55results about How to "Capture comprehensive" patented technology

The invention discloses an internet information intelligent monitoring method and system, a medium and a product, and relates to the technical field of data processing. The method comprises the following steps: firstly, acquiring internet information data containing text, image and audio contents, extracting features of the three contents respectively, and constructing a semantic map; calculating semantic similarity among different map nodes, and establishing a cross-modal mapping relation to form a mapping set; node attribute characteristics are extracted for the mapping relation to calculate consistency, and mapping lower than a threshold value is marked as a conflict relation; counting the conflict node proportion of each map to obtain a modal conflict degree, calculating the similarity with the reference map to obtain a modal matching degree, and calculating the modal credibility according to the modal matching degree; normalizing the modal credibility to obtain a weight coefficient, and performing weighted fusion on the map structure features to obtain a fusion feature vector; and finally, matching the fusion feature vector with a monitoring rule base to obtain a monitoring result. By implementing the technical scheme provided by the invention, the monitoring accuracy of Internet information can be improved.

The invention discloses a large-range landslide mass displacement early warning method based on GNSS and radar data fusion, and the method comprises the steps: obtaining GNSS monitoring sequence data in a landslide region based on the position information of a reference station and a GNSS monitoring station in a GNSS monitoring system, and further obtaining the displacement related information of the landslide region; collecting data through a satellite InSAR, and generating deformation field data of a landslide area by comparing radar images at different time points; performing space-time alignment on the obtained data, and performing two types of data fusion S by adopting an adaptive Kalman filtering method after alignment; decomposing time series data of the fused landslide mass displacement by using wavelet transform, and extracting multi-scale features of a time domain and a frequency domain; the proposed features are combined with historical data of landslide mass displacement, a landslide mass displacement trend prediction model using an LSTM method added with a memory decline factor is input, and future trend prediction of landslide displacement is carried out; setting a multi-level threshold value, judging the change trend of the landslide mass displacement according to the displacement rate, the acceleration and the prediction result, and generating a multi-level early warning signal.

The invention discloses an electromechanical equipment fault cross-domain diagnosis method and system based on minimum category confusion, and the method comprises the steps: selecting an external public data set or labeled historical data as a source domain data set, and then collecting the real-time operation data of electromechanical equipment to be diagnosed, taking the real-time operation data of the electromechanical equipment to be diagnosed as a target domain data set; constructing a fault diagnosis model; performing joint adversarial training on the fault diagnosis model by using the preprocessed source domain data set and target domain data set, and performing lightweight processing and parameter solidification after training is completed to obtain a final fault diagnosis model; and deploying the final fault diagnosis model to an edge computing equipment end, and diagnosing the preprocessed real-time fault data of the electromechanical equipment by using the final fault diagnosis model to obtain a diagnosis result. According to the method, the multi-scale feature extractor is adopted to perform multi-scale feature extraction and fusion, so that the feature extraction capability is improved, and missing of key features is avoided.

This invention provides a method for detecting sensor anomalies in nuclear power plants. The method includes: constructing a principal component analysis model, calculating the Q-statistic, and determining a first threshold; training a Bayesian long short-term memory neural network model and calibrating its prediction interval; calculating a second and third threshold based on the predicted value and uncertainty output by the calibrated model; inputting test data into the two models to obtain the Q-statistic, predicted value, and uncertainty, and comparing them with the corresponding thresholds to obtain three judgment results; and making a final judgment on the three results using a logical judgment table. This invention integrates spatiotemporal characteristics and uncertainty assessment, achieving high-precision and high-reliability sensor anomaly detection with only normal operating data, exhibiting high reliability and robustness.

The application provides a scene classification method and device for automatic driving, which comprises the following steps: acquiring static element and dynamic target data of a current scene of an automatic driving vehicle; extracting combined static element features, matching a preset static scene type library, and determining an initial scene category; generating a motion trajectory based on different time positions of dynamic target data, combining static element data to construct a time sequence correlation graph, and presenting time sequence correlation between scene elements and dynamic targets; identifying relative motion states through dynamic target data, combining speed and direction, and determining interaction relationships between dynamic targets; and finally determining a target scene category of the automatic driving vehicle according to the time sequence correlation graph and the interaction relationships for each subcategory of the initial scene category. The application effectively solves the problem of insufficient analysis of dynamic element interaction in the prior art, meets the high-precision requirement of automatic driving for scene judgment, and can more accurately identify potential risks in the scene.

