17results about How to "Improve discrimination accuracy" patented technology

The invention discloses a ring main unit fault processing method and device, a storage medium and a program product, and belongs to the technical field of power distribution automation. The method comprises the following steps: deploying a current sensor, a voiceprint sensor, a vibration sensor and an arc light sensor in a ring main unit, and synchronously collecting an electrical signal, a sound signal, a vibration signal and an optical signal when a fault occurs; carrying out preprocessing and feature extraction on the collected multi-modal data to obtain multi-dimensional feature parameters such as current amplitude, spectrum features, vibration energy and light intensity features; and calculating fault confidence based on a weighted fusion algorithm, and outputting fault type and position information when the confidence exceeds a judgment threshold. By fusing electrical quantity and non-electrical quantity information, real fault and non-fault disturbance can be distinguished, atypical faults such as high-resistance grounding and arc faults can be accurately identified, the false alarm rate is reduced, the reliability and the intelligent level of a ring main unit protection system are improved, and the method is suitable for intelligent operation and maintenance scenes of an urban power distribution network.

This invention provides a method and system for early warning of debris flow initiation based on dynamic weighting of multiple influencing parameters, belonging to the field of geological disaster prevention and early warning technology. The method includes: constructing a set of multiple influencing parameters by acquiring source conditions, topographic features, and real-time rainfall parameters; constructing an intrinsic weighting function based on the initiation physical mechanism to calculate the intrinsic weights of the source and topographic parameters; adjusting the intrinsic weights using a dynamic correction function with rainfall intensity or cumulative rainfall as input to obtain dynamic weights; establishing a critical function coupling the dynamic weights and critical conditions to calculate the debris flow initiation critical index, and issuing an early warning when the threshold is exceeded. This solves the problem that existing debris flow early warning methods are unable to simultaneously integrate physical mechanisms and dynamic weight adjustment, and accurately reflect the critical state of debris flow initiation under the coupling effect of multiple factors.

The invention relates to the technical field of quality classification, in particular to a mutton quality classification method and system based on multi-source data analysis, and the method comprises the following steps: constructing a distribution curve mapping feature space based on fatty acid data to generate a variety attribution region, constructing a decoupling coefficient characterization shearing force by using a spectral peak difference, and constructing a multi-source data analysis model; and calculating the molar ratio of trace elements to construct a competition vector, solving a steady-state deviation angle to judge the pollution risk, and matching multi-dimensional features to generate a quality classification result. According to the method, the distribution characteristics between the chain length and the unsaturation degree are constructed and mapped to the high-dimensional space to distinguish the variety attribution areas, the shear force intensity is quantified by combining the difference distance between near-infrared characteristic peaks, the metabolic steady-state deviation angle is constructed by using the molar ratio between antagonistic elements, and the pollution risk level is evaluated accordingly. In the classification process, a multi-source heterogeneous data fusion and quantitative feature extraction mechanism is introduced, so that the stable judgment capability of the mutton quality and the judgment precision of the pollution risk are enhanced.

The application discloses a mineral identification analysis method and system based on calculation of a polarized light field reconstruction, belongs to the technical field of geological mineral analysis and microscopic imaging, and solves the triple bottleneck problem of physical contradiction-algorithm fragmentation-experience dependence in the prior art. At least one group of original light field image sequences of a sample to be measured in multiple visual angles and multiple polarization states are acquired; registration processing is performed on the original light field sequence images, so that the images under all visual angles correspond in spatial positions, and registration data are obtained; a degradation relation model for describing the original light field image sequence and a target light field image is established; based on the degradation relation model and the registration data, a polarization synthesis image is generated through reverse solving; morphological texture features and optical physical features are extracted from the polarization synthesis image, and mineral types and distribution information are determined. The application realizes automation, quantification and intellectualization of mineral identification, and improves the accuracy and efficiency of analysis.

The application relates to the field of artificial intelligence, and discloses a kind of artificial intelligence's slub yarn fabric visual inspection method and system, to solve the problems of low recognition rate of micro-defects, high false detection and missed detection rate, poor environmental adaptability and model updating lag in prior art under complex texture background.The method comprises: collecting fabric images through high-resolution line array camera and multispectral programmable light source; constructing a five-layer multiscale texture feature pyramid network, fusing local gradient and global semantic features; deploying a cloud dynamic defect discrimination model, combining defect prior knowledge to guide pixel-level segmentation and classification; establishing an adaptive parameter optimization engine and incremental learning mechanism, the application realizes high-precision automatic detection of slub morphology anomaly and yarn breakage defect through the above-mentioned technical cooperation, the detection rate reaches 99.7%, the false detection rate is less than 0.5%, the detection speed reaches 120 meters / minute, and the online quality control demand of high-speed production line is met.

