70 results about "Detection threshold" patented technology

The application discloses a display panel defect detection method and system, comprising the following steps: obtaining original image data of a display panel; preprocessing the original image data to generate standardized image data; extracting multi-dimensional defect features based on the standardized image data; inputting the multi-dimensional defect features into a pre-trained defect classification model to output a defect type and a confidence; generating a defect distribution map based on the defect type and the confidence, and adjusting a detection threshold based on the defect distribution map; optimizing the defect classification model based on the adjusted detection threshold, and generating a dynamic quality control strategy. The application has the following advantages and effects: breaking through the limitation of a single data source by multi-spectral fusion and multi-dimensional features, covering surface and internal defects; adapting to changes in the production line environment by dynamic threshold and model optimization, and ensuring long-term detection accuracy.

The present application relates to the technical field of image processing, more particularly, the present application relates to a boring and milling machine operation area safety monitoring method based on video analysis, the method comprises the following steps: acquiring real-time operation video of the boring and milling machine and extracting real-time operation images, recognizing the operation area through a trained neural network; calculating the motion dispersion index of each pixel point in the operation area, and combining the pixel gray difference value and the instantaneous motion rate to construct a dynamic interference factor; based on the dynamic interference factor, the detection threshold of each pixel point is adaptively adjusted, and the target pixel point is screened out through the gray difference; the target pixel point is subjected to clustering processing, and the safety state of the machine tool is determined according to the distance between the cluster center and the center of the main shaft tool. The present application significantly reduces the false alarm rate caused by environmental noise, realizes the hierarchical safety protection of the boring and milling machine operation, and guarantees the continuity of the machining process.

The present application relates to a kind of sonar system compensation circuit and method based on multi-channel switching;Wherein the circuit uses cascade architecture, the core is to use domestic CBMG708 multiplexer to build at least one program-controlled gain stage, by the gain resistance of different resistance in series after each selection end is connected to operational amplifier, multiple signal channels with 6dB gradient basic gain are formed.Combined with FPGA+ARM dual-core control module, according to target action distance, corresponding channel is gated in real time, the fast and accurate gain compensation of sonar signal of different distances such as near, middle and far is realized.The method ensures that the output signal amplitude in the whole distance range is higher than the detection threshold, while having the advantages of small gain error, low power consumption and supply chain controllable, effectively solves the problem of insufficient flexibility and slow response of traditional single-channel compensation circuit.

The application discloses a long context processing robust enhancement method and system based on neuron dynamic adjustment, and belongs to the technical field of text sequence processing, and comprises the following steps: embedding a nerve regulation unit in each Transform layer of a large language model used for a long context processing task; and the nerve regulation unit sequentially performs the following steps: dynamically activating abnormal perception, that is, adaptively calculating an abnormal detection threshold of a current layer according to activation value statistical characteristics of the current layer; activating a hard-gate truncation, that is, generating a binary hard mask according to the abnormal detection threshold, and the binary hard mask realizes deterministic isolation of abnormal components in the activation value while keeping normal components unchanged; and performing hierarchical response compression, that is, performing learnable dimension-level linear compression on the isolated abnormal components, re-integrating the compressed abnormal components and the normal components through gate fusion, and outputting an activation tensor after suppressing the abnormality. The application can prevent rank collapse and gradient oscillation, and enhance the robustness of long context task execution.

This invention provides a method and system for monitoring and adaptively calibrating the blade condition of a hawthorn slicer based on multi-sensor fusion. It collects spindle motor current signals, blade holder vibration acceleration signals, and slice images, and extracts feature values. When the vibration feature value changes abruptly and the current feature value decreases simultaneously, blade breakage is detected, triggering an emergency stop. If breakage is not detected, the average current trend is monitored to generate a wear warning. Visual acquisition is initiated according to preset conditions to obtain the actual slice thickness. Using a pre-calibrated mapping relationship between slice thickness and current and vibration feature values, the threshold compensation amount is calculated in reverse based on the thickness deviation, adaptively correcting the detection threshold. This invention reduces the false blade breakage rate, achieves proactive wear warning, compensates for sensor drift, and ensures a high slice yield.