This invention discloses an abnormal driving recognition method based on pedal data and facial expression multimodal data. The process is as follows: Step 1: Collect vehicle driving data and facial video data, and label real facial abnormal expressions to obtain original multimodal data composed of multiple original multimodal samples; Step 2: Train an original abnormal driving model based on the original multimodal dataset; Step 3: Perform noise diffusion on each original multimodal sample to obtain diffused multimodal samples, and select effective multimodal samples from the diffused multimodal samples to construct a multimodal dataset; Step 4: Process the multimodal dataset using the original abnormal driving model, and select prediction-consistent multimodal samples to construct a prediction-consistent multimodal dataset; Step 5: Train the original abnormal driving model again using the prediction-consistent multimodal data to obtain an abnormal driving recognition model; Step 6: Use the abnormal driving recognition model to predict the abnormal driving prediction result.

The utility model discloses a concrete manufacturing dust treatment device which comprises a dust collection main box and square-opening right-angle air inlet pipes, and the square-opening right-angle air inlet pipes are connected to the two sides of the dust collection main box. The front end and the rear end of the dust collection main box are connected with air collection covers, one end of each air collection cover is connected with an air collection pipe, one end of each air collection pipe is connected with a front air box, the bottom end of each front air box is connected with a centrifugal fan, and a spraying dust falling mechanism is arranged at the bottom end of the dust collection main box. The centrifugal fan is driven, the air collecting pipe, the air collecting cover, the dust collecting main box and the square-opening right-angle air inlet pipe are matched so that dust in the concrete manufacturing process can be sucked into the dust collecting main box, dust removal is achieved, the filter screen is installed in the air collecting cover, and the filter screen is covered with the nylon fiber synthetic resin film. And dust is prevented from being sucked into the centrifugal fan.

The present application relates to the technical field of water environment monitoring and water conservancy informatization, and particularly relates to a reservoir sediment-pollutant migration coupling numerical simulation method and system. The method comprises the following steps: obtaining flow velocity profile data, suspended matter concentration data and spectral data of a monitoring area, and performing dynamic scattering compensation; obtaining entity water samples at different depth layers, and calibrating the entity water samples; constructing a pollutant migration model; inputting real-time obtained reservoir scheduling parameters as dynamic boundary conditions into the pollutant migration model, and introducing a preset adsorption lag coefficient to nonlinearly correct a pollutant release rate in the pollutant migration model; using the nonlinearly corrected pollutant migration model to calculate a composite migration flux of sediment and pollutants, and generating a sediment resuspension risk thermal map and water quality prediction results of the monitoring area. The present application helps to improve the prediction accuracy of heavy metal migration flux in a water source and the response speed of water quality risk early warning.

This invention relates to the field of time series forecasting technology and discloses a stock index forecasting method based on 1DCNN-ResBiLSTM, comprising the following steps: S1, data acquisition and sample construction; S2, data preprocessing; S3, local feature extraction; S4, long-term dependency learning; S5, residual feature fusion: extracting the feature vector of the last time step output by the 1DCNN module in S3, linearly mapping it to the same dimension as the output feature of S4 through a fully connected layer, and then adding it element-wise with the output feature of S4 to achieve residual feature fusion; S6, prediction result output; S7, model training and optimization. By adopting a technical solution that combines the 1DCNN module with time-dimensional sliding convolution and ReLU nonlinear activation, the method accurately captures the local fluctuation features and short-term trend information of stock time series, solving the problems of insufficient characterization of short-term stock price changes and inability to effectively mine local time series patterns.

This application provides a multimodal model-driven video recognition method and apparatus, relating to the field of video content analysis technology. The method includes: acquiring a target short video and performing frame segmentation, audio separation, and text content extraction to obtain video frame sequences, audio data, and text data; extracting visual features, speech features, and text features based on the video frame sequences, audio data, and text data, respectively; inputting these multimodal features into a multimodal fusion network based on an attention mechanism, and performing weighted fusion by calculating cross-modal interaction weights between the features to generate a fused feature representation; inputting the fused feature representation into a business order recognition classifier to obtain the business order recognition result of the target short video. The technical solution of this application can improve the accuracy of short video business order recognition and is applicable to scenarios such as short video platform content review, advertising monitoring, and business data analysis.

