76results about How to "Improve capture ability" patented technology

The invention relates to the technical field of intelligent evaluation of garment flatness, and particularly discloses an automatic evaluation method of a non-ironing suit seam flatness grade, and the method comprises the steps: obtaining image data and physical parameter data of a seam region in response to a detection trigger signal of the seam region of a non-ironing suit; and performing key scale screening based on texture contribution degree on the image data, calculating a relevance index between feature distribution and flatness grade probability distribution under different texture scales, and determining the texture scale of which the relevance index meets a preset contribution degree condition as a target key scale. According to the method, the entropy screening mechanism based on the texture contribution degree is introduced, so that the key scale which is most effective for flatness representation can be locked from the multi-source data in a self-adaptive manner, and the refinement and pertinence of feature extraction are realized.

This invention discloses a method for extracting the contours of point clouds on the surface of castings based on neural networks. The method comprises the following steps: S1: acquiring point cloud data of the casting surface; S2: preprocessing the point cloud data; S3: constructing a point cloud edge detection network based on a UNet++ neural network and training the network to obtain a trained point cloud edge detection network; the point cloud edge detection network includes an input layer, an encoder module, a feature extraction module, an attention module, a decoder module, and an output layer connected in sequence; S4: identifying the contours of various parts of the casting surface based on the trained point cloud edge detection network. This invention enhances the ability to capture detailed features of the target point cloud by constructing a point cloud edge detection network and extracting attention features of the center point features, while also exhibiting stronger robustness to sparse and noisy point clouds. Extracting the point cloud contours of parts on the casting surface using the point cloud edge detection network improves the accuracy of robots in identifying different shaped parts of the casting surface and provides better adaptability in complex industrial scenarios.

This invention discloses a multi-source electromagnetic noise suppression method based on noise classification and deep learning, belonging to the field of geophysical electromagnetic exploration technology. The method includes: segmenting and preprocessing the original electromagnetic observation sequence; using an improved U-Net network with an encoder embedded in a Mamba time-series modeling module for low-frequency noise suppression; identifying strong noise types in the data segments using a ROCKET classifier; based on the classification results, calling a second improved U-Net network trained for the corresponding noise type for class-based targeted denoising; and finally, splicing the data segments to obtain complete, high-quality data. This invention, through a phased processing framework of "low-frequency pre-suppression—noise classification—class-based denoising," combined with the strong time-series modeling capabilities of the Mamba module and the efficient classification performance of ROCKET, significantly improves the suppression accuracy and signal fidelity for complex, multi-source electromagnetic noise, and is particularly suitable for processing ground and airborne electromagnetic exploration data.

The application discloses a face expression capturing model training method based on unsupervised face expression capturing model training method system and medium, obtains a group of preset prompt words for image generation, and randomly extracts a preset prompt word to generate a first two-dimensional face image; obtains a first face expression score vector and a face parameter based on the first two-dimensional face image; calculates three-dimensional face key points, and obtains a second two-dimensional face image according to the three-dimensional face key points; inputs the second two-dimensional face image into an emotion regression network to obtain a second face expression score vector; calculates a preset loss according to a preset loss calculation mode, and trains the face expression capturing model in a gradient back propagation mode. The application can train a model with a driving effect closer to a real face by generating the first two-dimensional face image corresponding to the preset prompt word, combining the face expression score vector, the face parameter and a large amount of unsupervised data, and simplifying the data acquisition process and reducing the cost.

This utility model relates to a magnetic separation device for recovering manganese-iron ore slag, belonging to the field of manganese-iron ore slag recovery technology. It mainly includes a conveying mechanism, a magnetic separation component, a lifting mechanism, and a weak magnetic enhancement component. The conveying mechanism is used for material transport; the magnetic separation component forms a magnetic field coverage area through a main magnet and an auxiliary magnet; the weak magnetic enhancement component is located below the conveyor belt and enhances the adsorption capacity for weakly magnetic substances through an electromagnetic coil group and a magnetic guide plate group; the lifting mechanism adjusts the height of the weak magnetic enhancement component. This application can significantly improve the capture capacity of weakly magnetic substances, reduce the loss of valuable metals, adapt to various material characteristics, and improve resource recovery efficiency, possessing high practicality and promotional value.

