64 results about "Threshold curve" patented technology

The invention discloses a method for vibration monitoring and fault diagnosis of a wind generating set, which comprises the following steps: A. presetting a monitoring point, arranging a sensor is arranged, outputting an original vibration signal through the sensor, and finishing the storage of the signal through data acquisition equipment; B. sending feature value data to a receiving module through a component bus module and storing the feature value data in a data center, or obtaining the original vibration signal and store the original vibration signal in the data center through the receiving module which avoids the component bus module; C. presetting a threshold curve in the data center, the feature value data grows continuously, and the receiving module performings fault early warning when the feature value data exceeding exceed a threshold and sends sending status early warning when the feature value data continuously exceed the threshold; and D. establishing a fault diagnosis module is established in an analysis module, presetting a fault diagnosis threshold is preset, the analysis module obtaining envelope signal data from the data center by the analysis module, calculating a comprehensive fault evaluation value through the fault diagnosis module, and comparing the comprehensive fault evaluation value with the fault diagnosis threshold, thereby performing fault alarming.

The present invention discloses a real-time detection system based on hybrid background modeling for detecting parked vehicles along the side of a road. The hybrid background model consists of three components: 1) a scene background model, 2) a computed restricted area map, and 3) a dynamic threshold curve for vehicles. By exploiting the motion information of normal activity in the scene, we propose a hybrid background model that determines the location of the road, estimates the roadside and generates the adaptive threshold of the vehicle size. The system triggers a notification when a large vehicle-like foreground object has been stationary for more than a pre-set number of video frames (or time). The present invention is tested on the AVSS 2007 PV dataset. The results are satisfactory compared to other state-of-the-art methods.

The invention relates to a loudspeaker sound detecting method, which comprises: the computer of sound card collects the time wave; the computer distinguishes the signal head of said wave to judge the initial position of one section of effective signal; computer processes Daub4 wavelet discompose on attained signal and reset preset parameter to 0 to process the wavelet rebuilding, to attain the rebuilding signal; the computer transforms the display method of attained result, to build the special sound curve; and the computer compares the special sound curve with the preset threshold curve, when there is one point of special sound curve is higher than the threshold of said point, the sound is detected as disqualified. The invention has high reliability, wider application and it can be used on online test.

The invention discloses an intelligent cat-eye system triggering shooting based on luminous intensity and distance detection. Based on results of a luminous intensity analysis module and a distance measurement module, an external processor controls the activation of a camera. When the change curve of environment luminous intensity data exceeds a preset threshold curve, and the distance D between a target and a laser emitter measured by the distance measurement module is no greater than a preset distance threshold, the external processor outputs a trigger signal to activate the camera. When a visitor is close to a door, the intelligent cat-eye system can automatically start a doorbell calling system, and is intelligent and convenient to use. An emission mode having laser emission time less than 0.1 S, interval time greater than 0.1 S, and a period about 0.2 S can prevent eyes from being damaged by laser. Based on the determination of luminous intensity change and object distances, the intelligent cat-eye system activates a camera when luminous intensity drops suddenly and a target distance is closer, thereby avoiding misjudgement caused by going out of aisle lamps (illumination intensity can drop suddenly), and obtaining higher determination accuracy than a control method which only depends on luminous intensity detection.

The invention relates to an audio test method, the method comprises the following steps: a to-be-tested terminal and a test device are placed in a sealed box; the test terminal sends a first test signal to the to-be-tested terminal so as to enable a to-be-tested audio output device of the to-be-tested terminal to output a first audio signal; the test device sends a first audio electric signal corresponding to the first audio signal to the test terminal; the test terminal outputs a first test curve according to the first audio electric signal and compares the first test curve with a preset first threshold curve so as to automatically judge whether the to-be-tested audio output device of the to-be-tested terminal is qualified. The invention also provides an audio test system. The audio testmethod and the audio test system provided by the invention directly perform audio test for the to-be-tested terminal and separately test the to-be-tested audio output device and a to-be-tested microphone of the to-be-tested terminal. Thus, reliability of test can be improved further, and stability of the to-be-tested terminal can be improved.