The application relates to an ASM-oriented radio frequency fingerprint identification method and system, and relates to the field of radio frequency fingerprint identification, wherein the method comprises the following steps: acquiring an ASM signal emitted by a ship and extracting a multivariate time sequence of the ASM signal; extracting time domain local features and frequency domain structural features of the multivariate time sequence, splicing the time domain local features and the frequency domain structural features to obtain first spliced features, fusing the first spliced features to obtain fused features; performing feature enhancement and spatial information fusion on the fused features to obtain multi-scale fused features; extracting global statistical information and key detail information of the multi-scale fused features, splicing the global statistical information and the key detail information to obtain second spliced features; and performing identification on the second spliced features to output a radio frequency fingerprint identification result. The application can effectively identify a radio frequency fingerprint, and further judge whether an ASM signal is tampered with.

This application discloses an ultrasound image processing method, apparatus, electronic device, and storage medium. The method includes: acquiring multiple types of data signals during ultrasound imaging; wherein the multiple types of data signals include at least two of the following: ultrasound echo signal after beamforming, initial wave image before digital scan conversion, and ultrasound image after digital scan conversion; performing no-load detection on each of the multiple types of data signals to obtain corresponding no-load detection results; and combining multiple no-load detection results to determine the final no-load detection result of the ultrasound image. This application performs multiple no-load detections on multiple types of data signals in multiple stages of ultrasound imaging, and combines the results of multiple no-load detections to obtain the final no-load detection result of the ultrasound image, improving the accuracy of no-load image identification, avoiding subsequent functions from making incorrect diagnoses or measurements of the no-load image, and ensuring the reliability and accuracy of ultrasound diagnostic and measurement functions.

The application is specifically a mediation record legal fact keyword extraction system and method based on AI, which comprises the following parts: a data collection module; an analysis and processing module: the video data corresponding to the high-frequency words are analyzed to obtain visual values, and the voice data are analyzed to obtain phonetic values; the visual values and phonetic values corresponding to the high-frequency words are comprehensively analyzed to obtain evaluation values; a correction module.In the application, the video, voice and electronic record data in the mediation record are comprehensively analyzed, which can more comprehensively capture case information; through this multi-modal data fusion mode, information omission and inaccuracy caused by a single data source are avoided, thereby improving the accuracy and comprehensiveness of keyword extraction, and helping legal professionals better understand the overall situation and dispute focus of the case.

The invention discloses a photovoltaic power generation prediction method and device, relates to the technical field of intelligent power generation, and solves the problems that data errors exist in data acquisition and power generation processes, and a certain deviation exists between a predicted value and an actual value due to the fact that an output result is directly used as the predicted value. The method comprises the following steps: acquiring historical power generation amount, real-time weather data, historical weather data and corresponding historical total power generation amount of a plurality of photovoltaic panels; dividing the photovoltaic power generation area into a plurality of power generation areas based on the historical power generation amount of the plurality of photovoltaic panels; analyzing a power generation threshold value based on the historical weather data and the corresponding historical total power generation amount; predicting the generating capacity of the plurality of areas based on the real-time weather data and the generating threshold to obtain predicted generating capacity; integrating the predicted generating capacity of the plurality of areas to obtain photovoltaic predicted generating capacity; the generating capacity is accurately predicted, and prediction errors caused by uncertain factors are reduced.

The present application relates to a Laos language unsupervised phoneme segmentation method based on multi-feature interaction fusion, and belongs to the technical field of natural language processing. The present application comprises the following steps: collecting and processing Laos language audio data sets; extracting self-supervised features, spectral features and pitch features; fusing the self-supervised features, spectral features and pitch features to train a phoneme segmentation model; based on the principle of probability contrast loss, distinguishing adjacent frames and random interference frames to realize optimization of the phoneme segmentation model; in the inference stage, inputting the phoneme segmentation model output vector into a peak detection algorithm to generate the final phoneme boundary. The present application can better interact and fuse the advantages of different features, overcome the limitations of the direct addition method, and thus achieve better results in overall performance.

The application provides a text topic extraction method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining a Bert word vector and a Bert sentence vector of a to-be-processed text; taking the Bert word vector and the Bert sentence vector as nodes to construct a topological graph model; optimizing the nodes of the topological graph model by using a graph autoencoder with an attention mechanism to obtain multiple features of each node; and analyzing the multiple features by using a clustering algorithm to extract topic information of the to-be-processed text. The method can improve the accuracy of text topic extraction.

