400 results about "Quantized neural networks" patented technology

The invention discloses an intelligent duck shed monitoring method and system based on image recognition, and the method comprises the steps: obtaining environment monitoring data from an environment monitoring device, obtaining original image data of a breeding region from an image collection device, and processing the original image data through an image enhancement algorithm, thereby obtaining a preprocessed image data set; according to weight parameters output by the duck group behavior and environment correlation model, performing weighted fusion on the environment monitoring data and the preprocessed image data set to generate a multi-modal feature data set; if the group behavior index in the multi-modal feature data set deviates from a preset threshold range, extracting a spatial-temporal feature map through a convolutional neural network, and generating individual behavior trajectory data in combination with a multi-target tracking algorithm; and inputting the individual behavior trajectory data into the lightweight neural network model, and outputting a health state classification result. According to the intelligent duck breeding system, accurate monitoring and intelligent adjustment of the duck breeding environment are achieved, and the breeding efficiency and the duck group health level are effectively improved.

The invention discloses a mobile terminal streetscape image real-time segmentation method based on a lightweight neural network, and relates to the technical field of image segmentation. The method comprises the following steps: firstly, carrying out 320 * 320 adjustment, Z-score standardization, adaptive histogram equalization and 3 * 3 Gaussian filtering preprocessing on an input streetscape image; then, an improved MobileNetV3 backbone network is used, and a five-scale feature map is output in combination with DropBlock regularization through eight feature extraction stages including depth separable convolution and an SE attention module; multi-scale features are fused through a U-shaped structure, and a fusion feature map is generated through up-sampling, element-by-element addition of dimension reduction low-layer features and an attention gating module; and during reasoning, outputting a segmentation mask by using a convolutional layer, Softmax and a conditional random field, and finally performing knowledge distillation, weight pruning, 8-bit quantization and TensorRT optimization. According to the invention, high-precision real-time street view segmentation is realized, the robustness is high, and the method is suitable for different devices and scenes.

The invention discloses an intelligent security and protection monitoring system and method fusing multi-modal data analysis, and relates to the field of intelligent monitoring, and the method comprises the following steps: collecting a video stream, infrared thermal imaging data and an environment audio signal in real time through a heterogeneous sensor array; operating a lightweight neural network at an edge computing node to carry out space-time alignment and feature level fusion on the multi-source data, and generating an enhanced environment sensing matrix; a dual-channel anomaly detection mechanism is adopted, a first channel identifies short-term sudden abnormal events through a dynamic threshold self-adaptive model, and a second channel early warns potential risk behaviors through a time sequence prediction model; according to the system and the method, a dynamic privacy protection mechanism is adopted, behavior analysis is completed on the premise of ensuring biological feature safety, and transformation from passive monitoring to active early warning is realized in combination with a space-time prediction model; the problem of multi-view target tracking is solved through a distributed consensus algorithm, and the monitoring reliability in a complex scene is remarkably improved.

The invention discloses an intelligent electric equipment monitoring and optimizing method, and relates to the technical field of electric power, and the method comprises the steps: collecting high-dimensional voltage-current time sequence data and transient event marks, calculating topological invariant features, inputting the topological invariant features to a lightweight neural network model, and recognizing the features of all electric equipment; performing abnormal attribution and anti-fact energy efficiency prediction by using causal reasoning and dynamic regularization regression based on the identified characteristics of each electric device, and constructing a multi-objective optimization function through the abnormal attribution and anti-fact energy efficiency prediction; and based on the multi-objective optimization function, generating an equipment operation scheduling strategy through a deep reinforcement learning agent, based on the operation scheduling strategy, sending a control instruction to the electric equipment, and collecting an operation result feedback in real time for optimization and updating. According to the method, through fusion of topological features, causal reasoning and safety reinforcement learning, the precision, robustness and safety of monitoring and optimization of the intelligent electric equipment are improved.

