898 results about "Confidence threshold" patented technology

The invention discloses a multi-modal perception and optimization method and system for a low-power-consumption AR device, and the method comprises the steps: collecting multi-modal data, task demands, resource state data and environment data for the AR device; lightweight processing is carried out on the multi-modal neural network model through model pruning, parameter quantification and distillation technologies; inputting the collected data into a lightweight multi-modal neural network model for dynamic reasoning to obtain a multi-modal recognition result; comprising the steps of executing modal adaptive weight acquisition based on task requirements and environment data; executing energy consumption constraint scheduling according to the equipment resource state data, dynamically selecting a reasoning path strategy, and obtaining corresponding modal feature output; multi-modal feature fusion is carried out, task reasoning is completed, and a multi-modal recognition result is obtained; early-leaving control is executed based on a middle-layer confidence coefficient threshold value in the reasoning process; and interactively outputting a real-time multi-mode identification result. According to the invention, energy efficiency and precision balance and multi-mode fusion low-power-consumption optimization can be realized.

The invention relates to the field of visual processing, and provides a visual classification processing method and device based on large model and multi-modal data fusion. The method comprises the following steps: inputting a to-be-classified input image and a corresponding category text description into a text encoder for multi-level feature extraction to obtain global text features and local text features; performing fine-grained cross-modal alignment on the local visual features and the local text features, calculating association weights between the image regions and the text phrases through a bidirectional cross attention mechanism, and generating aligned intermediate features; splicing and fusing the aligned middle features and the global visual features, and inhibiting background noise in a fusion result and reinforcing discriminative features in the fusion result through a feature mask algorithm in combination with the global text features to obtain multi-modal fusion features; and synchronously inputting the multi-modal fusion features into a multi-space classifier to generate respective classification results, and adaptively outputting an image classification result according to a confidence threshold in combination with a dynamic routing mechanism.

Artificially intelligent systems and methods for financial coaching provide personalized, fiduciary-compliant financial guidance through advanced machine learning architectures with measurable performance criteria. The systems implement privacy-preserving processing pipelines that detect personally identifiable information using multi-layered pattern recognition including regular expressions for formatted data sequences, named entity recognition with confidence thresholds above 0.85, and contextual analysis algorithms. A multi-step artificial intelligence processing workflow includes automated language detection, emotional tone classification with confidence scoring, financial profile transformation using predefined templates, context-aware question rephrasing, and semantic similarity matching employing vector embeddings with financial domain vocabulary weighting applying multiplier values between 1.3-2.0. Specialized training methodologies expand datasets through mathematical transformation functions utilizing statistical standard deviations with incremental variations between 0.5-2.0. Mood-based escalation logic automatically transfers users to human advisors when emotional indicators exceed confidence thresholds above 0.8. The systems maintain response times below 5 seconds while providing regulatory compliance through curated content sources and predefined fiduciary instruction parameters.

The invention relates to the technical field of transformer substation intelligent inspection, in particular to a meta-learning-based small-sample transformer substation equipment state adaptive inspection system, which comprises a state acquisition module for acquiring the current feature vector and environmental parameter data of a target node; the drift detection module is used for comparing environment parameters to judge data drift and dynamically adjusting a confidence coefficient threshold value; the risk assessment module inputs the feature data into a meta-learning model to output an initial risk probability, and generates an effective risk probability based on threshold filtering; the blind area measurement module is used for acquiring unobserved nodes and calculating system state blind area entropy; the scheduling decision-making module is used for comparing the blind area entropy with a threshold value and generating an entropy reduction bottom instruction or a self-adaptive routing inspection distribution instruction; the strategy updating module is used for extracting an actual inspection result and feeding back to the model for parameter updating; according to the invention, the scheduling difficulty when the resources are limited is solved, and the self-adaptive capability of the system under different environment interferences is improved.