The application discloses a smart community police management system and method based on Internet of Things security, and belongs to the technical field of smart communities, and solves the problem that existing system data processing cannot utilize edge computing nodes for low-delay real-time analysis, and cannot meet timely early warning requirements, the method comprising preprocessing of multi-source heterogeneous community monitoring information, analysis and processing of the community monitoring information based on edge computing, and generation of a risk propagation path in combination with front-end analysis results and heterogeneity characteristics; in the application, analysis and processing of the community monitoring information based on edge computing avoids the delay of centralized processing, can quickly trigger early warning in the early stage of a risk event, thereby gaining golden disposal time for risk event disposal, and a risk analysis model realizes a leap from single-node anomaly detection to composite risk quantification through multi-dimensional feature fusion and dynamic space-time reasoning, and realizes multi-dimensional feature fusion and dynamic reasoning capability of the risk event.

The present application relates to the field of computer vision, and specifically discloses a DHA algal oil soft capsule filling amount insufficient detection method based on a deep learning neural network, comprising: accurately separating each capsule instance from a production line transmission image through a lightweight instance segmentation model; synchronously analyzing the global contour and local density features of a single capsule region using a multi-scale feature extraction network, which innovatively introduces an auxiliary density estimation module to enhance the perception of the content filling state; and fusing the above features to complete the accurate classification of the filling state. The present application effectively solves the practical problems of feature confusion, insensitivity to minor changes, poor dense detection effect, and the difficulty in balancing model efficiency and accuracy, and realizes high-precision and high-efficiency automatic detection of DHA algal oil soft capsule filling amount insufficient.

This invention discloses an AI-based physical education exam scoring method, belonging to the field of exam scoring methods, including the following steps: (1) activating the cameras deployed in the exam room to collect motion images and facial expression images of the examinee during pull-ups; (2) analyzing the collected facial expression images using AI image recognition technology to divide the exam process into three stages; (3) for each stage, analyzing the examinee's motion images using AI technology to determine whether each pull-up motion is a qualified motion; (4) counting the number of qualified motions in each stage and scoring them separately, adding the scores of the three stages to obtain the examinee's final exam score. This invention can divide the exam into three stages based on facial expression recognition, perform differentiated analysis on the motion characteristics of each stage, better fit the examinee's physical fitness change patterns, and make the evaluation results more targeted.

This invention relates to the field of data analysis technology, specifically to an intelligent analysis method and system for pile foundation vibration monitoring data. The method includes: acquiring a current vibration subsequence and a sample vibration subsequence; acquiring the energy intensity characteristic value and main frequency characteristic value of the vibration subsequence; obtaining a reference subsequence set corresponding to the current vibration subsequence based on the difference in influence data between the current vibration subsequence and the sample vibration subsequence; obtaining the relative amplitude anomaly index value and the relative frequency anomaly index value of the current vibration subsequence based on the characteristic value difference, time distance, and difference in influence data between the current vibration subsequence and the reference subsequences in the reference subsequence set; and performing pile foundation vibration anomaly early warning based on the relative amplitude anomaly index value and the relative frequency anomaly index value of the current vibration subsequence. Furthermore, this invention can improve the accuracy and reliability of pile foundation vibration anomaly early warning.

This invention discloses a method, system, and storage medium for multivariate time-series feature extraction and grade prediction in flotation processes. The method includes the following steps: Step S1: Raw data input and encoding; encoding and structured input of time-series data composed of various process variables; Step S2: Extracting dynamic features; Step S3: Condition-guided encoding modeling; Step S4: Enhancing the saliency of key variables and important time segments; employing a multi-head attention mechanism to match key-value pairs generated from target-guided query vectors and multi-scale features; Step S5: Outputting a prediction module; performing feature fusion and nonlinear mapping on the attention mechanism output to output the predicted concentrate grade and recovery rate for future times. The system and storage medium are both based on the above method. This invention has advantages such as higher intelligence, better controllability, and improved prediction accuracy and model adaptability for key indicators in the flotation process.