The application discloses a multi-modal satire detection method based on a large language model, and comprises the following steps: constructing a large-scale high-quality text data set; constructing a pre-training language model; adopting a supervised fine-tuning strategy to optimize parameters of the pre-training language model, training the pre-training language model through a cross-entropy loss function of self-recurrence language modeling, and obtaining a multi-modal large language model; and inputting the large-scale high-quality text data set into the multi-modal large language model for processing, and obtaining a detection result.

The application relates to the technical field of geological prospecting, and provides a method for comprehensively optimizing a prospecting target area. In the method, a grade interval boundary limit value of a target area metallogenic possibility is determined, a comment set score model corresponding to the grade interval is constructed, a score matrix of a geochemical exploration index of the prospecting target area and a score matrix of a geological index are generated; then, according to a corresponding weight matrix, a membership score matrix of the geochemical exploration index and the geological index of the prospecting target area in the grade interval is calculated respectively, so as to determine an index score comment of the prospecting target area in the grade interval; and according to the grade membership of the prospecting target area in the grade interval and the interval upper limit value of the grade interval in the index score comment, the comprehensive membership of the prospecting target area is calculated, and the optimized prospecting target area is output by sorting the comprehensive memberships of all the prospecting target areas. Therefore, the optimized prospecting target area is scientifically determined through theoretical calculation, and the deviation and interference of human factors are effectively avoided.

The invention relates to the technical field of student learning state monitoring, in particular to a student learning state intelligent monitoring method and system based on behavior analysis, and the method comprises the steps: a data collection unit obtains a glancing path sequence and glancing drop point coordinates of a student in real time through eye movement tracking equipment; the glancing judgment unit constructs a glancing feature judgment model, dynamically optimizes an angle deviation accumulated value threshold value, strengthens abnormal steering contribution in combination with a direction weight factor, and distinguishes target glancing and roaming glancing through speed fluctuation dual-threshold secondary verification of a critical interval; the region matching unit generates a keyword region through natural language processing and a visual saliency model, quickly calculates an Euclidean distance based on R tree index and coordinate normalization, maps the Euclidean distance into a matching degree score and divides a high matching region and a low matching region; and the state decision-making unit outputs a concentration state through a quaternary matrix, continuous N times of targeted glancing and a high matching area, accurately monitors a learning state, and provides a reliable basis for teaching intervention.

This invention relates to the field of gravity mineral processing technology, and provides a vision-based method for detecting and controlling the slurry flow rate of a spiral concentrator. The method includes: acquiring slurry images at the inlet of the spiral concentrator; extracting the edge geometric features of each slurry image to construct a slurry edge feature image sample set; constructing and training a slurry edge recognition model based on the sample set; identifying the slurry edge position using the recognition model; constructing a geometric relationship model between the slurry edge and the outer edge of the spiral concentrator using the identified slurry edge position; calculating the geometric distance from the slurry edge to the outer edge of the spiral concentrator using the geometric relationship model; determining the slurry flow rate using the geometric distance, and controlling the slurry flow rate. This invention enables visual determination of the slurry flow rate in a spiral concentrator, efficiently performing automatic control of the slurry flow rate without relying on human experience and subjective judgment, thus significantly improving the accuracy of the determination.

This invention discloses a method and system for identifying waste materials based on multimodal fusion, belonging to the field of intelligent waste material identification technology. The method involves acquiring visible light image data, depth structure data, and near-infrared spectral response data of the waste materials to be identified; obtaining candidate regions from the visible light image data and extracting a set of morphological features based on the depth structure data; extracting material reflection curves from the near-infrared spectral response data; extracting texture distribution features and color stability features from the visible light image data and generating a set of structural consistency features based on the morphological features; constructing a fusion feature vector by performing cross-modal association mapping on multiple features and inputting it into a waste material identification model to obtain candidate identification results; identifying key feature subsets through local sensitivity analysis and reconstructing the weights of the fusion features; and performing classification again to output the final identification result. This invention can effectively improve the accuracy and stability of waste material identification in complex recycling scenarios.

This invention discloses a Fe3O4@SiO2 / polyurethane composite photonic crystal sensor, its preparation method, and its applications, belonging to the field of photonic crystal sensor technology. This invention achieves spatial modulation of the C6 fluorescence signal and enhancement of the local photonic field by spectrally matching the reflection band of the structured photonic crystal with the C6 emission band. This results in significant fluorescence quenching response even under extremely low PA concentrations, overcoming the limitation of low sensitivity in traditional fluorescence detection. Compared with traditional HPLC or conventional fluorescence spectroscopy methods, the photonic crystal enhancement system of this invention can improve PA detection sensitivity by approximately 3-4 orders of magnitude, while exhibiting good reproducibility and environmental stability.

The invention relates to the technical field of geological disaster monitoring, in particular to a landslide analysis method based on deep learning and satellite images. The method comprises the following steps: S1, acquiring a satellite remote sensing image and landform data of a landslide area, and preprocessing the satellite image; S2, constructing a feature set according to a satellite image preprocessing result in combination with the landform data; s3, constructing a training sample set based on the feature set, classifying and positioning the areas conforming to the initial features of the landslide, and outputting the boundary and the distribution range of the landslide; s4, outputting an expression result according to the landslide boundary and the distribution range, and performing multi-index precision evaluation; and if the precision is smaller than the set value, returning to the step 2 and / or the step 3 to carry out adjustment and optimization. The invention aims to provide a landslide analysis method based on deep learning and satellite images so as to realize rapid and high-precision landslide analysis through the satellite images.