The invention provides a multi-sensor fusion welding temperature monitoring method and a related device, and the method comprises the steps: dividing a welding region into a plurality of sub-regions according to the spatial distribution of the welding region, and obtaining the initial temperature data corresponding to each sub-region according to various temperature sensors; performing time alignment on the initial temperature data through difference resampling of different sampling points to generate a time sequence temperature matrix; performing real-time quality detection on the time sequence temperature matrix to generate a real-time quality score, and performing weighted combination on the real-time quality score and a preset confidence coefficient parameter table to obtain a dynamic confidence coefficient result set; fusing the effective temperature data exceeding a preset confidence threshold in the dynamic confidence result set according to a fusion rule corresponding to each sub-region to obtain a fused temperature result; and the fusion temperature results of the sub-regions are subjected to time sequence splicing to obtain a global temperature curve of the welding process, so that the accuracy of temperature detection in the welding process can be effectively improved.

The invention provides a BIM component automatic identification and storage method and system based on multi-modal artificial intelligence fusion, and relates to the technical field of building information, and the method comprises the steps: receiving a BIM component model file, and preprocessing an obtained attribute parameter table; inputting the information into a multi-modal recognition engine, and outputting a classification result and confidence through Rapidfuzz fuzzy matching file names, OPEN-CV + YOLOv8 image recognition and knowledge graph analysis parameters; the dynamic fusion decision-making module calculates the final confidence coefficient according to the dynamic weight, and solves classification conflicts in combination with a dynamic rule base of industry specification rule priority; according to a confidence coefficient threshold value, judging automatic warehousing or triggering grading manual auditing; manual audit data are collected and fed back to the multi-mode recognition engine, and incremental knowledge self-learning is achieved. Through multi-modal fusion and dynamic decision making, the BIM component recognition accuracy and warehousing efficiency are remarkably improved, personal errors are reduced, and dynamic updating of a component resource library and system self-evolution are achieved.

The invention provides a knowledge base and large language model-based product quality anomaly reason searching method, apparatus and device, and a medium. The method comprises the following steps: collecting quality anomaly-related historical multi-source heterogeneous data from a product life cycle-related system in real time or at regular time; the collected data are preprocessed; constructing a structured domain knowledge base; model training; performing abnormal feature extraction based on abnormal event triggering, and matching the extracted abnormal features with related causal association rules and causal atlas nodes in the constructed knowledge base to form a preliminary suspected reason set, and performing corresponding triggering condition conformity evaluation and historical case matching degree; calculating the final confidence of each candidate reason in the possible reason set through a preset fusion algorithm; at least one root cause is determined according to a set confidence threshold; and generating a targeted solution suggestion report based on the output root cause and the corresponding reasoning explanation text in combination with solution measures pre-stored in a knowledge base.

The invention belongs to the technical field of semantic recognition, and discloses an extraction analysis method and system based on semantic expression understanding. The method comprises the following steps: performing semantic analysis and feature extraction on an original query statement input by a user in an interactive interface to generate a structured semantic unit; performing interaction scene judgment on the structured semantic unit based on a preset multi-dimensional judgment condition and then outputting a scene identifier; calling a corresponding target intention recognition strategy according to the scene identifier; inputting the structured semantic unit into a domain classification model, and outputting a plurality of candidate intentions and corresponding initial confidence; and determining a target user intention from the candidate intentions based on an interaction clarification result of the user and an intention query rule by using the initial confidence and a preset confidence threshold, and generating a standardized intention recognition result. According to the mode, the user semantics can be deeply understood, the recognition strategy is dynamically adjusted according to the context, and accurate, efficient and sustainable extraction analysis is carried out on the user intention through man-machine cooperation and closed-loop feedback.

