32results about How to "Avoid analysis" patented technology

The invention provides a method for quickly training an LSTM (Long Short Term Memory) model by time series data, which relates to the technical field of data processing, and comprises the following steps: S1, a data integration stage: efficiently processing and utilizing telemetering real-time stream data through Kafka and Flink, s2, a data storage stage: storing the time sequence data processed in the integration stage into a high-performance database IoTDB specially designed for the time sequence data according to a time sequence, and S3, an algorithm application stage: constructing an LSTM model, setting parameters, optimizing the LSTM model, and performing time sequence data prediction by using an LSTM algorithm model. According to the method for rapidly training the LSTM model through the time series data, the time series telemetry data generated in the test flight process are collected and processed through Kafka, real-time data of an aircraft can be rapidly captured, processed and analyzed, data transmitted by a sensor and an algorithm model are decoupled through the Kafka, data output by the sensor can be directly sent to the Kafka, and the LSTM model can be rapidly trained through the Kafka. And it is avoided that an algorithm model needs to be concerned to process data.

The invention discloses a saw blade tooth profile defect visual inspection system based on a deep learning model, and relates to the technical field of saw blade tooth profile defect detection.The saw blade tooth profile defect visual inspection system comprises a light source capable of providing a plurality of different specific wavelengths, and a first visual inspection device shoots a saw blade and generates a digital image; pixel information in the scanning image is subjected to edge detection, texture analysis and morphological operation, and then corresponding characteristic values are extracted; and the second visual detection device is used for receiving the light of different specific wavelengths reflected from the surface of the saw blade in a certain length range, and analyzing and generating reflection spectrum data. The method comprises the following steps: comparing a feature value extracted by a first visual detection device with a set threshold range to preliminarily identify an obvious tooth profile defect area, and taking a comparison result of the feature value and the set threshold as a starting condition for carrying out second accurate detection or not; and the secondary visual detection device performs more accurate detection on the area with the preliminarily detected potential defects, and deep defect features are identified.

The invention discloses a rapid batch switching authentication method for predictability and aggregation promise proof, which realizes batch authentication of a plurality of user terminals by generating system parameters and secret keys through a trust mechanism and combining predictability promise and aggregation signature technologies. In the authentication process, the edge base station to be switched exchanges with the current communication edge base station through an encrypted message, verification information such as a pre-authentication certificate and a signature is generated through the user terminal, and identity authentication is completed step by step. The method comprises the following specific steps: a trust mechanism generates a private key, a public key, a pseudo identity and an expiration timestamp for each device; through multiple rounds of message exchange, the edge base station and the user terminal perform identity verification by using a shared key and a hash function; and finally carrying out efficient verification on a plurality of users through batch signature and aggregation commitment. According to the method, on the basis of ensuring the security, the speed of switching the authentication process is greatly improved, and the method is suitable for a large-scale network environment, especially in edge computing and mobile communication networks.

The invention provides a network data asset affiliation identification method and system based on a stacking model, and relates to the technical field of network security. The method comprises the following steps: acquiring a network data asset data set comprising a plurality of network data asset samples; extracting a plurality of network features of the network data asset sample; constructing a plurality of base models; according to each network feature, determining a base classification result of each base model for the network data asset sample; determining the weight of each base model according to a base classification result; taking the output of each base model as the input of each channel of the multi-layer perceptron, taking the weight of each base model as the weight of the channel, and stacking the base models and the multi-layer perceptron to form a stacked model; determining a comprehensive classification result of the stacking model for the network data asset samples; obtaining to-be-identified network data assets; and determining a comprehensive classification result of the to-be-identified network data assets through the stacking model, and performing attribution identification on the to-be-identified network data assets.

