26 results about "Local statistics" patented technology

The invention relates to the technical field of image processing, and discloses a packaging material printing quality detection method and system based on machine vision. The method comprises the following steps: acquiring a multispectral image sequence and three-dimensional shape data of a moving packaging and printing material under different illumination, and constructing a dynamic three-dimensional physical attribute field; generating a virtual reference image and a dynamic reference image, constructing a multi-modal reference image, carrying out space-time registration on the multi-modal reference image and the dynamic three-dimensional physical attribute field, and calculating the difference between the multi-modal reference image and the dynamic three-dimensional physical attribute field in different dimensions to generate a multi-dimensional difference quality field; each dimension difference is enhanced through local statistics, and the comprehensive defect confidence coefficient is calculated based on the enhanced dimension difference; and extracting a defect region based on the comprehensive defect confidence, generating a defect evolution sequence and a defect track, analyzing defect track characteristics, constructing a correlation model in combination with process parameter time sequence data of the printing equipment, and positioning a defect reason. According to the invention, high-precision, multi-dimensional and self-adaptive printing defect detection and traceability can be realized.

The invention provides an underground structure boundary identification method and system based on distributed optical fiber sensing, and relates to the technical field of underground structure detection and boundary identification. The method comprises the following steps: firstly, arranging a sensing array in a to-be-detected area, establishing a channel corresponding to a space coordinate, and collecting a non-excitation base line; medium and noise are estimated through low-energy probe, excitation frequency band, energy and repetition rate are optimized, and the signals are emitted by an adaptive seismic source; reflection and transmission responses are synchronously collected under the unified time reference; performing wavelet denoising, temperature and dispersion compensation and time alignment on the data, and extracting amplitude, phase and frequency characteristics; gradient is calculated on the feature field, non-maximum suppression is carried out, and stable boundary points are obtained by adopting self-adaptive double thresholds based on local statistics and combining time continuity and space connectivity constraints; a two-dimensional section is obtained through spline fitting, a three-dimensional model is reconstructed under the constraint of multi-section consistency, a section map and the three-dimensional model are output, and high-precision, real-time and visual detection of small-size boundaries is achieved.

The invention relates to the technical field of pattern classification, in particular to a pattern classification method and system based on a regional label, and the method comprises the following steps: extracting a local label threshold based on a pixel gray range, analyzing the regional difference of a pattern image, screening a change optimal group as a reference, analyzing a gradient direction and amplitude mutation, and screening layered breakpoints. And integrating the regional features, and judging category attribution to obtain a discriminant quantity. According to the method, detailed pixel relation analysis and regional structure feature extraction are executed step by step, label affiliation accurate adjustment is achieved based on local statistics and spatial attribute association synchronization, boundary anomaly is judged through gradient and label change multiple parameters, layering is carried out, nodes are automatically positioned according to attribute mutation, and category affiliation is decided by multi-dimensional attribute interaction. Layered and partitioned analysis of high-detail and complex patterns is realized, hierarchical attribution requirements under multiple scenes are adapted, the region classification accuracy is enhanced, and the response capability to the internal differential and hierarchical relationship of the pattern structure is improved.

The invention provides a visual mapping method in a low-light environment, and relates to the technical field of visual mapping, and the method comprises the steps: separating a brightness channel from an input image, dividing the brightness channel into a plurality of tiles, calculating the local statistic of each tile to adaptively generate a contrast amplitude limiting parameter and an enhancement parameter vector, carrying out the CLAHE enhancement of the brightness channel based on the parameters, and obtaining a visual mapping result. After an image is reconstructed, candidate points in one-to-one correspondence with existing map feature points are extracted, comprehensive confidence scores of the candidate points are calculated, and whether the candidate points are adopted as new map points or not is determined by judging whether enhancement parameter vector distances between the candidate points and the corresponding map points are lower than a dynamic threshold value or not and combining the confidence scores. And finally, dynamically optimizing and enhancing parameters by the system according to the matching survival rate of each tile feature point, continuously improving the mapping quality and stability through iterative updating, and realizing robust visual mapping in a low-light environment.

The invention discloses a method for evaluating the health state of an energy storage battery pack based on a graph neural network. The method comprises the following steps: S1, collecting and preprocessing operation data; s2, establishing an undirected topological graph; s3, extracting local statistics from each pair of nodes in the undirected topological graph, and updating an edge weight; s4, training the undirected topological graph, and outputting an action value; s5, constructing a graph neural network model, extracting spatial features of each node, and obtaining state vector representation of each node; s6, calculating the health score of each single battery, performing weighted fusion on the health scores of all nodes, and generating an energy storage battery pack health state index; and S7, according to the health state index, generating a residual service life prediction result of the energy storage battery pack. According to the method, the dynamically optimized energy storage battery pack diagram structure is constructed, and the diagram neural network and reinforcement learning combined modeling is introduced, so that the high-precision evaluation of the health state of the energy storage battery pack and the accurate prediction of the residual life of the energy storage battery pack are realized.