The invention provides a distributed collaboration-based computing network intelligent agent system and a resource dynamic scheduling method thereof, and belongs to the technical field of computing network integration. The system comprises an edge agent used for sensing local computing power, network bandwidth and task load in real time, predicting task demand fluctuation by using a lightweight neural network, adjusting resource allocation weight in real time in combination with network topology change, and executing a preliminary task scheduling decision; the regional collaborative agent is used for aggregating multiple edge node states based on federated learning, generating a cross-node resource scheduling strategy, verifying the credibility of a computing power transaction smart contract and determining a cross-domain resource allocation scheme; and the cloud management agent is used for constructing a global resource portrait model according to the information provided by the edge agent and the regional collaborative agent, performing long-term strategy optimization, issuing global strategy information, and constructing and updating a computing power transaction smart contract based on a preset computing power transaction smart contract template. According to the invention, multi-level refined scheduling of computing network resources is realized.

The invention relates to the technical field of neural network model construction, in particular to a lightweight neural network model construction method, which comprises the following steps: based on a target task data set, in a lightweight basic operator library comprising a depth separable convolution, an inverted residual structure and an attention mechanism module, constructing a lightweight neural network model; and searching and jointly optimizing network structure parameters and weight parameters through the differentiable neural architecture to obtain an initial lightweight network model. And deploying the initial model in a target hardware simulation environment, and generating a Pareto optimal model cluster through structural re-parameterization and hardware-aware progressive channel pruning iterative optimization by taking model precision, reasoning delay and memory occupancy as collaborative optimization targets. And selecting a reference student model from the clusters according to deployment constraints, constructing a distillation framework taking the initial model as a teacher model, and performing fine adjustment by adopting a mixed strategy fusing multi-dimensional distillation loss to obtain a final model. The model gives consideration to precision and efficiency, and the detection efficiency and the quality control level are improved.

The invention belongs to the technical field of computer vision, and relates to a lightweight image feature point detection and matching method and related equipment thereof. The method comprises the following steps: performing feature extraction on a target image through a lightweight SuperPoint neural network model to generate feature information; performing super-pixel segmentation processing on the image to generate a super-pixel segmentation result; performing initial matching on the feature information through a lightweight SuperGlue neural network model to generate an initial matching pair; performing outlier filtering on the initial matching pair based on a superpixel segmentation result, and determining an optimized target matching pair; performing geometric consistency optimization on the optimized target matching pair by adopting an adaptive exponential moving average algorithm to generate a final matching result; the lightweight SuperPoint neural network model is generated by carrying out hierarchical progressive pruning processing on an original SuperPoint neural network model; the lightweight SuperGlue neural network model is generated by carrying out attention head pruning processing and graph network layer pruning processing on an original SuperGlue neural network model.

The invention discloses a layered energy management method of a fuel cell hybrid tramcar with self-adaptive working condition perception. In the recognition layer, a sliding window mechanism is adopted to extract time domain and frequency domain features of load conditions, feature data are clustered based on a spectral clustering algorithm driven by a deep auto-encoder, a data set with category labels is obtained, and a deep dynamic learning vector quantization neural network classifier is trained; in the strategy layer, a double-delay depth deterministic strategy gradient reinforcement learning algorithm is adopted, a reward function is constructed, and lithium battery SOC fluctuation penalty term limit parameters in the reward function are adaptively adjusted according to the real-time load working condition category output by the recognition layer; training the reinforcement learning agent to obtain an optimal power distribution scheme between the multi-stack fuel cell power generation system and the lithium battery; and according to the performance degradation degrees of different fuel cell stacks, a distributed cooperative control strategy considering performance difference is adopted to distribute the output power of each stack, so that the coordinated control of the running state of the multi-stack fuel cell power generation system is realized.