The application discloses a traffic event detection method, device and equipment and a storage medium. A ball camera is controlled to sequentially acquire video frames of each local field divided by a global field of view of the ball camera according to preset polling parameters. The video frames of all local fields acquired by the ball camera are rearranged to obtain a video sequence set of the global field of view. Multi-dimensional convolution operation is performed on each video sequence unit in the video sequence set to obtain input features of the global field of view at each moment. The obtained input features are input into a convolutional neural network to extract visual features at each moment. The obtained visual features are input into a recurrent neural network in time sequence for analysis to obtain occurrence probabilities of each event at each moment. Only one variable-focus rotating ball camera is deployed, the global field of view to be monitored is divided into a plurality of local fields, the hardware cost and the deployment and maintenance cost are reduced, the ball camera has the characteristics of light weight and low cost, and the real-time traffic event fine and high-precision discrimination and management capability is improved.

The application discloses a kind of capacitive capacity monitoring method and system of flexible direct current converter station bushing, it is related to the field of flexible direct current transmission equipment on-line monitoring and electrical parameter identification technique, including in the bushing of flexible direct current converter station bushing end screen and ground setting capacitor voltage divider and constructing series capacitor measurement structure, acquisition bushing primary voltage signal and end screen current.Primary voltage signal and end screen current are carried out synchronous signal preprocessing, using wavelet transform and filtering strategy to remove noise and extract frequency domain features.Capacitive capacity coupling relationship of bushing structure parameter is calculated based on frequency domain features, and abnormal point is judged by establishing capacitive capacity change identification model and combining temperature and humidity data.The method described in the application effectively overcomes the problems of static analysis lag, single variable misjudgment and insufficient frequency spectrum resolution, realizes real-time monitoring of flexible direct current bushing capacitive capacity, abnormal dynamic early warning and system improvement of model adaptability, and significantly enhances the safety operation guarantee capability of equipment throughout its life cycle.

This invention provides a dynamic mechanical structure wind vane ground-mounted landscape tower, relating to the field of communication landscape towers. It includes a landscape column, a flipping guide rod, a marker head, a marker tail, and an extension plate. The landscape column has two annular sub-tower structures on its upper and lower sides of its outer end face, symmetrically positioned. The landscape column's outer end face has four outward-facing adjustment cavities. This design uses a threaded connection to move a telescopic connecting cylinder outward, which in turn moves the connecting guide plate, and then, through transmission, moves the directional marker outward. As the directional marker moves outward, it makes it easier for pedestrians to understand the wind direction, providing a direct visual representation of the wind's direction. This design also deforms the extension plate, increasing the wind-receiving area of ​​the marker tail, thereby expanding the wind-sensing area and improving the wind vane's sensing accuracy. This facilitates adjustment and sensing of the wind direction angle, and the change in the dynamic mechanical structure enhances the wind vane's sensing effect.

The application discloses an intelligent ground rail welding defect detection system based on deep learning, which establishes a double-path parallel information representation channel through frequency-space domain multi-scale feature decoupling, introduces an adaptive wavelet phase attention module in the cross-domain feature extraction stage, implements amplitude attenuation by using a learnable frequency mask, based on the physical prior that the phase gradient of a crack is continuous when the crack propagates along the stress direction, implements secondary interception of residual invalid phase through a phase continuity gating mechanism, and enlarges the feature distance of true and false defects in the hidden space through a momentum comparison classifier to realize high-confidence discrimination. Through the synergistic effect of the dual physical constraints of amplitude aperiodicity and phase direction continuity, the system effectively eliminates the phase artifacts caused by frequency band aliasing, thereby significantly reducing the mis-detection rate of the boundary area and improving the discrimination accuracy and availability of the system in the industrial field.

This invention discloses a two-dimensional heterojunction array sensing method and device for high-precision monitoring of carbon pollution. The two-dimensional heterojunction array sensing method includes: emitting light signals with dynamically changing wavelength, power, and modulation frequency to a two-dimensional heterojunction array using a tunable light source according to a preset switching sequence; acquiring the electrical signals generated by the two-dimensional heterojunction array under the combined action of the target gas and the light signals, and converting the electrical signals into digital signals in multi-dimensional matrix form; and using a hybrid model integrating convolutional neural network branches and recurrent neural network branches to extract spatial pattern features and dynamic temporal features, ultimately outputting the type and concentration of the gas. This invention effectively solves the cross-sensitivity problem in gas identification by exciting the two-dimensional heterojunction array with optical sequence modulation and combining it with a lightweight neural network model, thus improving the identification accuracy and response speed in complex environments.