment adaptive post-processing and geographical semantic constraint, the segmentation precision, robustness and practicability in a complex remote sensing scene are remarkably improved, and an efficient solution is provided for the fields of natural resource management, disaster emergency response and the like.

The invention relates to the technical field of three-dimensional reconstruction, and discloses a monocular depth guided object-level NeRF reconstruction method, which comprises the following steps: firstly, through monocular video sequence input, generating a frame-by-frame initial depth map by using a depth estimation module, and estimating a relative camera attitude between adjacent frames through a relative attitude estimation module; calculating the absolute attitude of the camera in combination with the initial depth map and the relative camera attitude; then constructing a geometrically enhanced NeRF model, and optimizing scene representation through multi-resolution hash position coding and spherical harmonic direction coding; introducing photometric loss, depth contrast loss and density loss to jointly optimize parameters of the NeRF model, and constraining geometric reconstruction of the object by using depth information; and finally, through a four-stage iterative training strategy, alternately optimizing depth estimation, a camera attitude and a NeRF model, and generating an object-level controllable three-dimensional model. According to the method, depth information is fully utilized to optimize the NeRF training process, so that end-to-end reconstruction from a monocular video to an object-level controllable model is completed.

The invention provides a lodging area determination method and system based on a lodging monitoring spectral index image, and relates to the field of crop form prediction.The method comprises the steps that firstly, a target farmland remote sensing image and a digital surface model are obtained, radiation and geometric correction are completed in combination with GNSS positioning data and meteorological data, and a standardized orthoimage is generated; calculating a vegetation index on the image and carrying out difference to obtain change information; executing opening and closing operation by adopting a structural element adaptive to the image resolution and the crop row spacing, filtering noise according to the area of a connected domain or a pixel number threshold value, and extracting a spectral index abnormal region and a boundary thereof; and then fusing the boundary and the digital surface model under a unified coordinate reference, accumulating the area according to the pixel resolution, and outputting the area of the abnormal region. The method has parameter self-adaption and multi-source data fusion capabilities, can stably obtain a consistent abnormal region area in a multi-resolution and complex field environment, and provides reliable technical support for agricultural condition monitoring and disaster assessment.

The invention discloses a submarine topography super-resolution reconstruction method based on window displacement multi-source fusion, and aims to solve the problem of precision attenuation caused by absence of multi-beam truth value evaluation and resolution improvement of an existing neural network submarine topography reconstruction method. The method comprises the following steps of: obtaining gravity anomaly, gravity vertical gradient, vertical line deviation component, submarine topography background and ship survey water depth multi-source data of a target sea area; constructing a residual U-Net attention neural network model, and performing training by taking multi-source data extracted by a 11 * 11 window and position codes as input; moving the trained model sampling window in a staggered manner according to a target resolution step length; extracting data points in the window during movement and injecting position codes; and outputting a high-resolution water depth predicted value based on the position code and the window data. The reconstruction precision is improved by fusing physical quantities, the multi-resolution robustness is guaranteed by a window displacement mechanism, and the medium-frequency feature reconstruction capability is enhanced by a residual U-Net attention model.

The invention relates to the technical field of digital twinning, particularly provides a digital twinning scene adaptive optimization method based on point cloud data, and solves the problems that smoothness and detail presentation cannot be balanced during network fluctuation, and a scene prediction optimization mechanism based on space-time semantic analysis is not established. The method comprises the following steps: data preprocessing and hierarchical construction: carrying out preprocessing of abnormal point elimination, missing value complementation and density adjustment on original point cloud data, dividing key and common regions according to scene requirements, and constructing a multi-scale and multi-resolution hierarchical point cloud data system; the method can deeply analyze the spatio-temporal dynamic characteristics of the point cloud data through the technologies of spatio-temporal dynamic semantic segmentation, dynamic change monitoring, prediction optimization and the like, processes the continuous time sequence point cloud data after adaptive transmission, models a time dependency relationship and extracts local and global characteristics in time and space dimensions respectively, and improves the accuracy of point cloud data processing. Accurate identification of object types, positions and dynamic change information is realized.

