Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

238results about How to "The calculation formula is simple" patented technology

Temperature control and crack prevention design calculating method used for lining concrete of door-opening-shaped section structure

The invention discloses a temperature control and crack prevention design calculating method used for lining concrete of a door-opening-shaped section structure. The method comprises the following steps of (1) determining a temperature control and crack prevention target, (2) calculating the permitted maximum temperature and (3) making a temperature control scheme, specifically, the maximum temperature inside the concrete is calculated, and on the premise that the worked-out maximum temperature is smaller than or equal to the permitted maximum temperature, the temperature control and crack prevention scheme is designed. The calculation formula of the method is simple, the influences of surrounding rock performance, the lining structure size, concrete strength, intro-opening air temperature, water-introducing cooling and water temperature thereof, pouring temperature and the like can be reasonably reflected, the permitted maximum temperature of pouring construction each month in the construction period of the lining concrete of the door-opening-shaped section structure can be rapidly calculated, the maximum temperature inside the lining concrete is calculated rapidly for different temperature control schemes, and a temperature control and crack prevention construction scheme suitable for a project is provided under the conditions that the inside maximum temperature is smaller than or equal to the permitted maximum temperature and economy is achieved. The error is smaller than 3% compared with the simulation calculation result of a finite element method, and the temperature control and crack prevention design calculating method is particularly suitable for optimization and adjustment of primary design and an on-site construction scheme.
Owner:WUHAN UNIV

Temperature control and crack prevention design calculation method for lining concrete with round cross section structure

The invention discloses a temperature control and crack prevention design calculation method for lining concrete with a round cross section structure. The temperature control and crack prevention design calculation method comprises the following steps of determining a temperature control and crack prevention target; calculating a permitted highest temperature; drafting a temperature control scheme; calculating the highest temperature inside the concrete; and designing a temperature control and crack prevention scheme on the premise that the calculated highest temperature is smaller than or equal to the permitted highest temperature. According to the method, calculation formulas are simple; influences of surrounding rock performance, lining thickness, concrete strength, air temperature in a tunnel, water cooling and water temperature, pouring temperature and the like can be reflected reasonably; the permitted highest temperature of pouring construction in each month of the construction period of the lining concrete with the round cross section structure can be calculated rapidly; the temperature control and crack prevention construction scheme applicable to a project is provided on the condition that the inside highest temperature is smaller than or equal to the permitted highest temperature and the economic requirement is met; the error between the result of the method and the result of simulating calculation of a finite element method is smaller than 5%; and the temperature control and crack prevention design calculation method is particularly applicable to primary design and optimization and adjustment on an on-site construction implementation scheme.
Owner:WUHAN UNIV

Sinusoidal signal frequency estimation method based on interpolation DFT

ActiveCN110333389ASuppression of Frequency Estimation ErrorsEnhance the anti-image component interference abilityFrequency measurement arrangementPower qualityFrequency spectrum
The invention discloses a sinusoidal signal frequency estimation method based on interpolation DFT, being applicable to electric energy quality assessment of a power system. A real sinusoidal signal has a mirrored negative frequency component on spectrum, so that the directly use of DFT generates a relatively large estimation error due to the spectrum leakage of the negative frequency component under an incoherent condition. Aiming at the problem, a two-point IpDFT algorithm based on signal adjacent spectral lines is provided. The method comprises the steps of firstly performing discrete Fourier transformation on a time domain signal x(n), and acquiring an index value corresponding a peak spectral line through searching DFT amplitude spectrum, namely a rough estimated value k0 of frequency; then positioning a secondary spectral line of the amplitude through comparing the left and right spectral line amplitudes of the peak spectral line; and at last, constructing an interpolation proportion by the two maximum DFT spectral lines, computing a deviation [delta]0 of the rough estimated value of the frequency and thus acquiring the complete signal frequency. The method can effectively suppress the spectrum leakage and interference, has the relatively high estimation accuracy and anti-noise performance, and has more significant superiority in the scene with severe spectrum interference.
Owner:SOUTHEAST UNIV