The application relates to the technical field of medical safety, in particular to a patient violence behavior detection method and system based on an intelligent bracelet, which comprises a data acquisition module, a behavior analysis module, an abnormality judgment module and a prewarning execution module. The physiological and motion signals of a patient are collected in real time through multidimensional sensors, a behavior characteristic matrix is generated, a matching degree index is calculated, a violence tendency is analyzed and judged in combination with a threshold interval and time continuity, and graded prewarning and multi-channel notification are realized. The application can significantly improve the recognition accuracy of complex behavior patterns, reduce misjudgment and false positives, ensure that medical staff can respond in time, improve the safety management efficiency, simplify the medical management process, and provide a reliable solution for the whole-process automatic detection and prewarning of patient violence behavior.

This invention discloses a method for monitoring the state trends of hydropower units based on two-stage signal decomposition and the IBiLSTM model. The method involves collecting vibration signals from the hydropower units for preprocessing; constructing the ITGCOA optimization algorithm and designing a fitness function; using the ITGCOA optimization algorithm to adaptively optimize the SVMD and BAACMD models, achieving initial decomposition of the preprocessed signal and secondary decomposition of the sub-mode component with the highest center frequency obtained from the initial decomposition; calculating the fuzzy entropy values ​​of the remaining sub-mode components for reconstruction; and fusing the high-frequency feature sub-sequences obtained from the secondary decomposition with the reconstructed feature sub-sequences to construct the input sequence of the prediction model. This input sequence is then used to construct the IBiLSTM prediction model for monitoring the state trends of hydropower units, achieving high-precision prediction. Compared with existing technologies, this invention improves the efficiency and accuracy of hydropower unit state trend prediction, accurately warns of abnormal unit operating conditions, ensures the safe and stable operation of the units, and improves the overall efficiency of the power plant.

This invention provides a construction progress management method and apparatus based on asymmetric bidirectional complementarity, relating to the field of building construction progress management technology. This invention acquires construction semantic graphs, BIM semantic graphs, and component-level mapping relationship sets; performs incremental information identification and construction oriented towards differences to form incremental information units; constructs and determines the complementary direction based on virtual-real asymmetric quantification of complementary indices; performs hierarchical control and complementary calculation based on complementary indices to achieve asymmetric updates; performs unified expression of construction scenarios and progress reasoning driven by incremental information; and performs iterative updates and progress warnings based on difference-driven mechanisms. This application can transform virtual-real differences into engineering semantic increments, utilizes a complementary mechanism to solve the problem of inaccurate progress judgment caused by occlusion, and achieves dynamic closed-loop updates and accurate warnings throughout the construction process, significantly improving the accuracy and automation level of digital twin construction management.

This application discloses a method, apparatus, device, medium, and program product for identifying student psychological states, aiming to address the problems in existing technologies such as the difficulty in real-time, comprehensive, and dynamic identification of students' psychological states, and the lack of effective early detection and warning capabilities. The method includes: acquiring a multi-dimensional behavioral feature sequence of the target student, which includes at least an expression feature sequence, a behavioral trajectory feature sequence, and a voice emotion feature sequence; inputting the multi-dimensional behavioral feature sequence into a trained psychological state identification model to output psychological state representation information for characterizing the target student's psychological state; and identifying the target student's psychological state based on the psychological state representation information.

This invention provides a method and system for detecting knee meniscus injuries by integrating multi-sequence and contrastive learning, comprising: acquiring multi-sequence MRI training images and images to be detected of the knee joint; performing preprocessing and registration; inputting the preprocessed multi-sequence MRI training images into a feature encoding network based on contrastive learning for feature learning to obtain pre-training weights and optimize parameters; constructing a multi-sequence fusion detection model and loading the pre-training weights; optimizing the parameters of the multi-sequence fusion detection model using labeled data and a detection loss function; inputting the knee joint multi-sequence MRI images to be detected into the optimized multi-sequence fusion detection model to obtain the localization and classification results of the injury. The beneficial effects of this invention are: addressing the core challenges in knee meniscus injury detection, such as the scarcity of high-quality labeled data, insufficient utilization of multi-sequence information, and difficulty in detecting subtle injuries, this invention proposes a systematic solution.