The invention relates to the technical field of air conditioner control, in particular to a fusion circuit and method for machine room point position collection, and the circuit comprises an indicator light state obtaining unit, a first photoelectric coupler U1 is provided with a first pin PI N1 at the output end and used for being connected with a processor, and the input end is connected with the direct current output end of a rectifier bridge D1; the alternating-current input end of the rectifier bridge D1 is connected with two ends of a to-be-detected point through a detection line; the output end of a second photoelectric coupler U2 is provided with a second pin PI N2 used for being connected with a processor, the anode input end of the second photoelectric coupler U2 is connected with an isolation power supply VCC2 and then is connected with one detection line through a first relay S1, and the cathode input end of the second photoelectric coupler U2 is connected with the other detection line through a second relay S2; the first pin PI N1 provides a level signal to control the first relay S1 and the second relay S2 to be opened and closed through the relay control circuit; the circuit is designed from the button and the indicating lamp, the button state and the indicating lamp state are obtained, and then the process of analyzing and transforming a cabinet in an existing method is avoided.

The invention relates to the technical field of warp knitting machine control, in particular to a carbon fiber warp knitting machine operation method based on the Internet of Things, which comprises the following steps: determining a knitting control state of a warp knitting machine, determining an initial knitting speed and matching a traction speed according to the initial knitting speed; determining a central area and an edge area of the braided fabric according to the raw material diameter of the carbon fiber raw material and the target fabric density; determining the motion trails of the warp knitting needle and the sinker in the knitting process so as to determine the equipment characterization state of the warp knitting machine; determining whether the warp knitting machine acquires a surface image of the braided fabric or not according to the equipment characterization state; acquiring a surface image of the braided fabric at each preset time node to determine a density characterization trend of the braided fabric; determining an operation characterization trend according to the density characterization trend to determine whether to adjust the initial knitting speed; according to the center flat state and the fabric edge state, the operation characterization trend is secondarily determined so as to determine whether to adjust the traction speed or not. Therefore, linkage among links of monitoring equipment, controlling the quality of braided fabric and optimizing the production process is realized.

The invention belongs to the technical field of laser high-temperature measurement, and discloses a non-contact transient high-temperature measurement system based on LIBS (laser-induced breakdown spectroscopy), which comprises an LIBS heating device; a temperature control device; a base material posture adjusting device; provided is a spectrum modulation detection device. According to the non-contact transient high-temperature measurement system based on LIBS, through accurate control of laser, a tiny breakdown area is generated on the surface of a base material, instant temperature rise is achieved, atoms or molecules in the base material are promoted to be excited, spectral signals are generated, local instant analysis of the base material and rapid capture of the sample state are achieved, and the measurement accuracy is improved. The whole base material is prevented from being analyzed, so that time and cost are saved; meanwhile, the laser beams directly irradiate the surface of the base material for analysis, direct contact with the base material is not needed, damage and pollution to the base material are avoided, and spectral pollution of water vapor and carbon dioxide is also eliminated.

The present invention relates to the technical field of smart street lamps, and specifically discloses an energy-saving smart street lamp management method and system, the method comprising: determining all re-divided roads based on historical pedestrian flow data of each straight road section in a management area and control circuit grouping information of all smart street lamps; determining smart street lamp grouping information based on geographical distribution data of all smart street lamps and all re-divided roads; determining all pedestrian flow time periods of each re-divided road within a unit time period based on historical pedestrian flow data of each re-divided road section, and determining the average pedestrian flow data of each pedestrian flow time period of each re-divided road section within a unit time period; based on the average pedestrian flow data of each pedestrian flow time period of each re-divided road section within a unit time period and the smart street lamp grouping information, grouping and setting the lighting power of all smart street lamps to obtain smart street lamp energy-saving management results; for saving energy consumption of smart street lamps in a large management area.

The invention relates to the technical field of artificial intelligence, and provides an agent interaction-based document conversion method, which comprises the following steps of: splitting a target document according to pages, and distributing an agent for each split page; detecting the embedded resources of the distributed page by the intelligent agent which accepts the page distribution; if the agent accepting page allocation detects that the risk element exists in the allocated page, the risk element is removed, and the page with the risk element is converted into a page in a security format; if the agent receiving page allocation detects that an error exists in the allocated page, other agents are informed of the error, and the agents obtaining the error generate an error correction scheme of the error page through the shared knowledge base; error correction is carried out on the page with errors by adopting an error correction scheme, and the page after error correction is converted into a page in a safe format; and recombining the pages in the secure format into a complete document to be output. According to the technical scheme, the document conversion can be ensured to be safe, efficient and consistent.