In an elevator device, an elevator car and a counterbalance weight are suspended by a suspending body. The suspending body is wrapped over a drive sheave of a hoist. A drive sheave-side detecting means generates a signal corresponding to a rotation of the drive sheave. A suspending body-side detecting means generates a signal corresponding to a movement of the elevator car or the suspending body.A condition monitoring device performs condition monitoring by comparing the difference between a position signal, which is sent from the drive sheave-side detecting means and occurs from the start to the stop of the elevator car, and a position signal, which is sent from the suspending body-side detecting means, with a predetermined threshold profile.

The invention belongs to the field of traffic management and control, and discloses a real-time variable lane control method and a control system. The real-time variable lane control method comprisesthe steps of establishing a variable guiding lane equal-vehicle average-delay model, solving a threshold condition for changing the function attribute according to the variable guiding lane equal-vehicle average-delay model, and drawing a threshold curve map; when a point which is obtained through combining an inlet lane straight driving and left-turn flow value is above the threshold curve, changing the lane function, and otherwise, keeping the original function attribute unchanged, and performing preliminary determining for whether the lane function changes. According to variable guiding lane technology, through dynamically adjusting lane function conversion, a defect of insufficient utilization rate of a left-turn lane or a right-turn lane caused by fixed turn lanes at a crossing can besettled. When the left-turn and straight driving traffic flow of the inlet lane reaches a certain threshold, the function attribute of the variable guiding lane should be changed, thereby realizing apurpose of optimized matching between the lane function and the turning unbalanced traffic flow.

A gas turbine, computer software and a method for controlling an operating point of the gas turbine that includes a compressor, a combustor and at least a turbine is provided. The method comprises: determining an exhaust pressure at an exhaust of the turbine; measuring a compressor pressure discharge at the compressor; determining a turbine pressure ratio based on the exhaust pressure and the compressor pressure discharge; calculating a primary to lean-lean mode transfer threshold reference curve as a function of the turbine pressure ratio, where the primary to lean-lean mode transfer threshold curve includes points at which an operation of the gas turbine is changed between a primary mode to a lean-lean mode; and controlling the gas turbine to change between the primary mode and the lean-lean mode.

The invention discloses a bird repellent device based on bird feature recognition. The device comprises a pan-tilt, a recognition device, a signal selector, a driving module and a controller, wherein a loudspeaker is installed on the pan-tilt, a radar device, an infrared transmitter and a camera are arranged on the loudspeaker, the recognition device outputs bird attribute information according to a photographed bird image, the signal selector outputs a feeling threshold curve peak frequency of the target bird and feeling threshold sound pressure corresponding to the peak frequency according to the input bird attribute information, the driving module is connected with the loudspeaker through a power amplifier, the driving module drives the loudspeaker according to the peak frequency and the feeling threshold sound pressure, and the controller is in communication connection with the recognition device, the radar device, the signal selector and the driving module separately, wherein the loudspeaker emitting sound wave frequency is equal to the peak frequency, and the sound pressure of emitted sound waves is not smaller than the feeling threshold sound pressure. By recognizing birds, specific sound wave transmitting frequency and transmitting power are provided, the bird repellent process has the specificity, and the bird repellent effect is improved.

The invention relates to a method for automatically detecting interictal spike waves of epilepsy. The method comprises the following steps: extracting, by an MEMD (multivariate empirical mode decomposition), a component in which the spike waves present from scalp electroencephalogram of a epilepsy patient, performing the signal envelop, solving a dynamic threshold curve, and positioning a possibleposition of the spike waves. The method has the advantages that the frequency band range of electroencephalogram (EEG) can be adaptively adjusted according to the electroencephalogram, and the presence position of the interictal spike wave of the epilepsy can be accurately positioned; a novel concept is provided for reducing the burden of the clinical doctor for recognizing the spike waves, and the interictal spike waves can be more accurately automatically detected; and by virtue of comparison, the comprehensive advantage of the method in the aspect of the sensitivity and error rate is better than a method based on a signal envelop distribution model, and a better spike wave detection result can be achieved.