The invention relates to the technical field of nondestructive testing, and provides an electric power fitting crack nondestructive testing method based on bimodal and related equipment. Scanning the electric power fitting through an ultrasonic module to obtain an ultrasonic signal, and obtaining an infrared signal of the electric power fitting through an infrared module; performing space-time alignment and integration based on deviation parameters between the ultrasonic signals and the infrared signals to generate a bimodal data cube; the data cube is decoupled through a convolution solution network obtained through pre-training, and an ultrasonic characteristic spectrum and an infrared characteristic spectrum are generated; performing crack probability prediction based on topological characteristics of the characteristic spectrum, and generating a crack probability heat map and a topological structure map; and performing multi-scale verification on the crack probability heat map and the topological structure diagram to generate crack parameters. Crack parameters can be accurately obtained by integrating the internal structure and surface temperature information, crack prediction and multi-scale verification, and the accuracy and reliability of crack detection are effectively improved.

The present application relates to a kind of based on halftone noise and artifact enhancement document image tampering positioning method.The method constructs halftone noise perception stream, utilizes denoising residual mechanism to extract the micro halftone statistical law damaged by tampering;Frequency domain perception stream is constructed, decouples luminance and chroma in YCrCb color space, removes low-frequency background by combining adaptive frequency domain enhancement and adaptive high-pass filter, and accurately captures high-frequency quantization artifact;Covariance-based feature fusion mechanism is constructed, spatial domain, frequency domain and halftone noise domain features are cascaded and fused, and the conflict of multi-modal heterogeneity is relieved to accurately locate tampering area.The method is for the authenticity verification scene of document image disturbed by complex post-processing such as JPEG compression, scaling, and good results are obtained in positioning accuracy and robustness.

The utility model provides a self-adjusting wind turbine generator set pitch bearing crack monitoring device, which comprises an outer bearing, an inner bearing, bearing holes and an external control device, the inner bearing rotates in the outer bearing, the bearing holes are arranged in the inner bearing and the bearing holes in an annular array, and the wind turbine generator set pitch bearing crack monitoring device is used for monitoring the pitch bearing crack of the wind turbine generator set. The position of the detector can be dynamically adjusted according to the diameter and the distance of the bearing hole through sliding fit of the driving plate and the telescopic plate of the adjusting mechanism, when the size of the bearing hole changes, the external control device can drive the driving plate to slide in the first fixing plate, and meanwhile the telescopic plate stretches out and draws back synchronously, so that the detector is accurately aligned with the bearing holes with different hole diameters. Besides, the size of the detection window is matched with the distance between the bearing holes, a second fixing plate can be elastically extruded through a buffer pad in cooperation with an adjusting structure of a threaded rod and a bolt, it is ensured that the upper surfaces and the lower surfaces of the bearings of different sizes are stably clamped, and the problem that a traditional device cannot be matched with the bearing holes of multiple sizes due to a fixing structure is solved.

The present application relates to the technical field of in vitro diagnosis, and discloses a homogeneous immunoassay method, kit and application of soluble thrombomodulin and D-dimer. In further detail, the present application can realize one-step quantitative analysis of soluble thrombomodulin and D-dimer in the same reaction system based on the developed homogeneous immunoassay method. The detection method or kit can specifically recognize target antigens, and does not affect each other in the liquid phase system, and has excellent anti-interference performance, and is not interfered by common endogenous substances such as fibrinogen, fibrinogen degradation fragment D / E, coagulation factor X and active form Xa. Compared with the existing conventional single index independent detection or heterogeneous detection method, the present application has the advantages of no need for washing and separation, simple operation, fast detection speed, high sensitivity and the like. Through single detection, clinical information reflecting the double pathological mechanisms of 'endothelial injury' and 'coagulation activation' can be quickly obtained, the detection timeliness is significantly improved, and a new technical means with significant application prospect is provided for the clinical needs such as bedside rapid screening and efficacy monitoring of venous thrombosis and related basic scientific research.