Diesel engine DPF active regeneration control method

The invention discloses a diesel engine DPF active regeneration control method. The diesel engine DPF active regeneration control method comprises the following steps that the exhaust mass flow of anengine, the DOC inlet temperature and the air pressure difference between a DPF outlet and a DPF inlet are acquired, and whether the regeneration triggering condition at the moment is met or not is judged; when the regeneration triggering condition is met, the regeneration control is triggered, the initial oil injection quantity in the current state is calculated; an oil injection device is controlled to start to perform oil injection, and meanwhile, whether the DPF inlet temperature meets the set regeneration target temperature or not is detected; when the DPF inlet temperature meets the setregeneration target temperature, oil injection is carried out at the current oil injection quantity; when the DPF inlet temperature does not meet the set regeneration target temperature, the fuel oilinjection amount is regulated through a regeneration controller, so that a proper oil injection quantity is guaranteed to be regenerated, and meanwhile, whether the air pressure difference before andafter a DPF is reduced to a set lower limit value or not is detected; and when the regeneration end condition is met, oil injection is stopped, and the regeneration process is completed. According tothe diesel engine DPF active regeneration control method, parameters needing to be calibrated are few, the universality is high, the calculation fluctuation under a feed-forward oil quantity transientcondition is reduced, and the regeneration efficiency is improved.
Owner:DONGFENG COMML VEHICLE CO LTD

Parallel interpolation bit synchronization system in all-digital demodulation and synchronized method thereof

The invention provides a parallel interpolation bit synchronization system in all-digital demodulation and a synchronized method thereof. The method comprises the following steps: (1) delaying two paths of input parallel sampling data K1 and K2 respectively for three clock cycles, and obtaining K1_d1, K1_d2, K1_d3 and K2_d1, K2_d2 and K2_d3; (2) respectively carrying out interpolation processing to the above sampling data and uk; (3) delaying indicating effective signals to the time when the above interpolation processing obtains results, carrying out extraction to the interpolation processing results in the step (2) according to the indicating effective signals, obtaining maximum points of symbols and traversing points of symbols and outputting the maximum points of the symbols; (4) calculating timing errors according to the maximum points of symbols and traversing points of symbols obtained in the step (3), carrying out noise filtration on the timing errors and obtaining NCO control signals after adjusting amplitude processing; and (5) determining the interval uk of the indicating effective signals and the maximum points of symbols with the former sampling point, entering next clock cycle and starting execution from the step (1).
Owner:XIAN INSTITUE OF SPACE RADIO TECH

Correcting method for global satellite navigation system troposphere zenith delay

The invention discloses a correcting method for global satellite navigation system troposphere zenith delay, and relates to correcting technologies for troposphere zenith delay of radio signals in satellite navigation application. The method includes the following steps that (1), a global troposphere zenith delay original data base is built; (2), a global troposphere zenith delay data base is built; (3), a reseau and global troposphere zenith delay simplifying data base is built; (4), the annual mean value and annual variation amplitude of the troposphere zenith delay are calculated; (5), a global troposphere zenith delay correcting parameter table is stored; (6), the troposphere zenith delay is corrected through a receiver end. The correcting method for the global satellite navigation system troposphere zenith delay has the advantage that elaborate space and time changes of the global troposphere zenith delay are fully considered, simple calculation formulas and a reasonably optimized space reseau and parameter table storing method are adopted, so that the correcting method has the advantages of being high in accuracy, easy and convenient in calculation and high in efficiency, and meanwhile requirements for storing space of the receiver end are lowered.
Owner:INST OF GEODESY & GEOPHYSICS CHINESE ACADEMY OF SCI

A tunnel floor lining concrete temperature control anti-cracking tensile stress K value control design method

The invention provides a tunnel bottom plate lining concrete temperature control anti-cracking tensile stress K value control design method. The method comprises the following steps of 1, collecting the data for tunnel bottom plate lining concrete temperature control anti-cracking calculation; step 2, analyzing and determining a temperature control anti-cracking target and an anti-cracking safetycoefficient allowable value K; step 3, designing a temperature control anti-cracking measure scheme, analyzing the variables, and formulating a plurality of lining concrete temperature control anti-cracking construction measure schemes; substituting each proposed temperature control measure scheme into formulas 1 and 2 for calculation to obtain tensile stress omega 1 and omega 2 when the minimum anti-cracking safety factor Kmin of each scheme in the early stage and winter of the concrete construction period occurs; calculating the age d1 and the age d2 of each scheme at the early stage and inthe case of Kmin in winter; step 3-4, calculating the minimum anti-cracking safety factor Kmin of the lining concrete during the construction period of each scheme; step 3-5, under the condition thatKmin is greater than or equal to K, optimally selecting a measure scheme for construction application according to a simple, practical and economic principle.
Owner:WUCHANG UNIV OF TECH +2

