6results about How to "Realize continuous tracking" patented technology

The application discloses a layered sampling monitoring device and method for freeze-thaw cycle test, which comprises a freeze-thaw cycle box, a layered soil sample culture unit, an in-situ monitoring unit and a non-destructive sampling unit. The application provides a layered sampling monitoring device and test method for freeze-thaw cycle test, which has a reasonable structure design and is convenient to operate. The device can realize non-destructive or slightly destructive extraction of target soil samples after different freeze-thaw cycles without damaging the continuity of the whole test, and can realize real-time monitoring of the moisture content and temperature changes of the soil samples in the freezing and thawing process, thereby providing accurate and continuous test data for in-depth study of the evolution law of the freeze-thaw cycle on the soil properties. The device and method of the application significantly improve the continuity and accuracy of the data, and provide a powerful tool for revealing the effect of freeze-thaw on the soil erosion mechanism.

This invention relates to a health monitoring method for carbon fiber composite structures, comprising: acquiring electrical conductivity data of the composite structure under stress; generating an initial resistance distribution map and identifying abnormal regions based on the electrical conductivity data to obtain a preliminary signal set containing potential damage locations; extracting spatial feature vectors from the preliminary signal set; processing the spatial feature vectors using a feature extraction model to filter out noise interference to obtain an enhanced signal sequence; processing monitoring data at multiple time points using a time-series fusion model to generate a damage evolution trend map reflecting the damage propagation trend; performing hotspot region analysis based on the spatial distribution of signal peak points in the damage evolution trend map; and adjusting the distribution density of sensing channels to generate optimized monitoring system configuration parameters if the spatial dispersion of the peak points exceeds a preset range.

The application relates to a traffic flow detection method based on improved YOLOv8 and DeepSORT, which comprises the following steps: S1, acquiring a training data set; S2, obtaining initial values of hyperparameters of a target detection model; S3, after the target detection model is trained based on the training data set for multiple times, whether a first exit condition is met is judged, if yes, step S6 is executed, otherwise, step S4 is executed; S4, a Pareto Bayesian algorithm is used to update the hyperparameters to obtain candidate values of multiple groups of hyperparameters, and a genetic algorithm is further used to process the obtained candidate values of the multiple groups of hyperparameters to obtain a candidate value of a group of hyperparameters to replace the hyperparameters in the target detection model; S5, whether a second exit condition is met is judged, if yes, step S6 is executed; S6, each video frame of a to-be-detected video is detected by using the target detection model to obtain traffic flow data. Compared with the prior art, the application has the advantages of improving the detection accuracy in an unmanned aerial vehicle scene.

The present application relates to the technical field of wharf structure safety evaluation, and provides a wharf resilience evaluation method based on 4R four-dimensional coupling, comprising the following steps: responding to a wharf resilience evaluation request, constructing a 4R four-dimensional resilience evaluation system; mapping into a Bayesian network, wherein the root node of the Bayesian network is a basic index, the intermediate node is a 4R-dimensional index and corresponding sub-index, and the leaf node is a wharf comprehensive resilience grade; generating a root node probability according to historical data, and generating a conditional probability table of the intermediate node and the leaf node; converting the root node prior probability and the conditional probability of the intermediate node and the leaf node into high / low probability, inputting the Bayesian network, calculating important factors of each root node on the comprehensive resilience grade, identifying high sensitivity values as weak links through reverse diagnosis, and determining a priority repair or investment order based on the evidence high sensitivity value. The index system and the Bayesian network based on 4R four-dimensional coupling are constructed, and the probabilistic evaluation of the wharf resilience, the identification of key short boards and the dynamic and continuous tracking are realized.

The application is suitable for the technical field of data processing, and particularly relates to a data security risk early warning method and system based on artificial intelligence. The method comprises the following steps: obtaining metadata list and sensitive data list; matching data stream event set with sensitive data in the sensitive data list and data dyeing to obtain data stream event list; constructing data stream graph; calculating final risk value of each node; adding nodes with final risk value greater than initial threshold value into infection source set; calculating predicted infection probability of each node in future time window based on final risk value of each node and infection source set; determining dynamic early warning threshold value based on final risk value of each node and predicted infection probability; if final risk value is greater than dynamic early warning threshold value, triggering early warning and generating early warning report of nodes with final risk value greater than dynamic early warning threshold value. The method can quickly locate the affected nodes and predict the diffusion path after the data security problem occurs.

The invention relates to a lightweight convolutional neural network construction method and system for glue pudding defect detection, and the method comprises the steps: constructing a color intensity distribution interval for a brightness contrast value and a color saturation change rate in each region according to a preliminarily labeled gradient region distribution diagram, analyzing and calculating the intensity change trend in the region through a histogram, and obtaining a color intensity distribution interval; determining a quantitative description value of the color difference; adjusting local focusing parameters during image acquisition according to classification labels of defect severity in combination with crack depth estimation values and color intensity distribution intervals, optimizing the detail capture capability of high-risk areas, and determining a monitoring enhancement scheme for specific defects on the surfaces of the rice dumplings; and by monitoring a strengthening scheme, updating crack distribution density ratio and color distribution uniformity degree data processed by a production line image, generating a dynamic detection template aiming at the stuffing seepage defect on the surface of the sweet soup ball, and finishing continuous tracking and identification of a specific defect type.