This utility model discloses a support system for extending power generation by staggering peak hours of floating photovoltaic power generation at sea, belonging to the field of photovoltaic power generation technology. It includes a float with multiple columns fixedly connected side-by-side on both sides, each column having a hinged inclined beam at its upper end; multiple adjustment mechanisms corresponding to the inclined beams, used to adjust the rotation angle of the inclined beams relative to the columns and maintain a fixed position at a specific angle; multiple purlins arranged parallel to the column arrangement direction and symmetrically fixed to the inclined beams on both sides of the float; and two sets of photovoltaic modules, respectively fixedly installed on the purlins on both sides of the float, facing due east and due west respectively, with their tilt angle changing with the tilt angle of the inclined beams. This utility model can extend power generation time by staggering peak power generation through symmetrical east-west arrangement. Simultaneously, the change in azimuth angle reduces the shadow distance between the front and rear modules, reducing the size of the float and lowering relative costs. Furthermore, the maximum power can be tracked in real time by adjusting the angle, improving power generation efficiency and float area utilization.

This application provides a multimodal ultrasound intelligent identification method and precision diagnosis system for thyroid lesions. The method determines image data blocks with multimodal spatial registration and grayscale-contrast temporal alignment. Based on the grayscale ultrasound image sequence in the image data blocks, a first depth feature map of the target user's thyroid region is determined. A second depth feature map of the target user's thyroid region is determined based on the elastography map in the image data blocks. The hemodynamic spatial distribution characteristics of the parametric map in the image data blocks are captured to determine a third depth feature map of the target user's thyroid region. Cross-modal fusion is performed on the first, second, and third depth feature maps, and then the lesion segmentation mask, benign / malignant probability value, and risk classification of the target user's thyroid region are inferred and output in parallel. Using the scheme of this application, high-precision spatiotemporal alignment and cross-modal depth feature fusion of multimodal ultrasound images can be achieved to complete end-to-end intelligent identification of thyroid lesions.

The utility model relates to gas purification equipment technical field especially relates to a waste gas filter tank and gas washing equipment, and the device contains: fixedly arranged on the air inlet pipe of jar body, the air outlet end of air inlet pipe communicates with the inner chamber top of jar body for transmission toxic waste gas, the air outlet pipe is fixedly arranged on jar body, and the air inlet end of air outlet pipe communicates with the inner chamber top of jar body, the water supply pipe is fixedly arranged on jar body, and the output end of water supply pipe communicates with the inner chamber of jar body for transmission cooling water to the inner chamber of jar body, the shunt pipe is arranged in the air outlet end of air inlet pipe for the shunting of toxic waste gas that air inlet pipe output has carried out. The utility model discloses by adopting shunt pipe and shunt formation multiple small air current with toxic waste gas, to increase the contact area of toxic waste gas and cooling water, and then enhance the heat exchange efficiency of cooling water to toxic waste gas and the capture effect of solid phase particulate impurities, solved the low problem of cooling efficiency in the prior art.

This invention discloses a dual-sided synchronous spraying system for a guide trough, relating to the field of dust control technology in fully mechanized coal mining faces. It includes a guide trough body and an mounting plate; a sensing mechanism installed on one side of the top of the mounting plate; the sensing mechanism includes a storage box, a sealed door, two partition plates, multiple dust concentration sensors, multiple wind speed and direction monitors, and an intelligent control box. This invention achieves real-time monitoring and dynamic response to underground dust concentration and airflow conditions through the coordinated operation of the dust concentration sensors, wind speed and direction monitors, intelligent control box, electromagnetic control valve, and water pressure regulating valve. When the dust concentration exceeds the standard, the intelligent control box can drive the water pressure regulating valve to increase the spray intensity and the electromagnetic control valve to quickly open the spray branch. When the dust concentration reaches the standard, it controls the electromagnetic control valve to close and the water pressure regulating valve to revert, solving the problem of existing spray devices having a fixed and singular mode and being unable to adaptively adjust, thus avoiding the ineffective consumption of water and electricity.

This application belongs to the field of wind turbine technology, and particularly relates to a foldable horizontal blade device for wind turbines, including a rotating hinge base. The rotating hinge base includes a vertical shaft, on which multiple hinge seats are arranged. All the hinge seats are evenly arranged around the central axis of the vertical shaft. Blades are provided on the hinge seats and hinged to the hinge seats. A column is provided at the upper end of the rotating hinge base, and a fixed pulley group is provided at the upper end of the column. A rope winding and unwinding device is provided inside the column. A lifting rope is provided on the drive rod of the rope winding and unwinding device. The lifting rope passes through the fixed pulleys on the fixed pulley group and is connected to the blade. This application uses the connection between the blade and the rotating hinge base as the fulcrum, and uses the rope winding and unwinding device and the lifting rope to lift the blade, so that the blade can overcome its own weight and the blade length can be made longer. The angle between the blade and the horizontal plane can be changed, so as to enhance the wind energy capture capability.