Anchor wire prestress loss and rock-soil body creep coupling based computing method for side slope creep values

ActiveCN104179176ACorrectly reflect the change of stress stateThe calculation formula is simpleClimate change adaptationExcavationsRelaxation equationState variation
The invention discloses an anchor wire prestress loss and rock-soil body creep coupling based computing method for side slope creep values. The computing method comprises the following steps: selecting the side slope that is comparatively uniform in soil body materials and properties and low in stability; determining the physical and mechanics parameters of the rock-soil body of the side slope; determining the water content W, the volume weight r, the uniaxial compressive strength Rc, the tensile strength Rt, the deformation modulus Eo, the cohesive force C and the internal frictional angle Phi of the rock-soil body; determining the creep curvilinear equation of the rock-soil body; determining the instantaneous elastic modulus Eh, the lagging elastic modulus Ek and the viscosity coefficient Eta<k> of the rock-soil body; bringing all the parameters into the relaxation equation of the coupling effect computation module, and working out the relational expression of the variation of anchor wire stresses and time; computing the anchor wire stress value at each time point according to the relational expression of the variation of anchor wire stresses and time, and figuring out the anchor wire prestress loss according to the anchor wire stress calculated value at each time point. The computing method can correctly reflect the variation situation of the strained condition of the prestressed anchor wire.
Owner:SHANDONG UNIV OF SCI & TECH +2

Method for rapidly and accurately measuring low angle grain boundary orientation under transmission electron microscope

The invention discloses a method for rapidly and accurately measuring low angle grain boundary orientation under a transmission electron microscope. The method comprises the following steps: using a transmission electron microscope double bar-tilting rotation tilting sample, enabling a zone axis of a crystalline grain I in a coordinate system I to be at the position of a positive zone axis, and collecting a convergent beam electron diffraction pattern of the crystalline grain I at the moment; keeping a camera constant L of the transmission electron microscope and the parameter of the convergent beam diffraction condition unchanged, and collecting a convergent beam electron diffraction pattern of a crystalline grain II in a coordinate system II; superposing a kikuchi pattern of the collected crystalline grain I and a kikuchi pattern of the collected crystalline grain II by using software; measuring the distance S between a kikuchi pole of the crystalline grain I and a kikuchi pole of the crystalline grain II as well as the rotation angle gama of the kikuchi pole of the crystalline grain II vertical to the pattern surface relatively to the kikuchi pole of the crystalline grain I; calculating the angle theta according to the geometric characteristic of the kikuchi line, theta=arctan(s/L)=s/L; using a formula (6) cos theta=(cosgama+cosgamacostheta+costheta-1)/2 to calculate the orientation of the crystalline grain I and the crystalline grain II. The method does not need the additional installation of hardware and software, and the common transmission electron microscope is used to measure the orientation.
Owner:YANSHAN UNIV

Determination method for damping polarity supplied by power generator excitation system in oscillation process

An embodiment of the invention discloses a determination method for damping polarity supplied by a power generator excitation system in an oscillation process, and aims to solve the technical problem that the damping polarity supplied by each control system of a power generator unit in a low-frequency oscillation process is not clear caused by the fact that damping conditions supplied by different control systems of the power generator unit in the low-frequency oscillation process of the power generator unit are not analyzed independently at present while only the overall external characteristics of the power generator unit are obtained. The determination method disclosed by the embodiment comprises the steps of reading disturbance data of the power generator unit; extracting a power generator transient-state electric reactance Q-axis potential changing amount: a formula which is as shown in the description, and a power generator rotary speed changing amount: a formula which is as shown in the description, according to the disturbance data of the power generator unit; and calculating dot product values of the formulas which are as shown in the description; and determining the damping polarity supplied by the excitation system according to the dot product values.
Owner:ELECTRIC POWER RES INST OF GUANGDONG POWER GRID

Load balancing method and load balancing system

The invention discloses a load balancing method and a load balancing system. The method is characterized by through a wireless access control server, establishing a load balancing group, assigning main wireless access equipment, generating a load balancing information table of the load balancing group and storing to all the wireless access equipment; detecting whether the wireless access equipment with highest receiving signal intensity satisfies a load balancing threshold, if the wireless access equipment satisfies the load balancing threshold, associating equipment to be accessed with the wireless access equipment; otherwise, through the load balancing information table, measuring and calculating service quality coefficients of all the wireless access equipment of the load balancing group and determining the wireless access equipment which allows the equipment to access. By using the load balancing method and the load balancing system, a high quality network service is provided and simultaneously frequencies of states of all the wireless access equipment which can be accessed and is scanned by the equipment to be accessed can be reduced, and a connection and network access speed of the equipment to be accessed can be increased. Simultaneously, a condition that the equipment to be accessed acquires load information condition of the wireless access equipment can be avoided and safety of a WLAN network system is increased.
Owner:深圳市中天鼎通信息科技有限公司

