12 results about "Pulse flow" patented technology

The embodiment of the present application discloses a method for inhibiting integrated circuit from static damage, comprising: providing a first chip, and arranging an integrated circuit on the first chip; providing a second chip, and arranging an electrostatic protection circuit on the second chip, the electrostatic protection circuit comprising: a first output port, a second output port and a first resistor; wherein, when an electrostatic pulse flows through the first resistor, the resistance of the first resistor increases; connecting the first output port with an output port of the integrated circuit, so that the electrostatic protection circuit can inhibit the integrated circuit from static damage. Moreover, the electrostatic protection circuit is arranged on the second chip, and the integrated circuit is arranged on the first chip, so that the integrated circuit and the electrostatic protection circuit are arranged on different chips, avoiding the problem that the integrated circuit cannot be designed for electrostatic protection due to the limitation of the base material of the chip on which the integrated circuit is arranged, and realizing the electrostatic protection of the integrated circuit of new material.

The application discloses a mouse operation precision improving method based on AI posture reconstruction and relates to the technical fields of hand capturing and mouse. An asynchronous space-time pulse flow sequence carrying ultrahigh time resolution is generated; a continuous-state hand high-dimensional dynamics hidden space representation is formed; a Laplace domain continuous motion trajectory function is solved; a full-link system dynamic forward-looking compensation time is calculated; zero-delay hand motion state prior parameters eliminating hardware physical delay are deduced; sub-millisecond continuous mouse control signals are generated; and stepless smooth state refreshing and zero-delay interactive driving of a mouse pointer are completed. The application significantly improves the accuracy of hand super-speed motion trajectory extraction and the space-time calculation efficiency of the system, and the trajectory fidelity and anti-interference robustness in an ultrahigh-frequency interaction scene are qualitatively improved compared with traditional interpolation algorithms.

The invention relates to the technical field of educational informatization, and discloses a smart classroom interactive behavior real-time supervision method fused with attention recognition, and the method comprises the steps: collecting classroom video streams, carrying out the time domain differential coding through an integral issuing neuron model, and generating teacher and student end pulse streams; constructing a self-adaptive confidence mask based on the visual pulse stream of the student side, performing gating repair and interpolation on the original video signal, and reconstructing a clean physiological blood volume pulse wave signal; extracting a student physiological feature vector, and calculating a dynamic lag order parameter according to the teacher end pulse stream; and performing time-varying Granger causality test on the teacher end pulse stream and the student physiological feature vector by using the parameters, and outputting an interactive behavior judgment result. By quantifying the causal association between the teaching stimulation and the physiological response of the student, the problems that the false attention state is difficult to distinguish and the non-contact physiological signal is susceptible to action interference in the prior art are solved, and the accuracy and robustness of classroom supervision are improved.

The application discloses a kind of light weight multi-focus pulse fusion reconstruction methods based on compressed sensing, constructs light weight multi-focus pulse fusion reconstruction network, contains multi-focus pulse input module, feature fusion module, sparse coding optimization module, image reconstruction module;Multi-focus pulse input module handles two multi-focus pulse flow data, generates single first fusion feature map and single second fusion feature map, feature fusion module extracts and fuses two fusion feature maps, obtains dictionary matrix and long exposure image, based on dictionary matrix and long exposure image, by iteratively sparse coding optimization module solution, obtain optimal sparse coding, image reconstruction module utilizes dictionary transpose matrix and optimal sparse coding, reconstructs output clear image richer in detail, overall loss function improves the approximation degree of reconstructed image and real image.The application effectively solves the problem of pulse camera imaging depth of field blur.

The application discloses a turbine shaft torsional vibration in-situ detection method under an aero-engine whole machine state, relates to the field of aerospace engines, and comprises the following steps: converting the rotating speed fluctuation generated when the turbine shaft of the aero-engine occurs torsional vibration into the change on the triggered pulse time sequence when the rotor blade reaches the optical sensor, and collecting the blade pulse flow signal through the optical sensor; wherein the optical sensor is fixedly connected with the bore probe structure assembly of the aero-engine; the blade pulse flow signal is high-frequency sampled by using high sampling rate dynamic data sampling, and the rotating speed time domain signal is obtained; the rotating speed time domain signal is subjected to preliminary analysis of turbine shaft torsional vibration, and the torsional angle result of the turbine shaft is determined; and the torsional angle result of the turbine shaft is converted into the torsional vibration result of the turbine shaft based on the strain relationship function between the shaft torsional angle and the torsional vibration. The application can realize efficient and accurate detection of the turbine shaft torsional vibration under the aero-engine whole machine state.

The invention discloses an intelligent hydrostatic level communicating pipe device for deformation monitoring of a super high-rise structure, and relates to the technical field of deformation monitoring of the super high-rise structure. A three-dimensional position, an equivalent diameter, morphological characteristics and a moving track of a bubble are obtained through a bubble multi-dimensional characteristic monitoring module; the bubble risk grade evaluation and dredging decision module realizes bubble risk grade division and dredging strategy matching based on a digital twinborn model and an interference weight evaluation model; the differential bubble dredging execution module is used for regulating and controlling small-size free bubbles by adopting a low-pressure pulse flow field, stripping large-size adhesion bubbles through high-frequency micro-vibration and then pushing the large-size adhesion bubbles to an exhaust cavity; the deformation monitoring-bubble dredging cooperative regulation and control module ensures that dredging operation does not influence the structural deformation monitoring stability through time sequence data comparison; the dredging effect verification and strategy iteration optimization module achieves strategy dynamic optimization through secondary scanning and quantitative evaluation, and the problem of liquid level measurement distortion caused by bubble retention in the super high-rise long-distance communicating pipe is effectively solved.