The invention relates to an industrial data processing method, system and device based on an LAMBDA architecture. The method comprises the following steps: extracting periodic features of periodic historical data, constructing a periodic feature library, and setting corresponding periodic repair rules according to data with different features in the periodic feature library to obtain a periodic repair rule set; performing periodic feature recognition on the real-time data according to the periodic feature library to obtain periodic real-time data and non-periodic real-time data, defining a business rule according to the non-periodic historical data, and performing data repair on the non-periodic real-time data according to the business rule by adopting a rule engine to obtain a repair result. Performing data restoration on the periodic real-time data by utilizing the periodic restoration rule set; and performing data analysis according to the repaired periodic real-time data and the aperiodic real-time data. The method can adapt to periodic change and business requirements of data, and the flexibility and the intelligent level of the system are improved.

The present invention discloses a method for drawing and interpreting a real-time three-dimensional volume loop of the left atrium and ventricle, and belongs to the field of cardiac ultrasound. The method of the present invention comprises the following steps: S1, image retention: using a Philips echocardiogram instrument with a Philips Heart Model mode and equipped with an X5-1 probe to use the Heart Model mode to retain images of at least 6 cardiac cycles on the apical four-chamber section; S2, software analysis: analyzing the retained images to obtain a real-time synchronized three-dimensional volume curve of the left atrium and left ventricle; S3, point drawing: the corresponding left atrium and left ventricle volumes taken at different times on the three-dimensional volume curve in the software are drawn with the left ventricle volume as the X-axis and the left atrium volume as the Y-axis; the present invention can easily judge the left atrium and ventricle function and the left atrium and ventricle relationship through the real-time synchronized left atrium and ventricle volume curve, can identify abnormal heart function that cannot be identified by conventional echocardiogram, and avoid fragmented analysis of the left atrium and ventricle function; high repeatability and no harm to the human body.

The present invention provides a privacy protection method and device for audio information obfuscated reversible adversarial samples. The method proposes a new framework called "information obfuscated reversible adversarial examples". The framework can ensure the restoration of the original quality of the audio with authorization while ensuring the privacy security of the audio. By adopting the method of the present invention, the security and privacy protection level of audio data can be greatly improved, and unauthorized access and analysis can be effectively prevented. It is intended to solve the problems in the prior art of audio data privacy protection that adversarial attacks are easy to detect, the quality of audio is significantly reduced after processing, and the limitations of the existing reversible adversarial example (RAE) technology in audio application scenarios.

The embodiment of the invention provides a resource capacity expansion method and device, electronic equipment and a storage medium. The resource capacity expansion method comprises the steps that in response to a calling event of a set function, the set function is called through a proxy function, and the set function comprises an application programming interface related to a thread key; under the condition that calling of the set function fails, the set function is processed based on a target array associated with the set function; wherein the target array is an expansion resource stored in an expanded memory, and is used for storing contents associated with the set function. The technical problem caused by unreasonable use of thread key resources in troubleshooting is solved, under the condition that call of the set function fails, the set function is processed by expanding the target array in the memory, that is, the set function is processed by expanding the resources, use of the thread key resources is prevented from being analyzed, and the service life of the set function is prolonged. And the response speed of setting function calling is improved.

The utility model discloses a quartz tube sleeve for assembling a bonding wire annealing tube gas sensor, which comprises a tube body, the fixing pipe is used for being inserted into an annealing pipe opening to achieve sleeving connection of the pipe body and the annealing pipe, the limiting protrusion is used for abutting against the end face of the annealing pipe opening, the mounting base is used for mounting a gas sensor, and the limiting protrusion is arranged on the inner circumferential wall of the pipe body. One end of the limiting protrusion is connected between the fixing pipe and the installation base, the other end of the limiting protrusion is arranged in an inclined mode in the direction where the fixing pipe is located, a connecting table is arranged on the side wall of the installation base, and a containing cavity used for containing the sensor body is formed in the connecting table. The gas concentration monitoring device is simple in structure and convenient and reliable to assemble, facilitates monitoring of gas concentration and has a good use effect.