The invention relates to the technical field of mine safety and geological environment evaluation, in particular to a mine goaf geological environment dynamic evaluation method based on digital twinning. The method comprises the following steps: constructing a digital twinborn model of a mine goaf, designing a data structure and a virtual-real mapping rule, collecting multi-source heterogeneous data, and performing spatio-temporal data indexing processing to form a spatio-temporal associated data stream; microseismic travel time data are extracted from the space-time associated data stream; performing parallel ray tracing and rapid inversion on the microseismic travel time data to obtain a wave velocity distribution initial value; carrying out wave velocity field self-adaptive grid refinement on the micro-seismic travel time data by utilizing the wave velocity distribution initial value to obtain a multi-resolution wave velocity field; and carrying out wave velocity change rate analysis on the multi-resolution wave velocity field to obtain a wave velocity variability tensor. According to the method, the abstract energy transfer process is concrete into a data field capable of being quantitatively analyzed, so that the geological environment of the goaf is accurately and dynamically evaluated.

A monocular self-supervision depth estimation method fusing multi-resolution features and global context belongs to the technical field of computer vision, and comprises the following steps: firstly, providing an encoder based on a cross-stage feature high-resolution network, and enhancing the extraction capability of multi-scale features by improving a cross-stage connection mechanism of an HRNet network; secondly, a visual Transform module is introduced to improve complex down-sampling operation, and a multi-head self-attention mechanism is utilized to establish global pixel association; then, an improved decoder network is provided, a channel attention module is used for re-weighting coding features, multi-level features of an encoder are utilized more efficiently, and the balance between segmentation precision and calculation cost is achieved; and finally, proposing an improved attitude estimation network based on lightweight ResNet18, predicting 6-DoF relative poses of adjacent frames in combination with a four-layer convolutional decoder, and accurately estimating the depth of each object. According to the invention, the accuracy of depth estimation is improved, and the visual effect of the depth view is improved.

The invention relates to the technical field of artificial intelligence, and discloses a precise targeted AI recommendation system. The method comprises the following steps: firstly, acquiring multi-source heterogeneous data, constructing a global space-time database through space-time alignment, complementing and repairing the data by using a generative adversarial network, and cleaning and verifying; and constructing a three-dimensional topological model based on the improved graph neural network, and fusing multiple features to generate a multi-resolution model. A scene simulation platform is established by adopting a multi-objective reinforcement learning algorithm, multi-objective optimization is set, and a strategy library is constructed. And screening key factors by using a sparse Bayesian theory to construct a lightweight model. And designing a layered dynamic optimization framework, solving an optimal solution of each sub-layer by using a hybrid algorithm, realizing cross-layer constraint synchronous decoupling through a block chain smart contract, and outputting a global optimal three-dimensional design scheme. The method improves the application level of the digital achievement of the power grid design, gives consideration to stability, economy and environmental protection, and has important application value.

The invention provides a dark image preprocessing method and device based on machine vision, equipment and a medium, and relates to the technical field of agricultural automation, and the method comprises the steps: constructing a multi-scale Retinex enhanced fusion adaptive gamma correction illumination compensation model, carrying out the processing of an original garlic seed screening image through the illumination compensation model, eliminating illumination unevenness and light reflection interference, and obtaining an original garlic seed screening image; obtaining a first garlic seed screening image; according to a texture complexity detection method based on a gray-scale guide map, designing an adaptive joint bilateral filter, and processing the first garlic seed screening image through the adaptive joint bilateral filter to obtain a second garlic seed screening image; and performing multi-resolution contrast enhancement processing on the second garlic seed screening image through a pyramid decomposition and entropy-driven partitioning strategy to obtain a preprocessed garlic seed screening image. According to the method, the defect detection accuracy and the classification efficiency can be effectively improved under the conditions of high noise and complex backgrounds, and robust visual preprocessing support is provided for subsequent automatic sorting.