The invention discloses a sewage treatment plant equipment fault diagnosis method based on multi-scale adaptive feature extraction, and belongs to the technical field of equipment intelligent operation and maintenance. According to the method, after adaptive filtering and quality evaluation are carried out on a vibration signal, an adaptive decomposition strategy fusing VMD and improved EMD is adopted, and a modal number and parameters are optimized according to an information entropy criterion; extracting multi-scale features from three levels of micro-scale (instantaneous characteristics), mesoscale (local statistics) and macro-scale (global energy / entropy), and screening key features through mutual information and a two-layer fusion mechanism; a health index model is constructed, and normal, attention, warning and danger four-level dynamic early warning is achieved; and meanwhile, a self-adaptive fault knowledge base capable of being incrementally updated is established in combination with density clustering and a Markov chain. According to the method, the fault identification accuracy and the system self-adaptive capacity are remarkably improved, and the method is suitable for intelligent diagnosis of sewage treatment key equipment such as a centrifugal pump and a Roots blower.

The invention provides a small-sample defect detection method and system based on a visual language large model, and relates to the technical field of computer vision, and the method comprises the steps: obtaining an original image of a to-be-detected workpiece, and extracting a first feature; based on the image and / or prior information, semantic prompts are generated by utilizing a visual language large model and are converted into second features capable of being aligned with the first features, and cross-modal fusion is carried out to obtain fusion features; in a candidate position neighborhood, subspace projection is performed on the fusion features, anisotropy measurement is determined based on local statistics, end member number estimation and intra-packet or extra-packet discrimination are performed under the measurement, and an active end member number is determined in combination with a cross-scale stability criterion; a decoding channel and a threshold value are adjusted according to the number of the active end members, and a result is fed back to the fusion process; decoding the adjusted fusion features, and outputting a pixel-level defect segmentation image; therefore, the positioning precision and segmentation stability of weak feature defects are improved under the condition that defect samples are limited.

The invention relates to an intelligent detection method and system for differential settlement of a super high-rise building foundation, and belongs to the technical field of settlement detection. The method comprises the following steps: collecting settlement monitoring data, and constructing a training data set; calculating local statistics according to the spatial topological relation of the monitoring points, and combining with the adjacent point set to standardize the original data to obtain a standardized settlement value; calculating a curvature mutation factor through second-order difference, superposing a local variance weight factor to amplify the contribution of the mutation region, and splicing the standardized settlement feature vector and the curvature mutation factor to obtain an enhanced feature vector; constructing a differential settlement detection model, and inputting the enhanced feature vector into the model to obtain a differential settlement probability and a local abnormal thermodynamic diagram vector; optimizing the training process of the model through the loss function to obtain a trained model; newly collected real-time settlement monitoring data are processed and then input into the trained model, and an intelligent detection result is obtained. According to the invention, the detection capability of the differential settlement of the foundation can be improved.

The application discloses a strip steel hot rolling process multi-performance index abnormality monitoring method and system, and belongs to the technical field of industrial process control and fault diagnosis, and the method comprises the following steps: collecting process variable data and performance index data in a historical strip steel hot rolling process and performing standardization pretreatment; dividing the strip steel hot rolling process into multiple different process subblocks; for each process subblock, performing space-time feature extraction on the corresponding process variable data to obtain a space-time feature representation; constructing a local statistic quantity of each performance index; fusing the local statistic quantities to obtain a global statistic quantity and a global control limit; and based on this, realizing multi-performance index abnormality monitoring and alarm. The application effectively solves the problems that cross-process time lag dependence is difficult to capture, space-time feature extraction is not comprehensive, and single performance index monitoring leads to abnormality missed reports in the strip steel hot rolling process, significantly improves the fault detection rate and reduces the false alarm rate.

The application provides a visual mapping method in a low-light environment, and relates to the technical field of visual mapping. In the application, a brightness channel is separated from an input image and is divided into a plurality of tiles, local statistics of each tile are calculated to adaptively generate a contrast limiting parameter and an enhancement parameter vector, the brightness channel is enhanced based on the parameter, candidate points corresponding to feature points of an existing map are extracted after reconstruction of the image, a comprehensive confidence score is calculated, whether the candidate points and the corresponding map points are below a dynamic threshold in terms of the distance of the enhancement parameter vector is judged, and the confidence score is combined to determine whether the candidate points are adopted as new map points. Finally, the system dynamically optimizes the enhancement parameter according to the matching survival rate of feature points of each tile, continuously improves the mapping quality and stability through iterative updating, and realizes robust visual mapping in a low-light environment.