Embodiments of the present application provide an eye movement tracking method and device based on pulse data, equipment and a storage medium. The method comprises determining a tracking feature at a starting time according to a labeled pupil position of the target pupil at the starting time of the target pulse stream and a target visual feature corresponding to the starting time; for each time, determining a tracking feature at the time and a predicted pupil position at the time according to a target visual feature corresponding to the time, a tracking feature at a last adjacent time, a predicted pupil position at the last adjacent time, and a labeled pupil position; and obtaining an eye movement tracking result based on the predicted pupil position at each time. With the help of an end-to-end network architecture and a timing iterative updating mechanism, the tracking drift problem caused by high-speed eye movement is effectively solved. At the same time, the slow changes in the appearance of the pupil during the tracking process are adapted, and the efficiency, stability and accuracy of the tracking are significantly improved by relying on the historical state to smooth the predicted trajectory.

The application discloses a neuromorphic camera denoising method for general lighting conditions, and belongs to the field of video denoising. According to the relationship between the intensity of the pulse stream noise and the scene lighting, the continuous space-time pulse stream is converted into the picture feature representation under a specific window, and the input of the traditional neural network is adapted; the dark current noise and the response inconsistency noise are removed through the calibrated dark current matrix and the response inconsistency matrix; the image features are converted into the frequency domain features through the two-dimensional discrete wavelet transform, the low-cost noise removal is realized by using the frequency domain noise separation characteristics; the shot noise and the clipping noise are removed by fusing the frequency domain features of adjacent windows; the residual noise on the frequency domain is thoroughly removed through the deep self-attention network; finally, in order to inhibit the excessive smoothing caused by the denoising, the fusion features on the frequency domain are used to refine the denoising result. The application is suitable for the field of video denoising, is used for reconstructing a clean image, reducing the noise in the image and reducing the influence of the noise on the downstream task.

The present application relates to a kind of solvent-resistant label stretch forming device in the field of label processing device, heating is carried out when the first heating roller is stretched in label base film, so that label base film is stretched more uniformly, the surface temperature of the first heating roller is monitored using contact temperature sensor, the temperature of heat medium injected into multi-channel rotary joint is controlled, and the temperature of heat medium injected into feeding inlet and flowed out of discharging outlet is detected respectively, to judge the heat exchange efficiency of heat medium in the first heating roller, further judge whether scale exists in the first heating roller, and in the early stage of scale formation, use pulse flow to break, finally use collection unit to capture and concentrate scale fragments, reduce the risk of multi-channel rotary joint blockage, discard the complex scale cleaning device set in the heating roller in prior art, increase its reliability, reduce the maintenance cost of heating roller.

The application discloses a brushless pump intelligent control method and system based on a single-chip microcomputer. The method comprises the following steps: in the same single-chip microcomputer, collecting a pressure signal output by an analog pressure sensor through an ADC module at a sampling rate of no less than 500 Hz, collecting a temperature signal through the ADC module, and collecting a flow signal output by a pulse flow sensor through a timer capture unit; performing cascade digital filtering on the pressure signal, wherein the cascade digital filtering comprises median filtering and sliding average filtering; performing PID calculation on the filtered pressure signal as a main feedback quantity to output a speed correction quantity; switching between multiple operation modes according to the temperature signal and the flow signal; driving the brushless pump according to the speed correction quantity through FOC vector control; and performing a communication task with an external device. The application realizes the coordinated operation of high real-time pressure closed-loop control and the communication task through the single-chip microcomputer, and helps to eliminate the false triggering caused by the sensor signal fluctuation.

This invention discloses a spiking neural network (SNN) mapping method and system for congestion awareness on neuromorphic hardware, belonging to the field of neuromorphic computing technology. The method includes: mapping neurons in a spiking neural network model to physical cores of neuromorphic hardware to form an initial mapping scheme; iterating through congestion hotspots and updating the mapping scheme in each iteration to alleviate network communication congestion on the hardware; calculating the distribution of congestion hotspot regions composed of physical cores under the current mapping scheme based on pulse flow evaluation between physical core pairs; for each congestion hotspot region, selecting multiple physical cores with high pulse flow contribution as hub cores to be swapped; determining the optimal swap candidate core for each hub core based on a heuristic swapping decision mechanism; and swapping hub cores with the optimal swap candidate cores to update the mapping scheme. This promotes the efficient deployment of SNNs on neuromorphic hardware and provides key technical support for high-energy-efficiency, low-latency neuromorphic computing.

The invention discloses a radar signal pre-sorting method based on simultaneous multiple beams, which comprises the following steps of: reading pulse description words in a time slice, identifying N beams of which the frequencies, the pulse widths and the time differences of arrival are smaller than a set threshold value as a group of simultaneous multiple beams, simultaneously completing pulse width division and storing until all current pulses are traversed; performing angle measurement on each group of obtained simultaneous multi-beams by adopting a three-point power centroid method, compensating a beam pointing error caused by frequency deviation of a radar and a radiation source to obtain a unique accurate direction of the beams in the group, taking a beam direction with the maximum amplitude as an output beam, and replacing the direction with the accurate direction to obtain an accurate direction of the beams in the group; keeping amplitude information of other beams in multiple beams at the same time to an output beam, and eliminating other beams to complete false beam suppression; and on the basis of the output pulse flow, interval division based on system or precision requirements is carried out in a time domain, a frequency domain and a space domain in sequence, and pre-sorting is completed. According to the invention, high-precision angle measurement, false wave beam and distorted wave beam removal and pre-sorting are synchronously completed.