Self-adaptive control machining method with constancy of main shaft power as target

The invention relates to a self-adaptive control machining method with constancy of main shaft power as a target. According to the method, constancy of the main shaft power is taken as the target in the machining process, the main shaft power is controlled by changing a feeding speed, and the calculation formula of the feed speed adjustment amount delta f is simple; and an experiment is used for carrying out fitting on the formula among a target power difference value delta p, a feeding speed f, a rotating speed s, a radial cutting depth d, and an axial cutting depth a and the feeding speedadjusting amount delta f, an acquisition system is used for acquiring the above machining parameters in real time, and then the machining parameters are transmitted to a control system, a speed deltaf required by machining is calculated in the control system, and the feeding speed f is adjusted, so that constant-power machining of a workpiece is guaranteed. The method has the advantages of beingsimple in algorithm formula, short in calculation time and easier to embed into a universal numerical control system such as SIEMENS and FANUC without arranging additional equipment, and meanwhile the numerical control system can be built through a build-in PLC and a build-in CNC of the system, so that good transplanting performance is achieved, and self-adaptive control in the high-precision machining process can be realized.
Owner:CHINA YITUO GROUP +1

Device and method for conducting vibration monitoring and fault early warning on centrifugal pump

The invention discloses a device and method for conducting vibration monitoring and fault early warning on a centrifugal pump. The vibration condition of the centrifugal pump is collected through a piezoelectric type vibration acceleration sensor and input into a host computer through a multichannel data acquisition card, so that centrifugal pump vibration signals are formed. The method for conducting vibration monitoring and fault early warning through the device includes the steps of simulating centrifugal pump faults, and collecting fault signals; conducting noise elimination processing onthe fault signals; conducting feature extraction on the noise elimination signals; establishing a centrifugal pump fault feature vector; operating a centrifugal pump set, and also conducting noise elimination processing and feature extraction on the signals collected by the vibration sensor, so that a comparison feature vector is obtained; and comparing the comparison feature vector with a simulation feature vector, so that whether faults exist in a corresponding measure point of the centrifugal pump or not is known. The device and the method have the beneficial effects of being low in cost, easy to implement and great in universality. Early warning results are concise and clear, and fault early warning accuracy and efficiency of the centrifugal pump are improved.
Owner:LANZHOU UNIVERSITY OF TECHNOLOGY

Rapid design method for temperature control and crack prevention temperature stress control over tunnel floor lining concrete

The invention provides a rapid design method for temperature control and crack prevention temperature stress control over tunnel bottom lining concrete. The rapid design method comprises the followingsteps that step 1, data for calculation of temperature control and crack prevention of the floor lining concrete is collected; step 2, a target of temperature control and crack prevention of the floor lining concrete and values of allowable temperature stress [sigma<max>] are analyzed and determined; step 3, temperature control and crack prevention measure schemes are designed, specifically, thevariables are analyzed, a plurality of floor lining concrete temperature control and crack prevention construction measure schemes are formulated; the maximum temperature tensile stress sigma<max> offloor lining concrete construction periods of the floor lining concrete temperature control and crack prevention construction measure schemes is calculated; and step 3-3, in the case that sigma<max> is less than or equal to [sigma<max>], a measure scheme is optimized according to the principle of simple practical economy for construction application. The rapid design method can quickly calculate the maximum temperature tensile stress in the construction period of pouring lining concrete of a gate hole type tunnel floor structure in any period (season), is small in calculation error, and is especially suitable for the preliminary design and real-time rapid design calculation during the construction period.
Owner:WUCHANG UNIV OF TECH +2

Method for calculating failure critical clearing time based on critical unit pair