The invention discloses a digital body interaction authentication method with behavior privacy protection and supervision functions. The invention relates to six components including a digital avatar, a digital avatar, a service provider, a voucher issuer, a distributed supervision authority and a behavior chain. The method comprises five stages: a system initialization stage, a digital avatar generation stage, an access-to-ID construction stage, an on-chain and off-chain interaction authentication stage and a malicious entity supervision stage. According to the invention, based on the chameleon hash and the block chain technology, a method for accessing one ID is constructed, and the construction of one service and one duplicate is realized by embedding an implicit identity authentication key, so that the non-linkability of a meta-universe digital body in the service access process is realized; and avatar behaviors, preference privacy protection and self-sovereignty control of a natural person on the avatar are realized. In addition, a combined supervision mechanism is constructed by using dynamic threshold secret sharing and a Paillier homomorphic encryption technology, and responsibility asking for malicious entities is realized under the condition of not invading privacy of legal avatars.

This invention belongs to the technical field of network security protection and relates to a hybrid attack detection and defense method based on deep learning. The method includes the following steps: extracting multi-dimensional traffic statistical features and application layer protocol semantic features; outputting a high-dimensional behavior-related feature vector; issuing an attack trigger signal; outputting an attack context description; generating a collaborative defense command stream; outputting real defense effect feedback data; and adjusting the convolutional autoencoder model and the graph neural network. This invention solves the problem that existing network attack detection and defense technologies struggle to simultaneously meet the core protection requirements of comprehensive detection dimensions, reasonable computational resource scheduling, dynamic adaptability of defense strategies, and adaptive model iteration in hybrid attack scenarios.

The invention relates to a buried hill fractured reservoir logging porosity calibration method, device, medium and equipment, and the method comprises the steps: carrying out the physical analysis based on an oil and gas field buried hill fractured reservoir sample, and obtaining the porosity and permeability; on the basis of the porosity and the permeability, a cross plot of the porosity and the permeability of the buried hill fractured reservoir of the oil and gas field is constructed, and according to the distribution range of the permeability of the full-diameter core sample and the permeability of the plunger core sample in the cross plot, the threshold value of the permeability of the full-diameter core sample and the threshold value of the permeability of the plunger core sample are determined; based on the logging data of the oil and gas field buried hill fractured reservoir, the fracture porosity, the permeability and the logging porosity of the oil and gas field buried hill fractured reservoir are obtained through calculation; comparing the calculated permeability of the buried hill fractured reservoir of the oil and gas field with the boundary value of the permeability of the full-diameter core sample and the permeability of the plunger core sample, and judging the development condition of the buried hill fractured reservoir of the oil and gas field.

The application is suitable for the fields of high-end flexible sensing preparation and multi-modal sensing detection technology, and provides a patch type ultrasonic biomarker integrated sensor and a detection method.The sensor comprises a patch type ultrasonic transducer array, which is used for real-time monitoring of muscle and bone diseases, realizes multi-parameter multi-modal data information fusion detection and analysis of muscle and bone structure-function, a microneedle electrode array, which is used for in-situ detection of biomarkers and collection of physiological electrical signals, a flexible electronic circuit, which is used for bearing and connecting the patch type ultrasonic transducer array and the microneedle electrode array, and an acoustic coupling adhesive layer, which is used for skin fitting and acoustic coupling.The application can systematically evaluate the structure, mechanics, metabolic inflammation level and nerve function activity coordination of muscle and bone, provides a new means for early identification, quantitative evaluation and personalized intervention of muscle and bone diseases, and has important clinical and industrial values.

The present application relates to the field of computer technology, and in particular to a resource analysis method, device, electronic device and storage medium, the method comprising: when submitting a resource to be analyzed, determining a preset call function corresponding to the resource to be analyzed; the preset call function includes a first storage path where the resource to be analyzed is located and a second storage path where a resource analysis tool is located; based on the preset call function, calling the resource analysis tool under the second storage path, and based on the preset call function, passing the first storage path to the resource analysis tool; based on the resource analysis tool, analyzing the static features of the resource to be analyzed under the first storage path, and obtaining a static analysis result of the resource to be analyzed. The present application can improve the efficiency and accuracy of resource analysis.