Providing impact detection programs and the like that can accurately detect impacts to a vehicle using portable devices brought into the vehicle's interior. [Solution] The smartphone 10 and application server 50 execute an impact detection program in their respective processing units 11 and 51, and detect an impact on the vehicle Am based on the acceleration measured by the smartphone 10 brought into the vehicle cabin. The smartphone 10 acquires measured values ​​of the acceleration in three axes, which are measured repeatedly. The application server 50 prepares composite quantiles, which are multiple quantiles at a predetermined time, for the composite acceleration obtained by combining the accelerations in three axes. The application server 50 detects an impact based on a comparison between each composite quantile and a detection threshold associated with each of the multiple composite quantiles.

This invention relates to the field of passive underwater target identification technology, and in particular to a propeller noise cyclic coherent demodulation method based on an improved real-number genetic algorithm. Based on cyclic stationarity analysis theory, an optimal coherent weighted quadratic envelope spectrum is constructed by replacing the fixed integration frequency band with a weighting function, and an adaptive detection threshold is constructed based on the false alarm probability. The observed spectrum is modeled as the sum of harmonics and noise, and a sparse non-negative optimization problem is solved to extract the harmonics. Using the total harmonic intensity of the envelope spectrum as the cost function, an improved real-number genetic algorithm is used to optimize and search for the optimal weighting coefficient vector. The demodulated spectrum is obtained by coherently weighting and integrating the normalized cyclic modulation spectrum. This invention achieves autonomous allocation of weighting coefficients and adaptive selection of demodulation frequency bands, effectively improving the signal-to-noise ratio gain of cyclic demodulation, and is applicable to passive target identification in data backtracking and review systems.

The application provides a radar monitoring and evaluation method for a shale gas well site safety situation, comprising the following steps: collecting original vibration amplitude data, initial clutter power spectrum information and target Doppler shift signals, and obtaining filtered vibration amplitude data, purified clutter spectrum data and refined Doppler shift values after filtering; performing clustering analysis on the current vibration intensity level of the energy distribution characteristics of different frequency bands in the overlapping interval range, evaluating the interference level caused by the target detection of the clutter power spectrum broadening amount, and obtaining the interference intensity grading result; performing boundary feature extraction and trend fitting processing on the interference intensity grading result and the frequency spectrum overlapping boundary by using a support vector machine, and identifying a dynamic threshold adjustment amplitude to determine an adjusted detection threshold value.

The present disclosure provides a cleaning robot and a movement control method thereof. Based on the method, during travel of the cleaning robot, three-dimensional information of an obstacle is acquired through a sensor system. When the obstacle moves within the detection range of the sensor system, when a distance between the obstacle and the cleaning robot along a central axis of the cleaning robot is less than a first preset detection threshold, and when a maximum value among included angles between connection lines constituted by a first reference point of the cleaning robot and second reference points of the obstacle and a current traveling direction of the cleaning robot is greater than a third preset detection threshold, an action away from the obstacle is intelligently performed.

A multiple-pulses-in-air (MPiA) laser scanning system configured for measuring the time-of-flight of laser pulses, wherein the computing unit is configured for triggering the transmitter to transmit the send pulses with a varying pulse frequency, wherein the transmitter is configured to have a dead mode, wherein in the dead mode an increase in pulse frequency is effected, wherein in the course of increasing the pulse frequency, a reduction in pulse energy per individual send pulse of the send pulses is provided such that the pulse energy of the resulting return pulses is reduced such that it is below a detection threshold, and a measurement mode, wherein in the measurement mode a reduction in pulse frequency compared to the dead mode is effected.