The invention discloses a method for calculating a failure critical clearing time based on a critical unit pair. The method comprises the following steps of collecting all the running parameter data of a power system by a wide-area phase measuring system to form a data file; for each failure in a predicted event set provided by a scheduling department, simulating the transient values of the power system by maximum clearing time based on the data file to generate simulated data; selecting a critical unit pair according to the simulated data; judging the stability of the critical unit pair and finally judging the stability of the power system; dividing failures into instability failures and stability failures according to the judging results of the stability; calculating the critical clearing time of the critical unit pair if the failures belong to the instability failures; and calculating the critical clearing time of the power system. The method has the advantages that the values of the critical unit groups and the critical units about agglomeration and the like do not need to be recognized, the critical clearing time of the system can be obtained based on the data of few generators in the power system, and the calculation process is simple and fast.
Owner:HARBIN ENG UNIV

A tunnel bottom plate lining concrete temperature crack control anti-cracking K value design method

The invention provides a tunnel bottom plate lining concrete temperature crack control anti-cracking K value design method. The method comprises the following steps of 1, collecting the data for tunnel bottom plate lining concrete temperature crack control design calculation; step 2, analyzing and determining a temperature control anti-cracking target and an anti-cracking safety coefficient allowable value K; step 3, designing a temperature control anti-cracking measure scheme, which comprises the following steps of step 3-1, analyzing a variable, and formulating a plurality of lining concretetemperature control anti-cracking construction measure schemes; step 3-2, calculating the minimum anti-cracking safety factor Kmin of each scheme in the lining concrete construction period by adopting a formula; and step 3-3, under the condition that Kmin is greater than or equal to K, optimally selecting a measure scheme for construction application according to a simple, practical and economicprinciple. According to the method, the temperature control anti-cracking scheme design can be carried out in the design stage, and the method can be used for optimizing and improving the constructiontemperature control measures in real time to achieve the temperature control target for finding problems and changing construction technologies, conditions and the like in the pouring construction process.
Owner:WUCHANG UNIV OF TECH +2

Diesel vehicle DPF active regeneration temperature control method

The invention relates to a diesel vehicle DPF active regeneration temperature control method. The diesel vehicle DPF active regeneration temperature control method successively includes the followingsteps that (1) the DOC inlet temperature T1 and the DPF inlet temperature T2<original> before fuel injection are red, the DPF inlet regeneration target temperature T3 is selected, the initial fuel-injection quantity q of an HC injection unit (9) is calculated or selected, and fuel is injected into the inlet of a DOC (1) according to the initial fuel-injection quantity q; (2) the DPF inlet temperature T2<k> is red again, and the next fuel-injection quantity q<k+1> of the HC injection unit (9) is redetermined; (3) the fuel-injection quantity injected into the inlet of the DOC (1) is adjusted tothe next fuel-injection quantity q<k+1> determined in the step (2); (4) the DPF inlet temperature T2<k> is read again, and the next fuel-injection quantity q<k+1> of the HC injection unit (9) is redetermined; and (5) the step (4) is repeated. According to the diesel vehicle DPF active regeneration temperature control method, rapid regeneration of the DPF is achieved with low fuel consumption and good regeneration temperature, and the regeneration efficiency of the DPF is high.
Owner:DONGFENG COMML VEHICLE CO LTD

Multi-antenna spectrum sensing method applicable to high-dimension finite sample conditions

The invention relates to a multi-antenna spectrum sensing method applicable to high-dimension finite sample conditions. By adopting relevance between multi-antenna receiving signal components to structure sensing decision and sensing decision threshold based on the random stochastic matrix, the multi-antenna spectrum sensing method includes firstly, continuously sampling multi-antenna receiving signals to form a receiving signal data matrix X; then, calculating relevance measurement indicators among the multi-antenna receiving signal components on this basis, and calculating to obtain sensing decision I; secondly, calculating sensing decision threshold t on the basis of the random stochastic matrix; finally, implementing sensing decision, to be specifically, judging that no spectrum hole exists when the sensing decision I is larger than the preset threshold t, or otherwise, judging that a spectrum hole exists. The multi-antenna spectrum sensing method has the advantages that the method is simple and low in calculation complexity in sensing application of the high-dimension finite sample capacity, efficient total blindness detection under the condition of deficiency in statistical information of master user signals, wireless channels and noise can be realized, and sensing results are reliable and the like.
Owner:JISHOU UNIVERSITY

Computing method for torsional rigidity of split conductor of overhead power transmission line