The invention discloses a water conservancy project intelligent monitoring and early warning system based on the Internet of Things and AI. According to the system, firstly, the problem that a traditional monitoring task is blind and disordered is solved by obtaining a potential monitoring demand index of the monitoring task, marking a target task in combination with a threshold value and generating a priority list, resource configuration is optimized, and deployment scientificity and pertinence are improved; secondly, a water conservancy target topology network model is constructed, monitoring links are formed, early warning links are screened, adaptive measures are adopted according to the number grading of the early warning links, full-chain closed-loop management is achieved, and early warning accuracy and response effectiveness are improved; and 3, when an early warning link appears, generating an update priority list, dynamically associating subsequent tasks, quickly responding to chain risks, avoiding omission or lag, enhancing the flexibility and coherence of the system to cope with complex risks, and ensuring the safe operation of the water conservancy project.

The film and television script extraction method and device, the storage medium and the computer device provided by the application, after the audio and video file of the film and television is split into a video file and an audio file, the video file is subjected to feature recognition to obtain a subtitle text, a speaker face information and a video understanding text; and the audio file is subjected to speech understanding to obtain a speech transcription text and a speech understanding text; wherein the speech transcription text can be corrected into a standard transcription text with higher accuracy through the subtitle text. Therefore, based on the speaker face information, the standard transcription text and the dialogue segment of each speaker in the audio and video file are aligned, the speaker information with accurate segmentation and coherent semantics can be obtained, then the script information is constructed by combining the character profile text generated by the profile analysis of the speaker based on the video, the speech understanding text and the speaker information, the script information can cover the related feature description of the character on the basis of the dialogue content, thereby enriching the content and depth of the script.

The application discloses an abnormal frequency response detection system and method of an electret microphone. The system can automatically collect audio data, analyze frequency response curves, extract key features, and match them with features in a pre-stored reference frequency response feature library, ultimately directly outputting specific fault types, greatly reducing the dependence on human professional skills. Moreover, the matching analysis based on the pre-set algorithm and the reference feature library completely eliminates the subjective judgment deviation, so that the same product can obtain highly consistent and repeatable diagnostic conclusions at different times, by different operators or on different equipment, improving the reliability of quality control. At the same time, the three-dimensional feature parameters of frequency response flatness, frequency band average sensitivity and resonance characteristics are introduced for joint analysis, so as to realize high-precision differentiation and accurate positioning of different fault types (such as "high-frequency blockage" and "diaphragm resonance"), and the diagnostic conclusion directly guides production and maintenance.

This invention relates to the field of marine multimodal data processing technology, specifically to a method and system for detecting abnormal changes in marine biological communities based on multimodal features. To address the low accuracy of existing biological community anomaly detection technologies, the method first acquires graph modal data, acoustic modal data, and biochemical modal data from observation points, extracts modal features from each, and then fuses them to obtain an ecological embedding representation. Next, each observation point is treated as a node and connected to form a marine spatial graph structure. Graph convolution processing is used to enhance spatial features, and temporal extraction is combined to obtain the spatiotemporal features of the nodes. Finally, based on the spatiotemporal features of the nodes, outliers and their levels are obtained. A heatmap is then constructed and fused with geographic mapping to form a regional anomaly heatmap. The anomaly levels of the observation points are then marked to obtain the biological community anomaly detection results. This invention, when applied to the field of marine biological community anomaly detection, can improve detection accuracy and efficiency.

The application discloses an intelligent unmanned aerial vehicle power system anomaly detection method, which comprises unmanned aerial vehicle data acquisition, data preliminary processing, pseudo label generation, anomaly detection model construction and power system anomaly detection. The original data is obtained through data acquisition; the improved graph clustering model is used to generate pseudo labels to expand the training data; through semi-supervised learning, the unlabeled data is fully utilized, and the time sequence, causality and physical constraints are fused, so that the abnormal characteristics in the system dynamic evolution can be more comprehensively captured, and the false positives and false negatives can be reduced; the double-branch deep learning model is used as the anomaly detection model, the short-time local features and long-time global trajectory are processed in parallel, the multi-scale evolution mode of the anomaly can be comprehensively captured, the limitation of single perspective is avoided, the unmanned aerial vehicle power system physical priori knowledge is introduced to enhance the feature fusion, and the output conforms to the system dynamics law, so that the detection accuracy is improved.