Transactional Neural Reasoning AI (TNRAI) is a novel class of artificial intelligence designed to simulate human-like reasoning during live, multimodal user sessions. TNRAI departs from traditional static inference models by integrating a five-pillar architecture: (1) delta-path modeling for real-time outcome deviation detection, (2) skew-based adversarial recognition, (3) vector memory recall for behavioral context, (4) ambient reasoning overlays to incorporate situational data, and (5) a multi-logic arbitration engine that fuses rule-based, statistical, situational, and historical reasoning. The system evolves continuously via a CI / CD feedback loop, adjusting its logic and thresholds based on live outcomes. TNRAI supports overlays such as time or location constraints to influence reasoning and can activate conditional triggers based on logic patterns or confidence thresholds. This architecture enables adaptive, deliberative decision-making in real time, extending the utility of AI across domains such as automation, compliance enforcement, customer engagement, and transaction-based system control.

The invention provides an offshore ship suspicious navigation behavior identification method and system, and the method comprises the steps: collecting AIS, radar, satellite and other multi-source data of a ship in a target sea area, and carrying out the data cleaning, fusion and standardization preprocessing; extracting navigation features, staying features, mode features and interaction features of the ship, constructing a behavior recognition model based on a neural network, inputting ship features, and then outputting a recognition result and confidence; and judging result validity through a confidence coefficient threshold value of dynamic calculation, triggering artificial review or multi-source data secondary verification through an invalid result, generating early warning information when a valid suspicious behavior is identified, calculating a risk value in combination with a ship type, a behavior severity degree and an environmental impact factor, and matching a pre-constructed decision library to generate a decision suggestion. According to the method, the recognition precision is improved through multi-source data fusion and an intelligent model, and real-time and efficient monitoring and response to suspicious navigation behaviors are realized in combination with a confidence coefficient mechanism and risk decision support.

The invention discloses an unmanned aerial vehicle visual angle multi-target tracking method and system based on improved ByteTrack, and belongs to the technical field of target tracking, and the method comprises the steps: training a target detection algorithm, and carrying out the target detection of a public data set through employing the trained target detection algorithm; setting a high-score confidence threshold and a low-score confidence threshold, classifying detection frames in the target detection result, and dividing the detection frames of the target detection result into high-score detection frames and low-score detection frames; an improved ByteTrack multi-target tracking algorithm is adopted, a shielding matching module is introduced, and target matching is carried out on a high-score detection frame and a low-score detection frame for three times; and updating all successfully matched track frames based on a matching result, reserving a preset frame number of unmatched track frames for subsequent frame matching, and outputting a tracking result of the current frame. The method improves the tracking accuracy and robustness, and is suitable for the multi-target tracking task in the aerial video of the unmanned aerial vehicle.

The invention discloses an emergency rescue real-time human body detection method and device based on time sequence motion feature enhancement, and the method comprises the steps: obtaining visible light and infrared image sequences of a rescue region and environment parameters (including smoke concentration and illumination intensity) in real time, and carrying out the spatial registration preprocessing; respectively carrying out frame difference processing on the two types of image sequences, generating a binary motion mask, calculating an optical flow amplitude, and carrying out adaptive fusion according to smoke concentration to obtain a multi-scale motion energy field; human body micro-motion frequency band energy is extracted through time-frequency transformation, a frequency domain dynamic attention mask is generated, and a saliency motion target area is extracted in combination with a multi-scale motion energy field; constructing a double-branch neural network, extracting time sequence motion features and multi-modal appearance features, dynamically distributing weights and fusing the weights, and outputting a human body bounding box and detection confidence; and calculating environment complexity according to the environment parameters, determining a dynamic confidence threshold, verifying a detection result by combining the average temperature of the human body bounding box region, and generating alarm information if a condition is met. The method aims at solving the problems that a traditional method is high in omission ratio and unstable in recognition in a complex environment.

The invention relates to the technical field of data processing, in particular to an intelligent supervision method and system based on a data visualization platform, and the method comprises the steps: collecting multi-source water conservancy data through dual-channel redundancy check, converting the multi-source water conservancy data into a standardized space-time matrix, carrying out the data cleaning through a containerization adaptation and variational auto-encoder, and marking the confidence; hierarchically storing the data based on a confidence threshold, and generating a routing strategy through reinforcement learning to distribute data fragments; calculating a flood peak evolution prediction result by using space-time diagram convolution, calculating an output risk assessment value by combining a Bayesian network and long and short-term memory, and activating federal learning parameter updating when the confidence coefficient is insufficient; the digital twin model loads physical constraint parameters to deduce a flood control scene, and a scheduling instruction set and a rehearsal animation are generated through multi-objective optimization; and dynamically rendering to generate an interactive visual interface, associating a causal deduction path, and triggering federal model parameter updating by user feedback, thereby realizing dynamic collaboration of the forecasting and early warning rehearsal plan function chain.