A method for dynamically setting a detection threshold of a vehicle contact sensor of a vehicle's force management system, comprising detecting an object outside the vehicle via one or more vehicle environment sensors and then determining, based at least partially on data from the one or more vehicle environment sensors, the current path of the vehicle, the current trajectory of the object, and whether the current path of the vehicle and the current trajectory of the object intersect. The method assigns a categorization of the predicted contact of the vehicle with the external object based at least partially on the data from the one or more vehicle environment sensors and, based on the categorization, sets a trigger threshold of the vehicle contact sensor at which an energy-absorbing component is activated.

This invention relates to the field of boiler pipeline leakage detection technology, specifically to a boiler pipeline leakage detection method and system. The boiler pipeline leakage detection method includes the following steps: real-time acquisition and preprocessing of initial signal data from multiple sensors to generate standard signal data from multiple sensors; designing a separate dual encoder and a symmetric decoder based on the standard signal data from multiple sensors, and constructing a TCAE model; automatically iteratively optimizing the hyperparameters of the TCAE model using Bayesian optimization to obtain an Opt-TCAE model; confirming an adaptive leakage detection threshold; inputting the standard signal data from multiple sensors into the Opt-TCAE model, and calculating the reconstruction error of the standard signal data from multiple sensors in real time, wherein: if the reconstruction error is greater than the adaptive leakage detection threshold and the duration is ≥5 minutes, a leak is determined.

A bearing device includes a bearing and a vibration sensor for measuring the vibration of the bearing, the bearing having an inner ring and an outer ring capable of rotating concentrically relative to each other. The bearing device further includes: - a first determining component (10) for determining the value of at least a first statistical parameter equal to the Hjorth complexity parameters (P1, P2, P3, P4, P5, P6); - a modeling component (11) for modeling the value of at least the first statistical parameter using a normal distribution; - a second determining component (12) for determining a detection threshold (Sd) at least based on a normal distribution during a training period; - a third determining component (13) for determining a score (Sc) during normal operation of the machine (1) at least based on the first value of the Hjorth complexity parameter and the normal distribution modeled on the value of the first statistical parameter; and a comparison component (14) for determining a failure of the bearing (4) based on the detection threshold (Sd) and the score (Sc).

The application discloses a PCB defect detection method and device, computer equipment and a storage medium. The method comprises the following steps: collecting resistance data of multiple test points in a to-be-detected area of a PCB to form a resistance data set; sorting the resistance data set and determining the median of the resistance data set as a data median Med; calculating the absolute deviation of each data in the resistance data set from the data median Med to form an absolute deviation set; sorting the absolute deviation set and determining the median of the absolute deviation set as an absolute deviation median MAD; constructing a robust detection threshold for defect determination; and comparing each resistance data in the resistance data set with the robust detection threshold to obtain a PCB defect detection result. The method of the application can realize PCB resistance defect detection with high anti-interference, high detection rate and low false alarm rate, and improve the quality control level of the PCB manufacturing process.

A method of adjusting a detection threshold in a frequency-modulated continuous wave (FMCW) light detection and ranging (LIDAR) system includes determining a first confidence threshold with respect to a confidence metric for detecting a first target within a range of frequencies corresponding to a field of view of the LIDAR system, determining a subset of frequencies within the range of frequencies for detecting a second target, wherein a frequency peak associated with the second target has a confidence metric value lower than the first confidence threshold, and adjusting the first confidence threshold to a second confidence threshold at the subset of frequencies for detecting the second target.

The invention discloses a power system fault detection method and device, equipment and a medium, and relates to the field of power fault detection, and the method comprises the steps: constructing a to-be-solved target state space based on the historical operation data of a target power system; performing low-rank decomposition on a data matrix constructed based on the historical operation data, solving the target state space to be solved according to an obtained decomposition result, and obtaining a target state space matrix of the target power system; determining a current system residual error based on a target Kalman filter constructed through the target state space matrix and the collected current input and output data, and splicing the current system residual error and a historical system residual error to obtain an augmented residual error vector; and calculating chi-square statistical magnitude based on the augmented residual vector, determining a target detection threshold according to a preset false alarm rate and the current degree of freedom, and if the chi-square statistical magnitude is continuously greater than the target detection threshold, performing power system fault alarm. Therefore, rapid and reliable power system fault detection can be realized.