The application discloses a kind of based on deep learning electric power load multi-period prediction method and system, it utilizes instance-level reversible distribution adaptation mechanism, and the original tensor is stationary mapping in input end dynamic extraction local statistics, convert time-varying distribution into unified feature space that model is easy to converge.On this basis, design time-frequency double-flow interactive network, and extract local mutation feature in time domain and global periodicity feature in frequency domain in parallel, and compensate the information missing of single perspective through cross-domain gate fusion mechanism.Finally, cooperate with future-oriented generative decoder and distribution recovery module, and the stationary prediction result is reversely mapped back to the original dimension space with physical meaning.Through this closed-loop logic of stationary reasoning first and distribution restoration later, effectively eliminate the negative influence of statistical drift on model generalization, realize high-precision multi-period prediction under the premise of retaining environment-driven physical characteristics.

A method for stitching forward-looking sonar images while retaining information is provided. In this application, forward-looking sonar is used as underwater detection equipment and acquired forward-looking sonar images are stitched together. A phase correlation method is employed for estimating displacements between images to determine the position of each single forward-looking sonar image within the stitched image. A method based on local statistics is used for image blending to obtain a stitched forward-looking sonar image that retains information. The method for stitching forward-looking sonar images proposed in this application adapts to intra-frame and inter-frame artifacts caused by non-ideal sonar imaging configurations, overcoming the drawbacks of image quality degradation due to the intra-frame and inter-frame artifacts. The method enhances the amount of information contained in the stitched image, which can assist observers in conducting rapid underwater exploration.

The invention discloses a road ecological landscape dynamic classification method and system based on landscape visual features, and relates to the field of artificial intelligence, and the method comprises the steps: constructing a data set containing image-level and pixel-level double-layer annotations; preprocessing the image through self-adaptive illumination-color normalization in combination with local statistics and global estimation; carrying out image block division and recombination by adopting semantic guidance; constructing an end-to-end classification model, deeply fusing visual and semantic features, modeling a spatial layout and semantic similarity relationship, and realizing landscape type discrimination; and utilizing the training model to automatically classify the new image. According to the method, the robustness to complex illumination can be effectively improved, structured image blocks rich in semantics are formed, multi-scale feature interaction and discrimination are enhanced, and feature representation consistent in semantics is learned.

This invention provides a method and system for identifying underground structure boundaries based on distributed optical fiber sensing, relating to the field of underground structure detection and boundary identification technology. The invention first deploys a sensor array in the area to be measured and establishes a correspondence between channels and spatial coordinates, acquiring an unexcited baseline; through low-energy probing to estimate the medium and noise, the excitation frequency band, energy, and repetition rate are optimized, emitted by an adaptive seismic source; reflection and transmission responses are synchronously acquired under a unified time reference; wavelet denoising, temperature and dispersion compensation, and time alignment are applied to the data to extract amplitude, phase, and frequency features; gradients are calculated on the feature field and non-maximum suppression is performed; a stable boundary point is obtained by employing an adaptive double threshold based on local statistics combined with temporal continuity and spatial connectivity constraints; a two-dimensional profile is obtained by spline fitting, and a three-dimensional model is reconstructed under multi-profile consistency constraints, outputting the profile diagram and the three-dimensional model, achieving high-precision, real-time, and visualized detection of small-sized boundaries.

The invention discloses a road section traffic accident risk spatio-temporal characteristic analysis method and system, and the method comprises the steps: selecting collision time as a traffic conflict index, and processing collision time data to obtain a conflict sample; determining a conflict threshold based on the conflict sample, and further obtaining an extreme conflict sample; on the basis of the extreme conflict samples, representing space-time distribution of extreme traffic conflicts of the microscopic road sections by adopting a space-time hot spot analysis method; the method specifically comprises the steps of performing data aggregation on extreme conflict samples, and creating a space-time cube; calculating local statistics in the space-time cube, and identifying extreme conflict space-time cold and hot spots based on the local statistics; and analyzing the local statistical magnitude by adopting a time trend analysis method to obtain a time change trend of the local statistical magnitude. The problem of subjectivity of traffic conflict index threshold selection and the problem of space-time dimension separation of road extreme traffic conflict space-time distribution characteristic analysis are solved, and the blank of research of microscopic road section extreme traffic conflict space-time dynamic mode evolution characteristics is filled.