The invention provides an enterprise employee demand recommendation method and device based on multi-dimensional data driving, and the method comprises the steps: collecting the associated data of enterprise employees, and generating a standardized data set; constructing a dynamic preference model based on the standardized data set; the method specifically comprises the following steps: mining strong association rules of employees-commodity categories by adopting an association rule mining algorithm to form a preference rule base; dynamically adjusting the weight of each rule in the preference rule base on the basis of a time attenuation coefficient; using the employee satisfaction score as a supervision signal, and iteratively adjusting the confidence threshold of each rule in the preference rule base to complete the iterative training of the model; and receiving welfare conditions input by an enterprise, screening commodities conforming to the preference rule base from the commodity pool based on the dynamic preference model, and generating multiple groups of personalized gift bag recommendation information containing the commodities. According to the invention, the defect that the diversified and personalized demands of enterprise customers on welfare demand schemes are difficult to meet at present is overcome.

The invention discloses a semi-supervised remote sensing image semantic segmentation method and device based on a large model, and the method comprises the steps: obtaining an unlabeled remote sensing image, carrying out the zero-sample dense prompt segmentation of the unlabeled remote sensing image through a pre-training general visual large model, and obtaining a third mask; obtaining a labeled remote sensing image, and finely adjusting the lightweight classifier according to the labeled remote sensing image; performing semantic prediction on the third mask, and filtering a semantic prediction result based on a classification confidence threshold to obtain an unlabeled remote sensing image pseudo label; constructing a model training data set and a joint loss function in combination with an unlabeled remote sensing image pseudo label, and constructing a remote sensing image semantic segmentation model; training a remote sensing image semantic segmentation model in combination with the model training data set and the joint loss function; and adjusting the weight of each part of the joint loss function by adopting dynamic course learning to obtain a trained remote sensing image semantic segmentation model. The problems of noise pseudo labels and error accumulation generated in the prior art are effectively relieved.

The invention provides a water quality abnormity early warning system and method based on multi-source data fusion, and relates to the technical field of abnormity early warning, and the method comprises the steps: recognizing a sensitive region in which the water quality is gradually changed and abnormal in a target water area through the time-space correlation characteristics of each piece of water quality multi-source data; performing sensitivity analysis on the response intensity of the water quality gradual change abnormity in the sensitive area to obtain a water quality sensitivity index of the sensitive area; determining an early warning response period when the water quality at the monitoring point is gradually changed and abnormal; performing confidence correction on a water quality abnormal fluctuation early warning threshold value at the monitoring point through the water quality sensitivity index and the early warning response period to obtain an abnormal early warning confidence coefficient of the water quality at the monitoring point; and when the abnormity early warning confidence exceeds a preset confidence threshold, sending a water quality gradual change abnormity verification instruction to an intelligent sensor at the monitoring point, and generating risk alarm information of water quality gradual change abnormity based on feedback verification data. According to the invention, risk early warning can be carried out under the condition that the water quality has gradual abnormal fluctuation.