The application provides a fault detection method for a European standard wall penetrating terminal, and belongs to the technical field of fault detection. In the application, a partial discharge characteristic component is input into an artificial intelligence model based on a sparse self-encoder to output a partial discharge characteristic vector and calculate an insulation degradation index. Meanwhile, a contact state observer based on a virtual impedance principle is constructed to identify contact resistance parameters online by using a recursive least square method. A double-layer game optimization model is used to dynamically adjust a detection threshold and a sampling strategy, and a power grid fluctuation robustness evaluation algorithm based on topology dynamics is started to identify energy weak points. Finally, a fault confidence adjustment factor is calculated according to the insulation degradation index, the contact resistance parameter and the system stability to realize a graded response, so that the technical problem that the European standard wall penetrating terminal is difficult to realize the cooperative real-time detection of insulation degradation and contact deterioration in a high-power charging environment is solved.

This invention provides an AI-based multimodal network information operation and maintenance system, belonging to the field of network information operation and maintenance technology. The system collects heterogeneous operation and maintenance data such as network traffic, device status, audio alarms, and thermal imaging through a data acquisition unit. It utilizes a multimodal feature extraction module to extract various features in parallel, forming a unified multidimensional feature vector. A hierarchical anomaly detection architecture is established, deploying lightweight and deep anomaly detection models at the edge and cloud respectively. A multimodal feature fusion model intelligently fuses different modal features through an attention mechanism. An adaptive threshold dynamically adjusts the Bayesian algorithm to optimize the detection threshold in real time based on network status. A multimodal fusion decision engine is constructed to weight and fuse anomaly detection results and dynamically adjust modal weights, thus solving the technical problem of insufficient accuracy in multimodal operation and maintenance data fusion processing.

This invention discloses a multi-user liveness detection method based on wearable RFID tags, belonging to the field of detection technology. The key technical solution includes the following steps: collecting raw signal data of RFID tags worn by users within an area using an RFID reader; the raw signal data includes Received Signal Strength Indication (RSSI) and phase; preprocessing the raw signal data to obtain processed data; filtering the RSSI differential value and the processed phase to suppress noise; calculating a comprehensive feature value for each RFID tag based on the processed data to characterize signal fluctuations; the comprehensive feature value consists of a first feature component and a second feature component; using the maximum comprehensive feature value indicating no liveness as the detection threshold for determining the user's liveness; the effect is to quickly obtain the user's liveness status for the current period.

The application discloses a method for testing deception interference of a combined navigation system, and belongs to the field of combined navigation system testing. The method is used for solving the problem that existing technologies are prone to misjudgment under the condition that a combined navigation system is subjected to GNSS deception interference testing under the motion of a carrier. The method comprises the following steps: acquiring GNSS receiver observation data and INS inertial data; calculating the average heading angle of an INS sliding window; differentially solving a GNSS heading angle; independently solving an INS predicted position and aligning the position with a GNSS position in time, constructing a position deviation sequence, and extracting the change slope and variance of the position deviation sequence; calculating a heading deviation sliding window sequence of the GNSS heading angle and the average heading angle of the INS sliding window, and extracting the change slope and variance of the heading deviation sliding window sequence; and performing joint consistency judgment based on the detection threshold of the position and the heading set based on historical normal navigation data, and determining that the combined navigation system is subjected to GNSS deception interference when the preset condition is not met. The method is mainly used for combined navigation systems of various dynamic platforms such as vehicles and unmanned systems.