The invention provides a geotechnical engineering slope deformation monitoring method and system, belongs to the field of geotechnical engineering monitoring, and is used for solving the problems of low monitoring precision and poor early warning reliability caused by spatial-temporal dislocation of multi-source data, static geological parameters, poor swivel scene adaptation and no closed-loop optimization in related technologies. According to the method, multi-source deformation data, geological parameters and rotation characteristics are collected, the type and progress of a rotation are identified, multi-resolution hierarchical fusion (dynamic weight adaptive data credibility and environmental interference) and dynamic inversion geological parameters are performed after space-time alignment to construct a prediction model, model parameters and an early warning threshold are dynamically scheduled in combination with the rotation characteristics, and the prediction result is obtained. The precision is improved through closed-loop feedback optimization iteration, the system provides hardware support for the method, and precise monitoring and efficient early warning of slope deformation are achieved.

The invention relates to an artificial precipitation enhancement operation area identification method and system based on multi-source data, and relates to the technical field of meteorological intervention operation, and the method comprises the steps: building a multi-resolution dynamic data grid system, and enabling a coarse-grained grid layer to cover the whole operation area, and to be used for bearing the macro data of a satellite cloud picture and a conventional radar; the fineness grid layer is used for receiving and mapping real-time data provided by the high-resolution radar; receiving real-time data of all cloud blocks, mapping the real-time data to corresponding grid layers, and projecting the data collected at different time points to a unified time reference frame to obtain a cloud system state view; predictive interpolation is carried out by using the latest data of the conventional radar and the macroscopic trend of the satellite cloud picture in combination with the motion vector and confidence weighting so as to obtain cloud system information; and forming a visual cloud system three-dimensional view, identifying and tracking a target cloud block with an artificial precipitation enhancement potential, and generating an operation instruction. The method has the effect that the target cloud block with the operation value can be identified and tracked to generate the accurate operation instruction.

The invention discloses a dynamic graph convolution-based target point cloud completion method for a battery replacement robot, and the method comprises the steps: 1, reconstructing a complete target fastener model from a multi-view image through SFM and MVS technologies, and obtaining complete point cloud data; 2, constructing an incomplete-complete point cloud pair as training data by using a geometric constraint-based adaptive cutting strategy; 3, multi-resolution point cloud processing is adopted, and feature extraction and fusion are carried out according to three-level resolution; 4, on the basis of DGCNN dynamic graph convolution, in combination with multi-stage Edge Conv dynamic edge convolution and a Transform coding module, local geometric feature capture and global relation perception are realized; 5, point cloud generation adopts a pyramid step-by-step refining method, geometric details are added to each layer based on a previous layer result, and the number of points is gradually expanded; and 6, the training process is guided through a multi-target joint loss function, and the point cloud quality is improved while the precision is ensured. The geometric structure integrity of the point cloud of the target fastener is improved, and the operation error of follow-up operation of the battery replacement robot is reduced.

The invention discloses a semantic-based text content index automatic identification method, which relates to the technical field of information retrieval, and comprises the following steps: initializing sparse projection and LSH signature, performing iterative optimization by using a Lagrange duality form and gradient update, adjusting hash digits, obtaining a fragment index through a k-d tree, and constructing an inverted index. According to the method, compression is performed through Delta coding, an index map is constructed based on Jaccard similarity, compression is performed through WebGraph, CSNMF is used in combination with Z-Laplacian regularization, a low-rank basis matrix and a low-rank coding matrix are generated, a compressed inverted index is reconstructed after iterative optimization, and reconstructed inverted index entries are generated. According to the method, through multi-resolution hash table initialization, joint feature optimization, local adaptive quantization and low-rank index reconstruction, the semantic expression ability and the compression effect of an index structure are improved, the index precision and efficiency are improved, and intelligent identification of index content is achieved.