The invention discloses a liquid chromatographic peak intelligent analysis method and system based on deep learning, and relates to the technical field of data processing. The method comprises the following steps: constructing a chromatographic peak intelligent analysis network based on deep learning, and training the network by using a sample chromatogram until convergence; inputting a to-be-analyzed chromatographic signal into the chromatographic peak intelligent analysis network, and evaluating the reliability of a prediction result based on the probability distribution of the key physical parameters output by the chromatographic peak intelligent analysis network to obtain a result confidence coefficient; dynamically adjusting the number and composition of the sample chromatograms based on the result confidence and a preset confidence threshold, and continuously updating the chromatographic peak intelligent analysis network; and recording an attention mechanism of the chromatographic peak intelligent analysis network as a weight fraction allocated to each data point, mapping the weight fractions back to a time sequence of an original chromatographic signal, carrying out visualization through a color gradient, and superposing the weight fractions on an original chromatographic curve in a semitransparent layer form to generate a chromatographic peak thermodynamic diagram. According to the invention, the efficiency and precision of liquid chromatography analysis are effectively improved.

The invention relates to an intelligent decision-making system for nutrition metabolism collaborative management of elderly chronic disease patients, in particular to the field of nutrition metabolism collaborative management of elderly chronic disease, which is characterized in that drug molecule characteristics and nutrient metabolism paths are dynamically integrated through a multi-modal knowledge graph, and a cross-domain associated three-dimensional knowledge network is constructed; based on a reinforcement learning real-time optimization rule confidence threshold value, the early warning sensitivity is adaptively adjusted according to the degree that the metabolic index of the patient deviates from the safety interval; the streaming conflict detection engine accurately identifies the potential risk of asynchronously input medication and diet data, and triggers graded early warning through space-time alignment and sub-graph matching; the closed-loop evolution mechanism fuses patient compliance feedback and blood potassium change trend, drives the taboo rule base to continuously and autonomously evolve under the constraint of renal function layering, and realizes personalized risk prevention and control and metabolic state collaborative optimization.

The invention discloses an agricultural tractor visual positioning method and system, and belongs to the technical field of agricultural tractor visual positioning, and the method comprises the steps: extracting the multi-scale features of a farmland scene image, and carrying out the feature fusion, and obtaining the multi-scale fusion features; performing adaptive weighting on the multi-scale fusion feature map to generate a final feature map so as to obtain a feature point probability heat map and a feature point descriptor; extracting coordinates and detection confidence of the feature points according to the feature point probability heat map; when the detection confidence of the candidate feature points meets a confidence threshold requirement, matching the descriptors of the candidate feature points with the descriptors of the feature points of the adjacent frames in space to obtain matching pairs; and calculating according to the coordinates of the candidate feature points in the matched pairs, the coordinates of the feature points of the adjacent frames and the internal reference of the image acquisition equipment, and outputting the real-time pose of the tractor. According to the method, the spatial distribution of the feature points can be optimized, so that the discrimination is stronger, and the robustness and the positioning precision of the feature points in a farmland environment are improved.

The invention relates to the technical field of image processing, in particular to a fall detection method, device and equipment and a computer program product. The fall detection method comprises the following steps: acquiring a collected infrared video; determining skeleton key points of a target object in the infrared video; and determining that the target object is suspected to fall under the condition that the confidence of the bone key points is greater than a confidence threshold and the position relationship among the bone key points meets a preset condition. The external video does not contain details such as textures and colors of the shot object, so that the privacy is relatively high; tumble is detected based on the position relation of the bone key points, and tumble detection can be more accurately carried out; by determining the confidence of the key points of the skeleton, the accuracy and reliability of fall detection can be improved, and the probability of misjudgment and misdetection is reduced.

The invention discloses a large language model reasoning acceleration method and device based on two-stage speculative decoding and a storage medium, and the method comprises the steps: constructing and initializing a Trie tree, and inserting a historical corpus, and phrase sequences in a document library or a code library into the Trie tree one by one; in the reasoning process, longest prefix matching is carried out based on a Trie tree, and a candidate draft sequence is generated by adopting branch backtracking and recursive search; performing confidence evaluation on the candidate draft sequence, calculating a joint confidence score of the sequence through probability multiplication and a Top-K screening mechanism, and judging whether the joint confidence score reaches a confidence threshold; if the accumulated confidence of the candidate sequence reaches a threshold value, skipping a small model generation stage, and directly entering large model verification; otherwise, entering a small model draft completion stage; and the final large model takes the replaced and updated draft sequence as final output. According to the method, adaptive acceleration of the decoding process can be realized, and the long text reasoning delay of the large language model is remarkably reduced while the generation quality is ensured.