The invention discloses a circuit breaker image edge detection method fusing spatial constraint fuzzy clustering and lightweight network optimization, and the method carries out the local adaptive threshold calculation through combining spatial constraint FCM and Otsu algorithms, and optimizes the edge detection process of a Canny operator. According to the method, fuzzy classification is carried out on a circuit breaker image by adopting spatial constraint FCM to obtain a strong marginal probability graph; the circuit breaker image is subjected to block processing through local adaptive threshold calculation, a global threshold is generated through integration, and then the global threshold is input into a Canny operator for accurate edge extraction. In order to further improve the detection effect, a lightweight neural network PiDiNet is used to correct a Canny output image. According to the method, the edge detection precision of the circuit breaker image can be effectively improved, and the method is suitable for edge extraction tasks in high-noise and complex background environments.

The invention relates to the technical field of industrial Internet of Things and wireless sensor networks, in particular to a large-scale industrial fault diagnosis system and method based on Bluetooth MESH, and efficient monitoring and fault early warning of industrial equipment are realized through low-power-consumption sensor nodes, edge computing, a Bluetooth Mesh network and a clustered Mesh tree architecture. The system comprises a client, a server, a gateway, a cluster head and a cluster node, the cluster node comprises a sensor node and a relay node, a lightweight neural network model is built in the sensor node, and edge reasoning and fault detection can be realized locally. Stable communication and real-time cooperation of large-scale nodes are ensured through a decentralized architecture of Bluetooth Mesh, establishment of a dynamic cluster head chain and optimization of an RSSI threshold value. And a low-power-consumption mechanism combining event driving and fixed polling is adopted, so that the energy consumption of the system is remarkably reduced. In-cluster communication loads are reduced through a clustering architecture, and the communication efficiency and the anti-interference capability of the system are improved in combination with channel separation of Mesh and BLE protocols.

The invention provides a gas extraction multi-parameter monitoring method and system based on edge calculation, and relates to the technical field of coal mine gas extraction. The method comprises the following steps: arranging multi-parameter sensor nodes in drill holes, pipelines and gas gathering stations, and accessing an edge calculation unit; performing time synchronization, zero drift correction, outlier elimination and normalization on the data of the gas concentration, the negative pressure, the flow, the temperature, the humidity and the hydrogen sulfide concentration to generate feature vectors; calling a lightweight neural network model for reasoning, and outputting an extraction efficiency score and a leakage risk degree; when the risk degree exceeds the limit, an acousto-optic alarm is triggered and a speed reduction instruction is issued; when the efficiency is insufficient, the negative pressure is adjusted; compressing and encrypting the result, and reporting the result to a cloud platform through a wireless network at regular time; and the cloud performs model incremental training based on historical data and feedback, and issues an update model through hot replacement. According to the invention, local intelligent analysis, rapid early warning and dynamic adaptive optimization are realized.

The invention discloses a millimeter-wave radar sparse point cloud-based human body posture recognition method, which belongs to the technical field of human body posture recognition, and comprises the following steps of: acquiring three-dimensional point cloud data of human body actions through a millimeter-wave radar, and preprocessing the three-dimensional point cloud data; and training a lightweight neural network model by using the preprocessed point cloud data, wherein the model comprises an edge convolution module and a grouping sparse Transform encoder module. The edge convolution module extracts spatial geometric features and detects dynamics, static postures are directly classified, dynamic postures capture a time sequence dependency relationship through a grouping sparse Transform module, attention calculation is only carried out on frames with feature changes exceeding a threshold value, and mean pooling aggregation is carried out on other frames. And finally, classifying the human body postures based on the spatial geometric features and the time sequence dependency relationship to obtain a classification result. The device is simple in structure, accurate in recognition and suitable for efficient deployment of edge equipment.

Embodiments described herein provide techniques to facilitate hierarchical scaling when quantizing neural network data to a reduced-bit representation. The techniques includes operations to load a hierarchical scaling map for a tensor associated with a neural network, partition the tensor into a plurality of regions that respectively include one or more subregions based on the hierarchical scaling map, hierarchically scale numerical values of the tensor based on a first scale factor and second scale factor via the matrix accelerator circuitry, the first scale factor based on a statistical measure of a subregion of numerical values of within a region of the plurality of regions and the second scale factor based on a statistical measure of the region that includes the subregion, and generate a quantized representation of the tensor via quantization of hierarchically scaled numerical values.