The invention discloses a noise reduction processing method of an ultrasound medical image. The noise reduction processing method of the ultrasound medical image is characterized by comprising the following steps of calculating an edge intensity image E of a gray-scale ultrasound image I; calculating a mask image P in a smooth organization structure area of the gray-scale ultrasound image I; calculating an edge intensity threshold curve Kcurve of the gray-scale ultrasound image I along a depth direction of an ultrasonic probe scanning line; extracting a corresponding edge intensity threshold from the edge intensity threshold curve Kcurve and utilizing a diffusion coefficient function g to obtain a diffusion coefficient for any ith row image Ei of the edge intensity image E; and utilizing a diffusion equation (formula as follows, wherein all portions of the formula respectively stand for an image the gray-scale ultrasound image I evolving to time t to be formed into, an evolution time step length, a divergence operator and a gradient of the gray-scale ultrasound image I) to perform filtering processing on the image according to the generated diffusion coefficient. The noise reduction processing method of the ultrasound medical image has the advantages of enabling the ultrasound medical image to be capable of utilizing different edge intensity thresholds along the depth direction of probe scanning and achieving the effect of speckle noise suppression optimization.

The invention provides a howling suppression method and device, a loudspeaker and a sound amplifying system. The method comprises the following steps: receiving a first signal, wherein the first signal is a sound signal acquired by a microphone within a first time period; acquiring the frequency spectrum of the first signal, determining a first hearing threshold curve according to the frequency spectrum of the first signal, and generating a first reference signal according to the first signal and the first hearing threshold curve, wherein the first reference signal and the first signal are played by the speaker together; receiving a second signal, wherein the second signal is a sound signal acquired by the microphone within a second time period after the first time period; and calculatinga correlation value of the second signal and the first reference signal, and determining whether to decrease a gain for the sound signals acquired by the microphone according to the correlation value.The sound intensity represented by the amplitude of the reference signal generated in the howling suppression method provided by the invention on any frequency point is lower than the sound intensityof the same frequency point on the hearing threshold curve, the reference signal is not perceived by audiences after being played together with the first signal, high fidelity is achieved, and the generation of howling is suppressed.

An apparatus for frequency-selective occupancy detection of a channel in a frequency band includes a detector for detecting a receive spectrum in the frequency band, a provider for providing node frequencies, a determiner for determining node level values, a determiner for determining the threshold curve and a comparer. The provider provides the node frequencies for a threshold curve with respect to a noise floor in the frequency band. The determiner determines the node level values at the node frequencies based on receive level values of the receive spectrum. The determiner determines the threshold curve in the frequency band based on the provided node frequencies and the node level values at the node frequencies. The comparer compares a current power of the receive level values in the channel with a reference power in the channel predetermined by the threshold curve to detect occupancy or non-occupancy of the channel depending on the comparison.

A method of allocating hydraulic fluid between actuators in a machine accepts a first command to provide a first requested fluid flow to a first actuator, wherein the first actuator is a bounded actuator such as a steering actuator, and a second command to provide a second requested fluid flow to a second actuator. The system adjusts the first and second commands to produce adjusted first and second commands corresponding to adjusted first and second fluid flows, such that the sum of the adjusted first and second fluid flows is less than or equal to a maximum available flow and the adjusted first fluid flow meets or exceeds the lesser of the first requested fluid flow and a threshold curve that is a function of engine speed or other variable.

The invention relates to a method for testing a laser damage threshold of an optical thin film. The method comprises the following steps: S1, obtaining laser energy density Fth of the optical thin film during single-pulse laser damage through test; S2, irradiating the optical thin film by the single-pulse laser, recording laser damage area boundary coordinates (xi, yi) when the laser damage boundary on the surface of the optical thin film is not increased any more, and meanwhile, recording the number n of times of single-pulse laser irradiation; S3, comparing Gaussian distribution of the laserenergy density with laser damage area distribution to obtain a laser damage threshold FN of the optical thin film during multi-pulse laser irradiation damage; and S4, continuously changing the incident laser energy density, and repeatedly executing the steps S2 and S3 to obtain a laser damage threshold curve of the optical thin film under the irradiation of femtosecond laser with different pulsenumbers. The method has the beneficial effects that the measurement accuracy of the laser damage threshold of the optical thin film under multi-pulse laser irradiation is guaranteed, and meanwhile, the test efficiency of the damage threshold of the optical thin film under multi-pulse laser irradiation is greatly improved.