The application relates to the technical field of elevator intelligent monitoring, and discloses an elevator abnormal behavior recognition and early warning method and system based on AI vision analysis, which comprises the following steps: receiving a door machine state signal, waking up a static target analysis process when the door is opened, calculating a target physical size based on perspective transformation and outputting a static early warning, reading a high-frequency load sampling sequence to generate a physical dynamics sequence when the door is closed, triggering a hardware interrupt when a first-order difference of the physical dynamics sequence exceeds a threshold value, waking up a dynamic behavior analysis process in response to the interrupt, intercepting a video stream to extract moving foreground pixels and projecting the moving foreground pixels to a two-dimensional grid of a car bottom plane, calculating a visual kinematics acceleration sequence and a spatial distribution variance, using a dynamic time warping algorithm to obtain a time domain matching distance parameter of the physical dynamics sequence and the visual sequence, and constructing a joint confidence based on the spatial distribution variance and the time domain matching distance parameter to output an early warning. Through multi-modal data fusion and dynamic time alignment, the application reduces the consumption of computing resources and improves the recognition accuracy.

The invention belongs to the technical field of network security protection, and relates to a hybrid attack detection and defense method based on deep learning, and the method comprises the following steps: extracting multi-dimensional traffic statistical features and application layer protocol semantic features; outputting a high-dimensional behavior association feature vector; sending out an attack trigger signal; outputting attack context description; generating a cooperative defense instruction stream; outputting real defense effect return data; and adjusting the convolutional self-encoding model and the graph neural network. According to the method, the problem that the existing network attack detection and defense technology is difficult to meet core protection requirements of detection dimension comprehensiveness, computing resource scheduling reasonability, defense strategy dynamic adaptability and model iteration adaptivity in a mixed attack scene at the same time is solved.

This invention discloses an online monitoring method, system, equipment, and storage medium for the operating status of secondary equipment in substations, belonging to the field of power system monitoring technology. The method involves acquiring the status categories of each secondary device and constructing a secondary circuit model. The theoretical values ​​for each monitoring point are calculated using the model, and the residual vector is obtained by comparing the theoretical and measured values. The operating status is then determined based on the residual vector. A target fault set is selected by analyzing the similarity between the fault feature vector corresponding to a virtual fault and the residual vector. Within the target fault set, a correlation index for each virtual fault is determined based on the correlation between the disturbance vector and the residual vector, and a suspected fault set is selected based on the correlation index. Within the suspected fault set, the fault intensity of each virtual fault is calculated by constructing a fault location model. Virtual faults with a fault intensity greater than a set intensity value are used as detection results to provide early warnings for the operating status of the secondary equipment.

The present invention relates to a method for compensating positioning errors of a manipulator based on an improved WOA-GA-BP neural network, wherein the method comprises the following steps: inputting a motion type and selecting a motion compensation model, inputting an ideal end position coordinate, initializing the weight and threshold of a BP neural network prediction model; improving the BP neural network prediction model based on an improved whale optimization algorithm; calculating new control parameters and updating a coefficient vector A, updating individual positions according to the behavior of circling toward prey or according to the behavior of shrinking and surrounding or according to the behavior of hunting prey; calculating new individual fitness, tracking the new individual optimal position and the global optimal position, and outputting a global optimal solution; and outputting the actual compensated joint motion amount. The mutually coupled joint errors during the movement of the manipulator joints are simulated through multiple high-dimensional and high-order mathematical models, thereby avoiding the decomposition and analysis of coupling error factors, and having fewer unknown quantities than parameter compensation, and having stronger universality.