The invention provides an automatic detection method and system for the grinding machining surface of a precision workpiece, and belongs to the technical field of precision workpiece machining detection.The method comprises the steps that multi-mode image data of the grinding machining surface of a target precision workpiece and grinding parameters in the machining process are obtained; performing fusion preprocessing on the multi-modal image data to generate fused image features; feature extraction and cleaning are conducted on the grinding parameters, and grinding process features are generated; inputting the fused image features into a first-stage defect screening model to obtain a first detection result; when the existence confidence of any first defect is greater than a preset first confidence threshold, inputting the grinding process features into a defect mode diagnosis model to obtain a defect mode diagnosis result of the suspected defect area; determining a target defect detection model according to the defect mode diagnosis result; and inputting a part, corresponding to the suspected defect area, in the fused image feature into a target defect detection model to obtain surface defect geometric feature parameters of the suspected defect area.

The invention discloses a predictive vehicle control system and method for multi-modal consciousness recognition and electronic equipment, and relates to the field of remote monitoring, and the system comprises a multi-modal sensing module, an intention prediction module, a cross-domain pre-control module and a false trigger suppression module. The multi-mode sensing module is used for collecting three-mode data of sight, gestures and voice and dynamically distributing weights based on environmental parameters; the intention prediction module is used for performing time-space synchronization verification on the three-mode data and calculating comprehensive confidence in a grading manner; the cross-domain pre-control module outputs pre-control instructions to a cabin domain, a vehicle body domain and a power domain based on the effective intention; the false triggering suppression module is used for dynamically adjusting a confidence coefficient threshold value based on a driving scene and has a failure backspacing learning function; wherein each module performs data interaction based on a vehicle-mounted Ethernet / CAN bus.

The invention provides an intraoperative neurovascular bundle identification and injury early warning method and system based on artificial intelligence. According to the method, real-time image data are acquired through a multi-modal imaging device, multi-scale blood vessel features are extracted by adopting a three-stage cascade adaptive path convolution (APC) module, the long-range topology perception capability is enhanced in combination with a local-global interaction module (LGIM), global features are integrated by utilizing a cross-scale deformable convolutional network (CSDCN), and dimensions are compressed through a multi-branch feature distillation module (MFDM). And the multi-task classifier synchronously outputs the category, position and injury risk assessment of the blood vessel bundle, and real-time multi-modal early warning is realized through dynamic confidence threshold and mahalanobis distance anomaly detection. According to the invention, the accuracy and real-time performance of intraoperative nerve and blood vessel protection are obviously improved.

The invention provides a rapid broadband signal detection and identification method and device based on a time-frequency graph, and the method comprises the steps: converting a broadband signal into a time-frequency graph, and inputting the time-frequency graph into a YOLO-X network for signal time-frequency positioning after the time-frequency graph is preprocessed; constructing a complete signal instance containing a Welch power spectrum based on a positioning result; outputting signal category and confidence evaluation through two-stage feature analysis; and finally, realizing automatic processing or expert secondary discrimination according to a confidence threshold. According to the method, deep learning and signal processing technologies are fused, and the signal detection speed, the recognition accuracy and the decision reliability in a complex electromagnetic environment are remarkably improved.

The invention relates to the technical field of urban low-altitude safety monitoring, and discloses a low-altitude economic airspace passive safety early warning method and system. The method comprises the following steps: collecting multi-modal observation data streams of a target at different moments; generating multi-modal frame data with a unified timestamp; establishing a spatial confidence distribution map; extracting fusion discrimination features; executing multi-channel small target detection; continuously tracking the preliminary target; and introducing a constant false alarm rate framework to set a dynamic confidence threshold to execute alarm judgment. Compared with the prior art, the technical problem that high-confidence all-weather passive monitoring cannot be realized in urban core areas with dense buildings, serious signal shielding and easy GPS failure due to dependence on single vision or radar information in the prior art is solved. According to the method, the space confidence distribution diagram and the dynamic constant false alarm mechanism based on the shielding diagram are constructed by fusing the multi-modal sensing data, so that the safety early warning of the low-altitude small aircraft is realized, and the passive safety guarantee capability of the low-altitude economic airspace is improved.