A process for automatic control of a respirator changes between two phases of respiration by checking a detected respiratory breathing activity signal for a threshold criterion. If the threshold criterion is met, a changeover is made. A dynamic threshold curve, used for changing over into an inspiration phase, is held, after the beginning of the present expiration phase, at high values until the end of a selected inspiratory refractory period, and is then lowered monotonically, dropping to an inspiratory threshold target value at the expected point in time of the phase duration maximum of the present expiration phase. A dynamic threshold curve, which is held at low values, after the beginning of the present inspiration phase, until the end of a selected expiratory refractory period, is raised monotonically, increasing to an expiratory threshold target value at the expected point in time of the phase duration maximum of the present inspiration phase and is then changed over into the expiration phase when the breathing activity signal drops below the threshold curve for the expiration. The phase duration of the inspiration and expiration phases or breath duration (inspiration and expiration phases) is stored in each case, and the expected points in time of the phase duration maxima are derived from the distributions of the phase durations in relation to the beginning of the respective phase of respiration or from the distribution of the breath durations in relation to the beginning of the previous phase of respiration.

The invention relates to a blood vessel image recognition method and device, computer equipment and a storage medium. The method comprises the steps of obtaining initial boundary pixel points based ongray values of pixel points in an initial blood vessel image; Constructing a threshold curved surface by taking the initial boundary pixel point as a boundary condition based on a preset stable temperature field model; Identifying the initial blood vessel image according to the threshold curved surface to obtain an identified blood vessel image; Carrying out binarization processing on the identified blood vessel image; And identifying a target area in the blood vessel image after binarization processing, and obtaining a target blood vessel image according to the target area. According to themethod, threshold curved surfaces with different thresholds are selected based on different regions; The blood vessel image is recognized, the defect that the blood vessel image recognized through a single threshold value is large in segmentation error is avoided, it is guaranteed that the edges of the recognized blood vessel image and the background image are clear, therefore, the recognition error is reduced, and the blood vessel recognition accuracy is further improved.

The invention discloses an auditory non-invasive detection method based on composite nerve action potentials. The method comprises the following steps that a recording electrode is arranged at the lower rear quadrant of a human ear tympanic membrane and close to a tympanic ring, a stimulus sound is sent to an ear of a subject and a stimulus intensity threshold value capable of inducing a compositenerve action potential under the stimulation of each stimulation sound frequency in the test frequency range is extracted from a voltage signal near the tympanic membrane of the subject, a compositenerve action potential threshold curve is drawn to judge whether the hearing intensity sensitivity of the subject is normal, the stimulus sound and a suppression sound are sent to the ear of the subject, suppression sound intensity in the suppression sound frequency range is extracted from the voltage signal near the tympanic membrane of the subject, wherein the suppression sound intensity meets suppression rules at each suppression frequency; a composite nerve action potential tuning curve is drawn to judge whether the hearing frequency sensitivity of the subject is normal; the method realizes the hearing function detection of human ears in a non-invasive manner, and simultaneously realizes the comprehensive detection of a composite nerve action potential threshold curve and a composite nerve action potential tuning curve.

An apparatus for motion compensated noise reduction for input images is provided. The motion estimation and motion compensation circuit performs a motion estimation operation and a motion compensation operation on a current image and a previous image to obtain a first patch. The block matching operation circuit performs a block matching operation on the current image and the previous image to obtain a second patch. The motion detection circuit performs a motion detection operation on a target patch according to the first patch and the second patch to output a set of third patches. The current image includes the target patch. The noise reduction circuit performs a noise reduction operation on the set of third patches according to a threshold curve, so as to generate the target patch that the noise is reduced. A method for motion compensated noise reduction for input images is also provided.

The invention discloses an underwater acoustic signal design and feedback regulation method for large yellow croaker sound trapping, and relates to underwater acoustic signal design and parameter feedback regulation of large yellow croaker sound trapping. According to the signal design, linear frequency modulation signals are adopted according to the auditory threshold curve of the large yellow croaker and are processed through a designed reverse filter, so that emitted sound signals cover the auditory threshold range of the large yellow croaker, the sound power at the frequency band with thehighest auditory sensitivity of the large yellow croaker reaches the maximum, and the sound trapping effect of the large yellow croaker can be enhanced; in the implementation process, the transmittingtransducer is a cylindrical transducer, sound wave radiation in the horizontal all-direction is facilitated, and a uniform sound field is constructed. A feedback adjusting mechanism is added in the sound wave emission process; the amplitude adjustment coefficient and the passband proportion adjustment coefficient of the filter are calculated according to the larimichthys crocea ingestion sound received by the spherical hydrophone, and the waveform of the trapping sound wave can be flexibly changed according to a sea area background noise field through feedback adjustment, so that the reduction of the induction degree of the larimichthys crocea to the sound wave is prevented, and meanwhile, the emission power consumption can be reduced.