The application relates to an intelligent detection method and system for uneven settlement of a super-high building foundation, and belongs to the technical field of settlement detection. The method comprises the following steps: collecting settlement monitoring data and constructing a training data set; calculating local statistics according to the spatial topological relationship of monitoring points, combining adjacent point set standardized original data to obtain standardized settlement values; amplifying the contribution of the mutation area by a second-order difference algorithm curvature mutation factor and a local variance weight factor, splicing the standardized settlement characteristic vector and the curvature mutation factor to obtain an enhanced characteristic vector; constructing an uneven settlement detection model, inputting the enhanced characteristic vector into the model to obtain an uneven settlement probability and a local anomaly heat map vector; optimizing the training process of the model through a loss function to obtain a trained model; and inputting the processed real-time settlement monitoring data collected newly into the trained model to obtain an intelligent detection result. The application can improve the detection capability of the uneven settlement of the foundation.

The application discloses a kind of narrow river detection methods suitable for spaceborne wide swath interferometric radar altimeter image, comprising: reading into spaceborne wide swath interferometric radar altimeter image;It is enhanced river linear feature by Gaussian difference preprocessing to suppress background;Curvature structure perception detector is constructed, curvature response map is calculated based on Hessian matrix eigenvalue, and feature fusion enhancement is carried out in conjunction with multi-direction structure consistency score;The binary image obtained by adaptive threshold segmentation based on local statistics is used as region label by carrying out adaptive threshold segmentation to enhanced feature map, maximum value is extracted in the region corresponding to enhanced feature map, seed point set is formed, and adaptive region growth based on queue priority is executed, to generate initial river center line;Markov random field model is introduced, global structure optimization and topological connectivity correction are carried out using graph cut algorithm, to obtain optimized river center line;Direction and radiation characteristic constraint is applied again, morphological dilation is carried out, and finally complete river mask is generated.

The invention relates to a splicing seam repairing method and device, electronic equipment and a readable medium, and the method comprises the steps: obtaining an original DR image collected by DR equipment, and determining a first candidate region corresponding to the position of a grid of the DR equipment on the original DR image; performing multi-algorithm joint calculation on the original DR image based on the first candidate region to determine a splicing seam region in the first candidate region; performing local statistics on the splicing seam region and the neighborhood thereof to obtain local gray feature information, and taking the local gray feature information as a dynamic threshold constraint; and under the constraint of a dynamic threshold value, a weighted diffusion equation is constructed based on the main structure direction and the diffusion weight determined by a structure tensor, iterative repair is carried out on the splicing seam area according to the weighted diffusion equation, and the structure tensor is obtained by calculating gradient information of the splicing seam area and a neighborhood of the splicing seam area. The problem of how to ensure the structure continuity and the detail integrity while repairing the splicing seam of the DR image is solved.

This invention provides a method for suppressing and enhancing infrared image defects and thermal noise, relating to the field of image enhancement. The method aims to construct thermal target protection weights and data fidelity weights by normalizing, performing local statistics, and estimating gradient and fringe principal directions on the original infrared image. The observed image is decomposed into structural components, point-like defective pixel components, structured thermal noise components, and low-frequency thermal bias components. Through a thermal consistency-guided defective pixel weight generation mechanism, direction-aware structured thermal noise separation, background-gated low-frequency thermal bias separation, anisotropic edge-preserving enhancement for thermal target protection, and multi-scale local thermal contrast preservation, a joint optimization solution is achieved, and the enhanced image is output after local thermal contrast remapping. This method effectively suppresses defective pixels, fringe noise, and background thermal bias, reduces the risk of accidental deletion of small thermal targets, maintains the target outline and the thermal difference between the target and the background, and improves the clarity, recognizability, and engineering applicability of infrared images.

The application discloses a kind of federal continuous learning methods and systems for long tail detection, it is related to machine learning technical field, method includes: cloud server initialization and issue global model;Each edge device statistics local class distribution, calculates logit and updates local model, extracts the multiple prototype vector of each class and uploads;Cloud server calculates each device update gradient, combines global prototype pool to evaluate each class aggregation credibility and calculates class-level aggregation weight, carries out class-level weighted aggregation to classifier, average aggregation to feature extractor, updates global model.The application is uploaded to cloud server by edge device local statistics class distribution and calibration Logit simultaneously, and class-level credibility evaluation and weighted aggregation are carried out to multiple prototype vector, cope with long tail data bias in federal continuous learning, improve the detection performance of class.