This invention discloses a cross-modal data processing system for the safe operation of hydrogen refueling stations, including a raw data acquisition and processing module, a full-variable safety scanning module, a physical relationship coupling diagnosis module, an adaptive operating condition clustering module, a question-answer pair construction module, a hydrogen refueling station time-series command data acquisition module, and a fault type identification module. By constructing a three-layer semantic enhancement logic and model adaptation strategy, this invention effectively solves the technical problems in the prior art, such as the lack of supervision signals in the raw data of hydrogen refueling stations, the difficulty in identifying hidden faults under complex operating conditions, and the difficulty in adapting heterogeneous feature space models.

This invention relates to the field of blast furnace dust control technology, specifically to an integrated dust control device for environmentally friendly charging and injection of nickel-containing raw materials in a blast furnace. The device includes a base plate, inclined bridge tracks, and guide rails. Two inclined bridge tracks, oriented towards the blast furnace, are fixedly installed on the base plate. Limiting rails for guiding blast furnace charging cars are provided on the inclined bridge tracks. A guide rail is fixedly connected between the two inclined bridge tracks. A dust suppression mechanism for suppressing dust generated during the loading and unloading of blast furnace charging cars is slidably installed on the guide rails. The dust suppression mechanism of this invention moves synchronously up and down along the guide rails with the blast furnace charging car via a drive mechanism, always aligning with the charging car's loading port. It achieves dynamic tracking dust suppression throughout the entire process of loading, unloading, and tipping. The spray pattern can be adjusted by the cooperation of the support frame and the protrusions on the mounting rod, ensuring uninterrupted dust suppression and effectively preventing dust overflow.

This invention discloses a paragraph-level generated text detection method based on deep maximum mean difference (MDD) metric. The method includes: modeling generated text detection as a nonparametric two-sample test problem; designing a deep composite kernel to fuse shallow syntactic features and deep semantic features; training a deep kernel network with the goal of maximizing unbiased estimation of test power; in the testing phase, introducing a Wild Bootstrap mechanism to generate random weight sequences with autoregressive structures; performing a weighted quadratic transformation on the MMD kernel difference matrix; preserving paragraph sentence sequence correlation; accurately estimating the null distribution of the test statistic; and completing the detection by comparing the observed statistic with the test threshold. This method improves sensitivity to high-dimensional semantic differences, alleviates the problem of decreased test power caused by sequence correlation, and can efficiently distinguish between human and AI-generated text. It is suitable for generated text detection in scenarios such as news and academic writing.

This invention discloses a method and system for predicting the remaining life (RUL) of a battery. The method categorizes modal components into trend modes and fluctuation modes based on their frequency domain characteristics. For the trend mode prediction, a differentiable monotonicity penalty term is incorporated into the loss function, combined with a Transformer network, to ensure the prediction result decreases over time, thereby improving the physical plausibility and stability of the prediction curve. For the remaining fluctuation modes, a BiLSTM model combined with an attention mechanism is used for modeling. This combination effectively models and extracts high-frequency disturbance information. The prediction results of the trend mode and each fluctuation mode are then fused and reconstructed to restore the complete RUL prediction curve. This invention can accurately model battery degradation trends and high-frequency disturbances, and comprehensively and accurately model both to output prediction results, achieving reliable prediction of the remaining life of lithium-ion batteries.

The application discloses an e-commerce platform consumer behavior prediction method based on edge computing, comprising the following steps: collecting the behavior event flow of consumers in the e-commerce platform at the corresponding edge node of user access, and performing preprocessing; constructing a session state feature vector; constructing a local behavior transition graph for a micro-session segment and performing timing diagram coding on the local behavior transition graph; aligning and calculating the local behavior representation vector with a long-term preference anchor point set; generating a state digest based on an edge side prediction network and a collaborative routing indication value, and sending the state digest to a cloud node; the cloud node performs collaborative prediction in combination with the historical behavior characteristics stored in the cloud; collecting real behavior feedback of the edge side prediction result or the cloud prediction result to obtain difficult samples, generate a parameter correction amount or an anchor point correction amount, and write back to the edge node. The edge collaborative prediction realizes accurate identification of consumer behavior.