The invention relates to the technical field of user portrait information processing, and particularly provides a user real-time portrait updating method and system, and the method comprises the steps: collecting a time sequence interaction track sequence generated by a user in a current interaction scene in real time, and constructing a heterogeneous behavior data flow based on a cross-modal behavior synchronous collection technology; detecting the deviation degree of the historical behavior sequence in the unified feature space, and generating a conflict state mode label; analyzing a behavior evolution path by applying a time sequence self-attention model, and outputting multi-intention probability distribution; dynamically adjusting a confidence coefficient threshold value based on a conflict state and filtering out a noise portrait label; and triggering cross-modal joint recoding and intention re-reasoning for a prediction deviation time period, and locally updating the portrait through a feedback closed loop. Static portrait updating lag limitation is broken through, second-level intention tracking and dynamic behavior mode capturing are achieved, and portrait adaptability and storage efficiency in a high-noise scene are remarkably improved.

The invention discloses a sea drift trajectory prediction model training method and device, and the method comprises the steps: obtaining the environment data of a target sea area, and generating a plurality of physical trajectories based on the environment data of the target sea area and a Lagrange physical model; training the basic sea drift trajectory prediction model by using the plurality of physical trajectories to generate a first sea drift trajectory prediction model; a plurality of real observation trajectories of the data dense area are collected, the plurality of real observation trajectories and a direction and distance combined loss function are utilized to perform fine adjustment on the first sea drift trajectory prediction model, and a second sea drift trajectory prediction model is generated; environment data of the data sparse area is collected, multiple pseudo label tracks distributed like the multiple physical tracks are generated based on the environment data of the sparse area and a second sea drift track prediction model, and multiple high-credibility tracks with the confidence coefficient smaller than a confidence coefficient threshold value are calculated and screened; and training the second sea drift trajectory prediction model by using the plurality of highly credible trajectories to generate a target sea drift trajectory prediction model.

The invention relates to the technical field of intelligent glasses payment, and discloses a payment method and system based on intelligent glasses. The method comprises the following steps: collecting user iris biological characteristic data through an intelligent glasses eye pupil recognition module to generate an iris characteristic sequence; performing feature point extraction and pattern analysis on the iris feature sequence to obtain an identity verification result; when the identity verification result reaches a preset confidence threshold value, activating the near field communication module to establish a payment channel; acquiring payment request data of the transaction terminal through the near field communication module to generate a transaction information packet; performing hierarchical encryption on the transaction information packet by adopting a block chain encryption algorithm to generate an encrypted transaction block; constructing a distributed verification chain based on the digital signature feature and the time sequence feature of the encrypted transaction block; performing consensus calculation on verification results of the nodes through the distributed verification chain to obtain a transaction verification state; according to the transaction verification state confirmation level and the network load, dynamically adjusting block chain network parameters; and finishing final execution of the payment instruction.

The invention discloses an infrared small target detection method and system based on a visual autoregression teacher model, and relates to the technical field of computer vision and image processing. The method mainly comprises the following steps: constructing and training a VAR teacher model, extracting multi-scale features by using a frozen pre-trained visual autoregression model, and adapting general features to an infrared small target detection task through a lightweight adapter module; using the trained teacher model to generate a high-quality pseudo label for the target domain unlabeled infrared image, and carrying out confidence threshold screening and non-maximum suppression; training a lightweight student detector by using a joint loss function based on a self-training framework in combination with the source domain labeling data and the target domain pseudo-tag; and updating the average teacher model through an index moving average mechanism for final reasoning and evaluation. According to the method, the problems of low precision, low robustness and weak generalization ability in cross-domain infrared small target detection are solved.