This invention discloses an on-chip spectrum sensing automatic gain control method and system for FPGA. The method includes: performing spectrum analysis on the received baseband signal, dynamically adjusting the detection threshold using a unit average constant false alarm rate algorithm, and calculating the minimum signal-to-noise ratio (SNR) required for signal detection based on the detection threshold; performing amplitude detection, logarithmic conversion, and filtering on the baseband signal based on the minimum SNR, and generating an attenuation control signal output to a digital attenuator at the RF front end to adjust the signal gain; wherein, the minimum SNR is dynamically updated based on the gain-adjusted baseband signal, and the attenuation is adjusted in real time based on the updated minimum SNR to stabilize the power of the input signal within the optimal dynamic range required for signal detection, forming a closed-loop feedback control.

This application discloses a method and system for detecting abnormal output impedance of a power amplifier, relating to the field of output impedance abnormality detection technology. This application utilizes sliding window technology to track and analyze the changing trends of power amplifier operating parameters in real time, enabling timely detection of threshold range adjustment needs and generating corresponding instructions (including range expansion and range contraction instructions). This provides a foundation for realizing dynamic abnormality detection of power amplifiers. By calculating the standard deviations of load impedance, temperature, and operating frequency, and based on the results of the standard deviations, the abnormality detection threshold range can be precisely adjusted to ensure detection accuracy. Each time the sliding window slides, the system updates the abnormality range, ensuring that the abnormality detection threshold can respond in real time to dynamic changes in parameters. This adaptive mechanism is suitable for various working scenarios, especially for equipment with high dynamic performance requirements.

This invention relates to the field of noise monitoring and intelligent analysis technology, and discloses a method and system for intelligent noise source tracing based on human-machine perception alignment. The method includes acquiring acoustic signals and public perception data from the acoustic environment of a social living area; performing time warping on both using a cross-similarity matrix to obtain a perception feature set; analyzing the perception feature set to obtain a deviation distribution and quantifying it into a global deviation correction coefficient; using this coefficient to calibrate the acoustic signal to obtain a signal flow; when the coefficient exceeds a threshold, reconstructing the spatial gain distribution of the microphone array based on the signal flow to locate the noise; combining the location coordinates with image data to identify the type of abnormal event; if it belongs to multi-source interference, predicting the noise intensity trend and generating a real-time parameter optimization scheme; adjusting the detection threshold accordingly and combining the signal flow to pinpoint the final noise source. This method can solve the problems of misalignment between early warning strategies and public perception, as well as the low reliability of noise source tracing.

An aerosol provision device is configured to receive at least a portion of an article comprising aerosol generating material, the aerosol provision device comprising an article sensor and a processor, wherein the article sensor comprises a receiver for receiving a receiver signal from the article, the receiver signal indicative of article information, wherein the processor is configured to: determine whether the receiver signal has a receiver signal strength below a detection threshold; and in response to the receiver signal strength being below the detection threshold, modify an operation parameter of the article sensor to increase the receiver signal strength.

This application relates to the field of computer technology, providing a method and computing device for recovering corrupted metadata. The method includes: detecting the activity status of a container runtime; when the activity status is inactive, creating a temporary test file in the working directory of the container runtime; if the temporary test file is successfully created, reading the runtime log of the container runtime; if the runtime log contains any preset keyword, determining that the metadata of the container runtime is corrupted, and updating the corruption detection count of the container runtime; when the corruption detection count is greater than or equal to a detection threshold, restoring the metadata file of the container runtime. Based on this solution, a host device equipped with a container orchestration platform can periodically detect and identify whether a fault involves the metadata file of the container runtime, achieving accurate identification and controllable repair of corrupted metadata, and improving the response speed for handling corrupted metadata files.