This invention discloses a method and system for target detection in hyperspectral complex scenes, belonging to the field of image processing technology. It utilizes an adaptive threshold to select high-confidence results as pseudo-labels from the network prediction results of weakly enhanced samples. Based on the difference between the network prediction results of strongly enhanced samples and the pseudo-labels, a consistency regularization loss is constructed, and the network parameters are updated through backpropagation. All samples are ranked by difficulty according to the confidence level of the network prediction output, and the proportion of samples participating in the next training stage is progressively increased until all data in the training dataset is used to obtain a trained dual-stream feature extraction network. Each pixel in the hyperspectral image to be detected is input into the trained dual-stream feature extraction network, and the network outputs the confidence level of each pixel belonging to the prior of the target, generating a detection map. This method can significantly improve the accuracy of target detection.

This utility model relates to the field of detector technology, specifically to a glass fiber filament breakage detector, comprising: a base; a steering mechanism including a universal joint, a locking part, and a steering part, wherein the universal joint is fixedly connected to the base, the steering part is rotatably connected to the universal joint, and the locking part is located on one side of the steering part and rotatably connected to the steering part; a support mechanism fixedly connected to the universal joint; and a pneumatic detection mechanism locked to the support mechanism. This utility model solves the problem of detection accuracy attenuation caused by the collection process, ensuring the continuous accuracy of filament breakage detection.

The present application relates to the field of data processing, and particularly relates to a sewer network pollutant tracing method, device and system. The present application fuses multi-source pollution data to construct an uncertainty quantification framework, which is used to determine adaptive pollution source tracing positioning. The experience probability of the relevant historical pollution path is obtained by mining the historical pollution events, and the suspected pollution source point and the suspected pollution path are obtained. Then, the asynchronous strength and the blind area gain of each suspected pollution path are accurately quantified. In the final adaptive tracing positioning process, the experience probability, the asynchronous strength and the blind area gain of the suspected pollution path are fused to obtain an adaptive search radius, and the tracing positioning is performed within the adaptive search radius. The final adaptive search radius obtained by the present application can differentiate the scheduling of search resources, reduce the waste of tracing resources and reduce the tracing cost.

The application discloses an electrochemical noise corrosion state identification and early warning method and system, and belongs to the technical field of corrosion monitoring and early warning. The method collects the potential noise signal and the current noise signal of the target component through an electrochemical workstation and calculates the noise resistance, adopts db4 wavelet for multi-layer wavelet decomposition and threshold denoising processing, extracts the kurtosis feature, the skewness feature and the wavelet coefficient feature, and constructs a corrosion state identification and early warning data set after feature screening; a fusion model including an LSTM time sequence feature extraction layer and an Attention attention weight distribution layer is constructed for training, and a corrosion state identification model is obtained; the real-time collected and processed feature parameters are input into the model for identification, the proportion of a specific corrosion state in a preset time period is counted, and a blue, orange or red graded early warning is triggered. The application realizes high-precision identification and accurate graded early warning of the corrosion state, the identification accuracy is above 85%, and reliable technical support is provided for industrial component corrosion protection.

This application relates to a method and apparatus for segmenting road target instances in remote sensing images. The method includes: acquiring remote sensing images of the target to be detected by a remote sensing satellite; preprocessing the remote sensing images through an input module; extracting features from the preprocessed image using a backbone network; fusing the extracted multi-resolution features using a dual-stream feature pyramid in the neck region; dividing the head region into two branches: a detection head and a prototype; and performing category prediction and instance segmentation refinement tasks on the image entering the head region; outputting the high-resolution remote sensing image road target instance segmentation result; using a DMH-YOLO model to perform instance segmentation on the target remote sensing image to determine the target objects in the target remote sensing image; and using the Darknet-53 architecture framework in the backbone network and incorporating a C2f module to implement a residual learning mechanism. This invention can accurately identify and segment small target objects.