The invention discloses a method for processing three-dimensional scanning data. The method comprises the following steps: acquiring the three-dimensional scanning data of a target object; performing fusion processing on the three-dimensional scanning data under the first resolution to obtain first point cloud data; in response to a second resolution, performing fusion processing on the three-dimensional scanning data to obtain second point cloud data, the value of the first resolution being greater than the value of the second resolution; and performing duplicate removal and fusion on the first point cloud data and the second point cloud data to obtain multi-resolution point cloud data of the target object, and finally displaying the multi-resolution point cloud data. According to the invention, the range of the point cloud displayed in real time in the scanning process is consistent with the range of the finally displayed point cloud.

Systems and methods are disclosed for anomaly detection using a “detect and collect” cybersecurity monitoring approach. Initially, a cybersecurity monitoring system obtains and analyzes a baseline subset of telemetry data from computing resources to detect potential anomalies indicative of cybersecurity threats. Responsive to identifying such anomalies, the system selectively determines additional, contextually relevant telemetry data for targeted collection. This selective data collection significantly reduces telemetry volumes, enhancing efficiency and scalability. An intelligent data fabric and dynamic security knowledge graph are employed to enrich telemetry data in real-time, enabling comprehensive anomaly characterization, risk scoring, and automated security responses. The disclosed techniques support multimodal and multiresolution anomaly detection, adaptive learning, and rapid threat response within diverse distributed computing environments.

The invention discloses an open vocabulary three-dimensional scene semantic query method based on a hierarchical semantic field, and the method comprises the steps: extracting instance masks of a sub-part, a part and an overall hierarchy through a zero sample segmentation model SAM, and generating a pixel-level multi-granularity semantic feature map in combination with a CLIP encoder; a neural radiation field architecture is improved, a multi-resolution hash grid and a multi-head multi-layer perceptron network are designed, efficient mapping from three-dimensional space coordinates to multi-level semantic features is realized, and region contrast loss and feature consistency loss are introduced to constrain cross-view semantic consistency; in the reasoning stage, the optimal hierarchy is adaptively selected by calculating the similarity between the text features and the rendering semantic features, and a high-precision semantic segmentation result is generated. According to the method, the problem of feature blurring caused by the fact that a traditional method depends on image block cutting is solved, the problems of low semantic query precision, insufficient multi-granularity understanding and cross-view conflicts in an open vocabulary scene are solved, and a natural language driven solution with fine granularity and high robustness is provided for three-dimensional scene interaction.

The invention discloses an ultra-low illumination-oriented SPAD active imaging data compression method, which comprises the following steps of: performing wavelet decomposition on histogram data of a single pixel in a time-frequency domain based on wavelet transform to obtain low-frequency data of each pixel after wavelet decomposition; by taking each pixel as a center, performing non-maximum suppression and data enhancement on the low-frequency data after the wavelet decomposition of the current pixel by using adjacent pixels to obtain processed compressed data; customizing different Gaussian kernel parameters for each pixel according to the possibility that each pixel is located at the boundary, and performing Gaussian filtering on the processed compressed data of each pixel to obtain filtered data; and performing depth estimation on the filtered data to obtain a final depth image. According to the method, the depth reconstruction performance of the laser pulse can be improved by utilizing the multi-resolution characteristic of wavelet transform, the space-time correlation of signal photons and the smoothness of Gaussian filtering.

The invention relates to the technical field of monitoring control, and particularly discloses a mixed bean disease monitoring system and method. Comprising an unmanned aerial vehicle remote sensing acquisition module, an image preprocessing and feature extraction module, an expert teaching and path recording module, a track replay and servo execution module, a gradient integrated classification module, a dynamic abnormal sample mining and adaptive reinforcement learning module and a visualization and prevention and control scheduling module. According to the method, hyperspectral multi-resolution data acquisition, expert interactive path standardization script generation and automatic reproduction, dynamic abnormal sample preferential capture and multi-terminal expert remote collaborative review, feature screening and gradient integrated modeling and variable prescription map intelligent generation are carried out; and spatial positioning, grading identification and prevention and control task closed-loop scheduling of the red kidney bean CBB scab in each stage of morning, noon and evening are realized.