InactiveCN102436546AFactors that reflect torsional stiffnessImprove calculation accuracySpecial data processing applicationsEngineeringForce analysis
The invention discloses a computing method for torsional rigidity of a split conductor of an overhead power transmission line. The method comprises the following steps of: firstly, analyzing length change of a sub-conductor due to a torsional movement of the split conductor, and deducing a practical level tension computational formula according to a linear formula of the power transmission line and material property of the conductor when the sub-conductor is torsional; then carrying out a mechanical analysis for a four-split conductor under the action of a torque, in the analysis process, utilizing a characteristic that the split conductor is divided into multiple subspans through a conductor spacer, firstly carrying out a force analysis for the subspans, and finally obtaining a torque balance equation under the action of the torque, and further obtaining a computational formula for the torsional rigidity of the four-split conductor. Finally, summarizing a common computational formula for the torsional rigidity of the split conductor for any split number. In the computing method, the method is adopted to calculate the torsional rigidity of the split conductors with parameters, such as different split numbers and span length and the like, and higher computational accuracy can be achieved, and effectiveness and adaptability of the computing method can be proved.
Owner:XI AN JIAOTONG UNIV

Method for correcting span during rock blasting damage cross-hole sound wave test

The invention discloses a method for correcting a span during rock blasting damage cross-hole sound wave test. The method is suitable for span correction between an emitting transducer and a receiving transducer in a sound wave hole for rock blasting digging damage sound wave monitoring in blasting digging engineering. A sample is measured through a sound wave hole before blasting to obtain coordinates A (xa, ya, za) and B (xb, yb, zb) of orifices A and B; after blasting, the orifices of the dug residual sound wave hole are cleared, and the sample is measured to obtain the coordinates A' (x'a, y'a, z'a) and B' (x'b, y'b, z'b) of the two orifices A and B; an actual distance between the emitting transducer and the receiving transducer at the moment can be precisely calculated through a formula, wherein eta=xb+(xb-x'b)tbi-xa-(xa-x'a)tai, delta=yb+(yb-y'b)tbi-ya-(ya-y'a)tai, phi=zb+(zb-z'b)tbi-za-(za-z'a)tai. According to the method, the span can be corrected only through the coordinates of the orifices of the conventional sound wave monitoring hole of the engineering before and after blasting without measuring an inclination angle and an inclination direction of the sound wave detection hole; extra workload and extra equipment are not added; the method is easy to operate, precise and reliable. The calculating formula of the span correction method provided by the invention is simple; calculation is easy, and sound wave test analysis software is easily programmed and implanted.
Owner:WUHAN UNIV

Lightning protection grounding method of shared iron tower

The invention relates to a lightning protection grounding method for a shared iron tower, a lightning protection method is provided according to the carrying position of a 5G communication antenna, and when an antenna is located at the tower top (not within the protection range of a lightning rod), a lightning protection grounding down lead is arranged on an antenna fixing rod; when the antenna islocated at other positions (within the protection range of the lightning rod), the tower body is used as a grounding conductor; and a grounding resistance calculation formula is optimized, meanwhile,the influence of the horizontal grounding body, the vertical grounding body and the radial grounding body is considered, and an earth screen transformation scheme is provided based on an existing grounding system of an electric iron tower, that is to say, an annular grounding device is added on the basis of an original grounding grid. The annular grounding device and the original grounding grid are located on the same horizontal plane, the annular grounding device and the original grounding grid are mutually welded and communicated once every 3-5 m, the annular grounding device comprises a horizontal grounding body, a vertical grounding body and a radial grounding body, and finally the shared iron tower grounding resistor meets the requirements of a power system and a communication systemat the same time.
Owner:NORTH CHINA UNIVERSITY OF TECHNOLOGY

False face video detection method and system based on face cross-parallel ratio under trust mechanism

The invention discloses a false face video detection method and system based on a face intersection ratio under a trust mechanism. The method comprises a network training step and a sample testing step, wherein the network training step comprises model training and model verification. In the model training step, a segmentation network is trained, and a segmentation network model and a weight valueare stored; in the model verification step, a trained segmentation network is used to predict a mask, and a binarization threshold and a decision threshold are determined by using a grid search method, wherein the sample test comprises the steps of performing video framing preprocessing to obtain an input picture, determining a to-be-detected area by utilizing a segmentation network, performing smooth denoising, binarizing the to-be-detected area, determining a face area and an undetermined tampered area, calculating an intersection ratio of the face area and the undetermined tampered area under a trust mechanism, and finally performing true and false face judgment according to a judgment threshold value. The false face video generated by the deep face changing tool is detected, high accuracy is obtained in different databases, the cross-database test performance is obviously improved, and an effective way is provided for false face video detection.
Owner:SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products