The application relates to a nugget diameter prediction method and device for resistance spot welding, computer equipment and a storage medium. The foregoing method comprises the following steps: acquiring welding process parameters of a to-be-detected welding spot, the welding process parameters comprising current data and resistance data at multiple sampling time points; performing sliding window processing on the welding process parameters; performing feature extraction on the data of each window obtained through the sliding window processing by using a convolutional neural network to obtain first feature data; performing time sequence feature extraction on the welding process parameters by using a gated recurrent unit network to obtain second feature data; performing feature fusion on the first feature data and the second feature data; and inputting the fused feature data into a pre-trained nugget diameter prediction model to obtain a nugget diameter prediction result of the to-be-detected welding spot. The method can improve the prediction efficiency of the nugget diameter prediction result and save costs without destructive detection.

The invention discloses a power transmission line icing galloping prediction method fusing physical and attention mechanisms, belongs to the field of power transmission line monitoring, and realizes efficient icing galloping prediction under complex weather and physical disturbance conditions through multi-level data processing, physical model guidance and time sequence feature fusion. The method comprises the following steps: constructing a galloping strength label by utilizing collected transmission line lead and meteorological data and combining an interpolation completion technology based on a tension-sag equation, and generating a preliminary prediction value through physical perception resampling; the high-dimensional time sequence features extracted by the feature coding module are fused with a time sequence network of a multi-scale sensing mechanism and physical guidance, so that the prediction result of the galloping intensity is further optimized; through introduction of physical constraint and an error tolerance mechanism, the collaborative constraint of a loss function ensures that a prediction result accords with a physical rule while time sequence accuracy is ensured; the corrected time sequence type icing galloping intensity prediction value provides accurate early warning information for safe operation of the power transmission line.

This invention discloses a method for human-vehicle association analysis based on spatiotemporal feature fusion, comprising the following steps: For all checkpoints in the target road network, two different devices are used to collect vehicle license plate numbers and mobile phone IMSI codes respectively; the collected data is preprocessed, and license plate numbers and IMSI codes collected from the same checkpoint within the same time period are grouped into a sample; a graph is constructed for each sample, with license plate numbers and IMSI codes as nodes; a graph attention network is used to capture the spatial association features of each license plate number or IMSI code within each time slice, and a temporal convolutional long short-term memory network is used to capture the temporal association features of the license plate number and IMSI code under different time slices, generating the final vector representations of the license plate number and IMSI code; finally, the similarity between all license plate number and IMSI code vectors is calculated, and the IMSI code or license plate number with the highest similarity is selected as the association result. This invention can effectively capture the spatiotemporal association between license plate numbers and IMSI codes, improving the accuracy of human-vehicle association analysis.

The present application relates to physiological micro-vibration signal processing technical field, specifically disclose physiological micro-vibration signal multi-dimensional time sequence feature extraction and reconstruction method and system, the method includes preprocessing physiological micro-vibration original mixed signal, obtain pretreatment signal;Based on the improved adaptive variational mode decomposition algorithm and blind source separation algorithm to the pretreatment signal processing, obtain time-frequency domain feature vector;The method of combining bidirectional long short term memory network and encoder is used to the pretreatment signal deep time sequence feature extraction and dimension reduction, obtain dimension reduction feature vector;Time-frequency domain feature vector and dimension reduction feature vector are fused, effective fusion feature vector is screened based on correlation threshold value, physiological micro-vibration signal is reconstructed based on the effective fusion feature vector screened out. Through the improved VMD-BSS algorithm, the adaptive determination of the number of modes and the penalty factor is realized, the dependence on prior knowledge is reduced, and the processing speed of signal decomposition is significantly improved.

The utility model discloses a high accuracy heart rate detector for sports training belongs to sports training heart rate detection technical field, this high accuracy heart rate detector for sports training, including heart rate detection mechanism, heart rate detection mechanism includes first photoelectric sensor and second photoelectric sensor, installs transmission frame, and the transmission frame includes transmission frame, and the transmission frame includes upper side board, and the one end of upper side board is hinged with lower side board, and the bottom fixed mounting of lower side board has base, the one end of transmission frame is installed with hinged component, and hinged component is connected with the one end transmission of transmission frame, and the other end transmission of transmission frame is installed with supplementary locking assembly, and supplementary locking assembly is locked with transmission frame spacing, adopts first photoelectric sensor and second photoelectric sensor, and double sensor detects from the wrist upper and lower different position, can more comprehensive, accurately capture pulse signal, effectively improves the precision of heart rate detection, and provides reliable data support for sports training.