The invention discloses a locomotive wheel polygon damage detection method based on a lightweight neural network, and the method comprises the steps: obtaining axle box vertical vibration acceleration response signals of a heavy-load locomotive under the combination of different operation speeds and different wheel polygon abrasion degrees, and carrying out the preprocessing of the signals, the method comprises the following steps: acquiring frequency domain characteristics of a signal through fast Fourier transform, constructing a sample data set based on the frequency domain characteristics, constructing a wheel polygon damage detection network model, training the model by using the sample data set, and identifying a wheel polygon abrasion amplitude by using the trained polygon damage detection network model. And completing damage detection of the polygon of the wheel. The method can realize accurate and quantitative detection of the polygon abrasion degree of the heavy-load locomotive wheel, has the characteristics of accuracy, high efficiency and strong robustness, and also has relatively good interpretability.

The invention provides an equipment abnormal voiceprint detection system. The equipment abnormal voiceprint detection system comprises a voiceprint collection module used for collecting original voiceprint signals in real time and preprocessing the original voiceprint signals; the edge end detection module is deployed on edge computing equipment and is used for carrying out real-time anomaly detection on the voiceprint features by utilizing a one-dimensional lightweight neural network model; the abnormity credibility evaluation module is used for converting a real-time abnormity detection result into a probabilistic abnormity credibility score; the incremental data screening module is used for screening high-value samples from the real-time voiceprint data based on a dynamic density trend sensing algorithm and caching the high-value samples in the edge; the cloud model evolution module is used for carrying out evolution training on the detection model by utilizing a continuous learning mechanism; the generation and playback module is used for jointly generating pseudo samples through a variational auto-encoder VAE and a generative adversarial network GAN; and the model updating module is used for compressing the evolved cloud model and then issuing and replacing the original model in the edge end detection module so as to form a cloud-edge collaborative sustainable evolution closed loop.

The invention relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle detection tracking device, method and equipment, which comprises a data sensing unit, a calculation unit, a mechanical and power supply unit, a communication and storage unit and a countering execution unit, and is characterized in that the data sensing unit is used for acquiring image data and servo turntable rotation angle data; the calculation unit is used for deploying a lightweight neural network model and a multi-algorithm fusion process, processing sensor data and realizing unmanned aerial vehicle target detection and tracking logic control; the mechanical and power supply unit is used for providing physical support, servo motion capability and energy supply for the device; the communication and storage unit is used for real-time data transmission and storage and encryption archiving of historical data and image videos; and the countering execution unit is used for implementing signal interference or physical interception on the detected and tracked target. Therefore, the problems of easy target loss, low small target identification precision, poor complex environment adaptability, insufficient countering means and the like of a traditional device are solved.

The invention relates to the technical field of intelligent gateways, and discloses a multi-mode industrial Internet of Things intelligent gateway based on edge computing and an implementation method thereof, and the method comprises the steps: obtaining sensor data streams of a door magnetic sensor, a human body sensor, a temperature and humidity sensor and a smoke detector in an intelligent region; constructing a space-time fusion data matrix based on the sensor data stream; inputting the space-time fusion data matrix into an edge layer quantization neural network, a fog layer recurrent neural network and a cloud layer large model for distributed reasoning to obtain a reasoning result set; and performing priority queue scheduling and zero-copy transmission on emergency events, important events and conventional events in combination with the reasoning result set to generate a control instruction sequence, so that parallel operation of data acquisition and processing is realized, and the intelligent level, the response performance and the operation reliability of an intelligent regional Internet of Things system are improved.