The application discloses a driving track determination method and device of a vehicle, a processor, an electronic device and the vehicle. The method comprises the following steps: acquiring space-time perception information of a rear vehicle relative to the vehicle; detecting the rear vehicle based on the space-time perception information to obtain a detection result; in response to the detection result that the rear vehicle is in a risk area, adjusting an initial detection threshold of a radar detection device in the vehicle based on the space-time perception information to obtain a target detection threshold, wherein the risk area is used to represent an area in which echo power of the radar detection device is attenuated, and the target detection threshold is lower than the initial detection threshold; and determining a driving track of the rear vehicle in response to the space-time perception information meeting the target detection threshold. The application solves the technical problem of low determination accuracy of the driving track of the rear vehicle.

The application provides a radar active jamming time-frequency statistical detection and suppression method and device, and relates to the technical field of radar anti-jamming. The method comprises the following steps: sequentially performing short-time Fourier transform, modulus value taking and logarithm processing on a radar echo signal to determine a time-frequency pixel matrix, so as to realize time-frequency analysis of the radar echo signal in each PRT; and performing positioning processing and time-domain reconstruction on the jamming signal in the time-frequency domain according to each estimated parameter, false alarm rate and detection threshold, so that the jamming signal can be determined for jamming suppression, without needing to search for a jamming area and accurately estimate parameters on a range-Doppler plane, so that jamming suppression can be performed when a range-Doppler plane target and jamming are highly overlapped; and the jamming signal can be sequentially subjected to positioning processing and time-domain reconstruction in the time-frequency domain, so that the division of a jamming signal area is realized, the use of a neural network for distinguishing the jamming signal is avoided, a large amount of jamming data set is not needed, and the method is suitable for engineering application.

The present application relates to a kind of threshold configuration and constellation fault determination method, device, equipment and medium, the method includes: including: the data set of the all detection threshold values set in log according to the division of fault type, after all detection threshold values are weighted average, corresponding type fault history detection threshold baseline is generated;According to the data set in log according to the division of fault type, the short-term fluctuation trend of detection threshold baseline is predicted;According to the data set in log according to the division of fault type and history detection threshold baseline, the medium-term fluctuation trend of detection threshold baseline, long-term fluctuation trend is predicted;According to the short-term fluctuation trend of detection threshold baseline and short-term alarm quantity trend change, whether short-term threshold baseline is changed is determined;According to the medium-term fluctuation trend of detection threshold baseline and medium-term alarm quantity trend change, long-term alarm quantity trend change, whether medium-term threshold baseline, long-term threshold baseline is changed is determined;According to the threshold baseline after changing, whether constellation is fault is judged.

This invention relates to the field of radio frequency (RF) communication technology, and discloses a control method, apparatus, device, storage medium, and card reader for RF communication. The RF communication control method includes: monitoring changes in a sensing signal at preset time intervals; acquiring sensing signal strength information when the sensing signal changes; analyzing the signal quality of the sensing signal based on the sensing signal strength information to obtain a target signal-to-noise ratio (SNR); determining a polling detection threshold based on the target SNR; acquiring a real-time sensing signal; determining a target judgment index based on the real-time sensing signal and a preset reference value; comparing the target judgment index with the polling detection threshold; and polling the electronic tag based on the target comparison result to control the RF communication. This invention improves the accuracy of triggering polling by determining the polling detection threshold through the SNR.

The application discloses a kind of robust target detection methods based on covariance matrix reconstruction under interference condition: first, training sample is used to form sampling covariance matrix;Then, the noise subspace is constructed using the sampling covariance matrix;Then, according to the angle information of target, the angle range to be used for covariance matrix reconstruction is determined, and the noise subspace is used to reconstruct the covariance matrix;Further, the reconstructed covariance matrix is used to construct detection statistics;According to the false alarm probability, the detection threshold is determined, and the size between detection statistics and detection threshold is compared, if detection statistics is greater than detection threshold, it is judged that target exists, otherwise, it is judged that target does not exist.The robust target detection method based on covariance matrix reconstruction under interference condition designed by the application has robust characteristics for interference angle mismatch, and when the interference angle information is inaccurate, the target can still be detected with a large detection probability.