The invention relates to a method of evaluating a resolution index in naked eye stereoscopic display. The method comprises the following steps: S1, the brightness of a naked eye stereoscopic display device to play a pure white picture in a 3D mode is acquired and marked as E; and the illumination L1 of the naked eye stereoscopic display device to play the pure white picture in the 3D mode and the illumination L2 when a pure black picture is played are acquired; S2, the E, the L1 and the L2 are used for building a spatial frequency-contrast threshold curve C(f); S3, the naked eye stereoscopic display device displays a tested picture in the 3D mode, the image displayed by the naked eye stereoscopic display device is photographed, the photographed image is processed, and the contrast of the photographed image is thus acquired; S4, the spatial frequency of the photographed image is solved; and S5, according to the contrast of the photographed image and the spatial frequency, a corresponding point is determined in the f-C space, and a resolution loss condition is judged through the relation between the point and the spatial frequency-contrast threshold curve C(f), wherein the resolution loss conditions are as follows: if the point is lower than the threshold curve, resolution loss happens; and if the point is higher than the threshold curve, the resolution has no loss and is unchanged.

The embodiment of the invention provides a fitting method for evaluating a threshold curve of a black matrix and an evaluation method of the black matrix, relates to the technical field of display, and can improve the black matrix evaluation accuracy. A fitting method for evaluating a threshold curve of a black matrix comprises the following steps: obtaining a contrast sensitivity value by the widths of a plurality of shading strips extending along the length direction or the width direction of each sample and the widths of sub-pixels included in the black matrix of each sample; the shading strips have N widths, and N is greater than or equal to 2; the width of the shading strips in at least part of the samples is not completely equal to that of the shading strips in other samples; obtaining spatial frequency values corresponding to the samples under different observation distances; marking the visible or invisible condition of each sample dark line under the same ambient brightness and the corresponding observation distance in a coordinate system which takes the spatial frequency as an abscissa and takes the contrast sensitivity value as an ordinate; and fitting a threshold curve,so that the coordinate points of which the identifications are visible or invisible are respectively positioned on two sides of the threshold curve.

The invention discloses a polynomial-curve-fitting-based optimization method and device for a false alarm threshold of an IFF system. The method comprises: automatic recording and storage of identified target data within an airspace power range are carried out; data preprocessing is carried out; a polynomial model coefficient matrix is constructed; polynomial fitting solution is carried out; iteration with different fitting number of times and error analysis are carried out; a false alarm threshold curve is generated; sector inputting and integration of false alarm threshold curve are carriedout; and effect adjustment and comprehensive evaluation are carried out. According to the optimization method and device, false targets generated by environmental factors of positions can be suppressed effectively in real time; only a false alarm probability parameter value needs to be set during the whole process and no manual calculation is needed, so that the optimization efficiency is high; the operational requirements of quick identification and accurate positioning are met; and the manpower and material costs are lowered substantially.

The embodiment of the invention discloses an information processing method and device and a storage medium, and the method comprises the following steps: receiving a test request of ear print data, outputting a standard test sound corresponding to a target frequency, and enabling the ear print data to represent listening thresholds corresponding to different frequencies; detecting a touch action on the touch screen, and adjusting a test line of a standard test sound displayed on the touch screen according to the detected touch action, the test line being a graphical interaction interface for visually reminding a user; obtaining an audition threshold value corresponding to the adjusted test line according to the adjusted test line, and generating an audition threshold curve according to theaudition threshold value; correcting the auditory threshold curve according to a standard isophone curve, and obtaining a sound effect parameter corresponding to the auditory threshold curve; and obtaining at least one sound effect scheme corresponding to the ear print data according to the sound effect parameters. By adopting the method, the diversity of sound effect schemes can be enriched.

The embodiment of the present specification provides a network payment risk assessment method. By calculating the payment risk concentration value from the user's network payment behavior data, the payment decision threshold curve based on different scores is determined to evaluate the risk of the user's network payment behavior, so as to release or disturb the transaction process in time, so as to achieve the optimal balance and reduce the overall impact of the disturbed object.