The invention relates to the field of emerging software and information technology service, in particular to an artificial intelligence-based application system integrated management method, which comprises the following steps of: analyzing an API (Application Program Interface) log, structured data, an unstructured document, business work order data and business image data by pre-training a large model, and generating a unified semantic representation in combination with multi-modal contrast learning; a knowledge distillation technology is adopted to migrate the large model capability to a lightweight neural network, and a domain knowledge base is generated based on fine tuning of business work order data; a user multi-mode instruction is analyzed, a DAG service agent template library is matched, and an execution chain is dynamically combined based on a reinforcement learning strategy, so that text processing, multi-system collaborative reasoning and RPA service process automation are realized; and collecting an execution log, and updating lightweight network parameters and an intelligent agent strategy library through an online distillation algorithm to form a closed-loop optimization mechanism. According to the method, the service response efficiency can be improved, and the problems of enterprise multi-service system data splitting and process stiffness are solved.

The invention discloses an intelligent interaction control method for AI glasses, and relates to the technical field of AI glasses, and the method comprises the following steps: synchronously collecting user input signals and environmental parameters through a multi-source sensor group built in the glasses, the sensor group at least comprising an IMU, a binocular camera, a microphone array and a physiological sensor; carrying out real-time fusion processing on the multi-modal input data by adopting a lightweight neural network to generate an interaction intention feature vector; based on real-time detection results of environmental noise intensity and illumination conditions, dynamically distributing weight coefficients of all interaction modes; generating a hierarchical response instruction according to the weight coefficient and a confidence threshold; and continuously optimizing the personalized interaction strategy of the user through a federal learning framework. According to the AI glasses intelligent interaction control method, through multi-modal weighted fusion, edge AI acceleration and federated learning optimization, the interaction success rate in a complex environment is improved to 93.6%; the end-to-end delay is controlled within 28ms; and a user-defined interaction strategy is supported.

The invention relates to the technical field of gas detection, in particular to a gas detection system and method based on an AI technology, and the system comprises a differential sensing module, a drift compensation processor, a dynamic risk assessment unit and a cloud-edge collaborative learning framework. Compared with the prior art which adopts a single-sensor independent compensation scheme and has the defects that the false alarm rate is high and long-term drift is difficult to correct under environmental interference, the scheme innovatively introduces a dual-mode differential sensing architecture and an adaptive drift compensation algorithm, a reference sensor is packaged in an inert gas environment, and the reference sensor and a main sensor are subjected to real-time collaborative analysis; the pre-trained lightweight neural network is combined to dynamically eliminate temperature and humidity cross interference, the detection precision and stability under complex working conditions are remarkably improved, and the aging distortion problem of the semiconductor sensor is effectively solved.

The invention relates to a lightweight neural network model implementation method and system for small target detection in a complex aerial photography scene, and belongs to the technical field of computer vision and unmanned aerial vehicle target detection. In order to solve the problem of low detection precision caused by small target size, complex background, easy feature submerging and the like in an aerial image of an existing unmanned aerial vehicle, the method comprises the following steps: constructing a multi-scale adaptive hierarchical feature enhancement module MSAHFE, and combining multi-scale pooling and differential edge enhancement to improve feature sensitivity; constructing a lightweight feature selection module LAFS based on an attention mechanism, and screening high-correlation features by using a space and frequency double-domain attention mechanism; a lightweight detection head LWDeect is constructed, and shared packet convolution and self-calibration convolution are utilized to reduce the calculation complexity. The method has the advantages of being high in detection precision, small in parameter quantity, high in reasoning speed and the like, and is suitable for complex aerial photography and other application scenes needing real-time small target detection.

The invention relates to an instant denoising method for electroencephalogram signals of students in a classroom scene. The method comprises the following steps: based on electroencephalogram acquisition equipment and a Daisy expansion board, acquiring electroencephalogram signal data of a student through a dry electrode device; and carrying out sliding mean filtering on the electroencephalogram signal data to remove baseline drift, eliminating noise by adopting a band-pass filter to obtain preprocessed electroencephalogram signal data, and inputting the data into a lightweight neural network model to carry out channel-by-channel denoising processing to obtain a denoised electroencephalogram signal. The model comprises a multi-scale feature extraction layer, a time sequence modeling module and a residual connection structure. According to the method, through lightweight neural network design such as multi-scale feature extraction, time sequence modeling and a residual connection structure, the denoising effect is improved, the real-time performance and high efficiency of electroencephalogram denoising processing are guaranteed, and an electroencephalogram data basis with a high signal-to-noise ratio is provided for subsequent teaching quality evaluation.