A collaborative learning framework is presented. The collaborative framework is implemented by multiple network nodes interconnected by a network. The network nodes belong to multiple client systems of the framework. A network node belonging to a first client system constructs a predictive model for the first client system by using a pattern dictionary that is a built based on a consensus among the multiple client systems. The network node calculates a set of local statistics for the first client system based on raw data of the first client system. The network node computes a consensus set of local statistics by aggregating sets of local statistics from the multiple client systems. The network node updates the pattern dictionary based on current values of the pattern dictionary and the consensus set of local statistics.

The invention relates to the technical field of vehicle-mounted intelligent perception, and discloses a multi-angle face image analysis method for driving state monitoring, which comprises the following steps: constructing a steady-state three-dimensional face template during a baseline acquisition period based on pre-calibration input of multiple paths of synchronous visible light and near-infrared cameras and depth and bottom layer sensors in a vehicle; during operation, performing geometric consistency judgment on each key point by adopting a distorted normalized image coordinate, a projection re-projection error and a projection depth, and comparing local statistics of visible and near-infrared channels in a template projection window so as to identify mirror surface highlight and channel distortion; and aggregating each single-source criterion by using deterministic binary logic to obtain a key point credible mark, and analyzing and outputting states of sleepiness, distraction and the like of the driver according to a rule set based on credible three-dimensional geometric quantities.

The invention discloses a narrow river detection method suitable for a satellite-borne wide-cradling interference radar altimeter image. The method comprises the following steps: reading in the satellite-borne wide-cradling interference radar altimeter image; gaussian difference preprocessing is carried out to suppress background and enhance the linear features of the river channel; constructing a curvature structure sensing detector, calculating a curvature response diagram based on a Hessian matrix eigenvalue, and performing feature fusion enhancement in combination with a multi-direction structure consistency score; performing self-adaptive threshold segmentation based on local statistics on the enhanced feature map to obtain a second map as a region mark, extracting a maximum value in a corresponding region of the enhanced feature map, forming a seed point set, executing self-adaptive region growth based on queue priority, and generating an initial river channel center line; introducing a Markov random field model, and performing global structure optimization and topological connectivity correction by using a graph cut algorithm to obtain an optimized river channel center line; and then applying direction and radiation characteristic constraints, performing morphological expansion, and generating a final complete riverway mask.

This application provides a method and system for few-sample defect detection based on a large visual language model, relating to the field of computer vision technology. The method includes: acquiring the original image of the workpiece to be inspected and extracting a first feature; generating semantic cues based on the image and / or prior information using a large visual language model, and converting them into second features that can be aligned with the first feature, and performing cross-modal fusion to obtain fused features; performing subspace projection on the fused features in the neighborhood of candidate locations, determining anisotropy measures based on local statistics, and estimating the number of endmembers, distinguishing between in-bag and out-of-bag operations under this measure, and determining the number of active endmembers in conjunction with cross-scale stability criteria; adjusting the decoding channels and thresholds according to the number of active endmembers, and feeding the results back to the fusion process; decoding the adjusted fused features and outputting a pixel-level defect segmentation map; thus improving the localization accuracy and segmentation stability of weak feature defects under the condition of limited defect samples.

The invention discloses a large model generation Chinese text detection method and system based on lexical probability statistics. The method comprises the following steps: firstly, calling a proxy large language model to reasone a target text to obtain the logarithmic probability of each token, and then respectively calculating global statistics and local statistics based on a token logarithmic probability sequence of the target text; and finally, dividing the global statistical magnitude by the local statistical magnitude to obtain a detection score of the target text, and comparing the detection score with a given threshold value to make a decision. According to the method, the global and local statistics of the token logarithmic probability sequence of the target text are combined, and low-cost and high-performance large language model generation text detection is realized. Compared with an existing detection method only utilizing global statistics, the method has the advantage that higher detection precision is realized in complex scenes such as cross-model and cross-domain scenes.

The invention discloses a community event data processing method and system based on a double-layer Dccu architecture, and belongs to the technical field of smart communities and grassroots governance, and the method comprises the steps: a cell Dccu server deployed in a cell collects data, and generates a trusted event through the recognition of a built-in AI model; a community Dccu server deployed in a community receives the trusted event through a private network, carries out local statistics and non-cloud processing, generates a report form and a work order and issues the report form and the work order; and finally, receiving a disposal result to realize closed-loop management. The system comprises the cell Dccu server and the community Dccu server. According to the invention, through a double-layer cooperative processing architecture of a cell and a community, in combination with private network transmission and local AI processing, full-process closed-loop management of event data from identification, generation, transmission, disposal to feedback is realized, and the real-time performance of event processing, the data credibility and the grassroots governance efficiency are effectively improved.