The invention discloses an aircraft flow field aerodynamic thermal characteristic convergence discrimination method, which belongs to the field of aircraft thermal protection, and comprises the following steps: based on the relevance between flow field parameters and aerodynamic thermal characteristics, constructing a weight function, carrying out weighted calculation on the average relative residual error of flow field thermal correlation parameters to discriminate the basic convergence of the flow field, and combining the thermal protection characteristics of the aircraft to determine the convergence of the flow field. Aerodynamic thermal characteristic convergence is judged by adopting different methods in a high-heat area and a large-area low-heat area. According to the method, the phenomenon that flow field convergence is inconsistent with aerodynamic thermal characteristic convergence can be effectively avoided, and unnecessary redundant iteration can be reduced while the precision requirement of the thermal protection engineering of the aircraft is ensured.

The present invention relates to the field of image processing techniques. Disclosed are an image recognition method and device, a terminal apparatus, and a medical system. The image recognition method disclosed in the present invention comprises: using a first recognition model to determine whether a medical image indicates a pathological change; and using a second recognition model to perform advanced recognition on the medical image showing a pathological change so as to obtain a pathological change degree recognition result indicating a degree of the pathological change indicated in the medical image. The present invention eliminates the need to perform manual analysis and formulate a feature extraction solution, thereby enhancing efficiency and precision of medical image recognition.

The invention discloses a restraint testing device for sports protectors, and relates to the technical field of testing devices. The device comprises a base, a stretching device is arranged at the top of the base, a moving device is arranged at the top of the base, an adjusting device is arranged at the top of the moving device, a supporting column is fixedly connected to the top of the adjusting device, and a human body model is fixedly connected to the top of the supporting column. The outer wall of the human body model is sleeved with a protector body. Through cooperation of a U-shaped rail block, a long rod, a moving block, a T-shaped handle, an elastic bag ring and liquid storage cotton, an anti-wear agent in the liquid storage cotton can be smeared to the surface of a human body model through a flow guide groove in the moving process of the liquid storage cotton, so that the situation that the human body model is abraded due to long-time use of the human body model, the friction coefficient is changed, and the service life of the human body model is prolonged is avoided. The friction force is not kept consistent any more, and the accuracy of a test result is influenced.

The invention discloses a work order processing method and device, a storage medium and electronic equipment. The method relates to the technical field of data processing, and comprises the following steps: classifying a target record work order recording a to-be-processed event through a data classification device to obtain target category information corresponding to the target record work order; processing the target record work order through a work order generation device corresponding to the target category information to obtain a target operation work order; and distributing the target operation work order to a first target object through a work order distribution device so as to process the to-be-processed event based on the target operation work order through the first target object. Through the work order distribution method and device, the problem that the work order distribution efficiency is relatively low due to the fact that the work orders are distributed in a manual mode in related technologies is solved.

The embodiment of the specification discloses a method, device and equipment for detecting abnormal information in a mine, which can comprise: acquiring a living body detection result in a preset area in the mine; if the living body detection result indicates that a target living body exists in the preset area, starting an image acquisition device in the preset area; acquiring a target image containing the target living body collected by the image acquisition device; inputting the target image into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; the abnormality detection model is used for detecting whether abnormal information exists in the target image; the abnormal information includes that the target living body in the target image has a rule violation behavior. Based on the scheme of the application, the abnormal situation in the mine can be automatically and efficiently detected.

The invention discloses an image processing method and device, a storage medium and electronic equipment, and relates to the technical field of image processing.The method comprises the steps that a first image of a vehicle parked at a target position is obtained based on a dual-domain private network, and one or more cameras used for collecting vehicle appearance information in an omnibearing mode are arranged at the target position; when it is monitored that the moving object passes through the vehicle, a second image of the vehicle is obtained; and comparing the first image with the second image to judge whether the vehicle is abnormal or not. Compared with the prior art, the method has the advantages that the images of the vehicle parked at the target position at different moments are acquired and compared to judge whether the vehicle is abnormal or not, image information acquisition is performed only when the moving object is monitored, continuous video data acquisition and analysis are avoided, and the accuracy of video data acquisition is improved. The consumption of computing resources is reduced, the safety of vehicle parking is improved, and meanwhile, the data transmission is safer and quicker by using the dual-domain private network.