The invention discloses a construction site three-dimensional modeling method based on multi-sensor fusion. The method comprises the following steps: synchronously acquiring data by using a laser radar, a camera, an infrared sensor, a GPS (Global Positioning System) and an IMU (Inertial Measurement Unit); the method comprises the following steps of: performing denoising and filtering operation on collected original data, aligning a timestamp and a space of the collected data, fusing the collected multi-source data, and aligning and fusing the processed multi-source point cloud data through a point cloud registration and fusion algorithm to generate a complete three-dimensional point cloud model; generating a grid model by using a surface reconstruction algorithm based on the point cloud data, and performing texture mapping in combination with texture information collected by a camera; the model is simplified and the precision of the model is improved through a model optimization technology, and a reliable three-dimensional visualization foundation for construction management and analysis is formed; real-time detection and modeling of dynamic targets such as mobile equipment and workers on a construction site are realized, and the real-time response capability of the system in a complex construction site environment is ensured through combination of a lightweight neural network architecture and an edge computing technology.

The invention discloses an intelligent management system for submarine cable recovery of an offshore wind plant, which belongs to the technical field of submarine cable recovery, and comprises a multi-source sensor data acquisition module comprising a shipborne meteorological sensor, a multi-source sensor array deployed on a recovery ship, and a streaming data processing module; the incremental online learning module comprises a lightweight neural network and an elastic weight solidification unit; the dynamic weight decision-making module is used for dynamically distributing weights based on multi-modal feature information entropy, outputting a cable residual value grade and supporting misjudgment case feedback optimization; and the edge cloud collaboration module receives the incremental update package issued by the cloud. According to the intelligent management system for recovery of the submarine cable of the offshore wind plant, real-time assessment and residual value judgment of cable damage are realized through a full-process closed loop of multi-modal data acquisition, dynamic noise suppression, incremental learning updating, intelligent decision and cloud collaborative optimization.

The invention discloses a target detection model system and method based on a lightweight network, and belongs to the technical field of target detection, and the system comprises an input module which is used for the reading and preprocessing of an input image, so as to optimize the format and quantity of input data of a model; the feature extraction module is used for extracting key features of the image by using a lightweight neural network; the feature pyramid module is used for fusing features on different scales to adapt to multi-scale target detection; according to the invention, the multi-scale feature fusion design enables the system to be more stable when processing a small target and a large target, and the detection precision is higher; flexible target classification and positioning: a detection head module in the system is designed with classification and regression branches, efficient target classification and bounding box prediction can be carried out according to features extracted by a feature pyramid, and in combination with a loss function based on cross entropy and smooth L1 loss, the system can effectively reduce classification errors and bounding box regression errors, so that the accuracy of target classification and positioning is improved. The detection precision is improved.

The invention discloses a power transmission tower attitude monitoring and vibration intrusion identification method based on multi-source data fusion, and the method comprises the steps: carrying out the deep fusion of GPS displacement data and accelerometer inclination angle data, employing a dynamic weight fusion mechanism: adjusting a fusion coefficient alpha in real time according to the GPS signal quality, guaranteeing the accuracy and stability of data fusion, and at the same time, carrying out the real-time adjustment of the fusion coefficient alpha, and carrying out the recognition of the vibration intrusion of a power transmission tower. And continuously optimizing the error propagation model by taking the final inclination angle theta final of the power transmission tower as a state variable, updating the attitude estimation of the power transmission tower, and realizing + / -0.01-degree high-precision attitude monitoring. According to the invention, the system can output a high-precision three-dimensional attitude angle in real time by combining the rapid dynamic response of the accelerometer and the long-term stability of the GPS. Besides, an LCD-MPE (feature extraction technology) is introduced, lightweight neural network classification is combined, vibration signal sampling, noise reduction and abnormal signal segment identification, analysis and extraction are carried out on the transmission tower through an MEMS accelerometer, and second-level detection and behavior type discrimination of invasion behaviors around the transmission tower are realized. The core technical barrier from'inability perception 'to'accurate identification' is overcome, and subversive upgrading of power facility safety protection from'passive disposal 'to'active defense' is promoted.

The invention relates to a ship unloader material transfer process adaptive flexible dust suppression system, which comprises an environment sensing module, an integrated multi-parameter dust concentration sensor array, a temperature and humidity sensor and a material characteristic detection unit, and is used for collecting dust concentration, environment temperature and humidity and material viscosity / granularity data in a transfer area in real time; the data processing module is based on a dynamic model library constructed by an embedded industrial computer, comprises a fuzzy control algorithm and a lightweight neural network, and is used for fusing multi-source sensing data and generating a dust suppression strategy; the flexible dust suppression execution module is composed of a variable-angle high-pressure atomizing nozzle array, a dynamic pressure regulating valve group and a flexible sealing cover body; and the self-adaptive control module is internally provided with a non-linear mapping table based on the material transfer speed and the dust generation amount, and real-time adjustment of the nozzle pressure and the spraying angle is achieved through PID and expert rule mixed control. Accurate dust suppression for different materials is achieved, and the reliability and environmental protection performance of the system are improved.

The present invention discloses a neural network construction method having average quantization mechanism that includes steps outlined below. A weight combination included in each of network layers of a neural network is retrieved. A loss function is generated according to the weight combination of all the network layers and target values. Corresponding to each network layers, a Gini coefficient of the weight combination is calculated and the Gini coefficients corresponding to all the network layers are accumulated as a regularized correction term. The loss function and the regularized correction term are merged as a regularized loss function to perform training on the neural network accordingly to generate a trained weight combination of each of the network layers. A quantization is performed on the trained weight combination of each of the network layers to generate a quantized neural network, in which each of the network layers thereof includes the trained weight combination.

The invention discloses a railway axle counting equipment management system which comprises a hyper-fusion sensing terminal, an edge intelligent diagnosis unit, a digital twin management platform, a block chain evidence storage subsystem and an intelligent interaction terminal. The super-fusion sensing terminal is integrated with an orthogonal magnetic head array, a vibration sensor, a temperature and humidity sensor, a tilt angle sensor, a laser ranging module, a current sensor and a dust concentration sensor; the edge intelligent diagnosis unit is internally provided with a lightweight neural network and a fault feature knowledge graph; the digital twinborn management platform can construct a three-dimensional digital twinborn body of axle counting equipment, and integrates a multi-physics field coupling simulation module; the block chain evidence storage subsystem can perform uplink evidence storage on the fault original signal and the maintenance work order record; the intelligent interaction terminal comprises a three-dimensional visual interface and a mobile operation and maintenance App. The intelligent level of railway axle counting equipment management can be effectively improved, railway transportation safety and operation efficiency are enhanced, and the method is suitable for various scenes such as high-speed railways and urban rail transit.

The invention discloses a debris flow infrasound recognition method based on a Mel spectrogram and a lightweight neural network, and belongs to the field of image data processing, and the method comprises the steps: collecting infrasonic wave signals generated by on-site debris flow and other events, and generating a Mel spectrogram data set; constructing a channel alignment fusion bridge; constructing an improved MBConv block; constructing a debris flow infrasound recognition network based on the channel alignment fusion bridge and the improved MBConv block; and training by using the Mel spectrogram data set to obtain a debris flow infrasound identification model, and identifying the infrasound signal to be identified. According to the method, the characteristics of debris flow infrasonic waves are combined, samples in the data set are constructed into the infrasonic mel spectrogram, the debris flow infrasonic recognition network is designed based on the multi-stage lightweight feature extraction module, the infrasonic mel spectrogram is classified and recognized, the calculation complexity can be effectively reduced, the parameter quantity can be reduced, low-power-consumption equipment can be adapted, and the method is suitable for popularization and application. The method can improve the classification accuracy, and is especially suitable for the real-time early warning scene of debris flow.