119results about How to "Avoid cumbersome calculations" patented technology

The invention discloses a face detection method. The face detection method comprises the following steps: acquiring a to-be-detected image; inputting the to-be-detected image into a trained convolutional neural network, recognizing whether a face is included in the to-be-detected image, and estimating a facial pose, wherein a training sample image in a training sample set for training the convolutional neural network comprises position data and pose data of the face; outputting a detection result whether the face is included in the to-be-detected image; and outputting pose information of the face in the to-be-detected image if the to-be-detected image includes the face. The invention further discloses a face detection device, a method for training parameters of convolutional neural network, a computer device and a computer-readable storage medium. According to the face detection method disclosed by the invention, the face pose can be synchronously estimated during face detection, and the face recognition efficiency is further improved.

An actual power curve fitting based theoretical power determination method for a wind power plant comprises: reading wind power plant data; preprocessing the wind power plant data; calculating wind speed at the height of a wheel hub of a fan; filtering preprocessed fan power data to remove fault data, establishing a sample data set for actual power curve fitting, and dividing the sample data set; calculating theoretical power of the fan; adopting a correlation coefficient weighting method for correcting the theoretical power of the fan; obtaining theoretical power of the wind power plant; and evaluating theoretical power recovery precision through a correlation coefficient and a secondary center distance, and generating a data recovery quality factor. With the method, the theoretical power recovery precision is high and the tedious and complicated calculation is avoided; and the method is simple to operate, easy to realize and high in applicability.

The invention discloses a method for designing an elliptic cylindrical surface transverse gradient coil based on spatial alternation. The method is characterized by comprising the steps that firstly, two shafts of the elliptic cylindrical surface where the gradient coil is located are defined to be in the direction x and the direction y, the shaft length of the elliptic cylindrical surface where the gradient coil is located in the direction x is 2a, the shaft length in the direction y is 2b, the wiring length in the axial direction is 2L, the strength of the gradient field is G, and the linearity is E; secondly, spatial alternation is utilized for converting the elliptic cylindrical surface into a cylindrical face, and the current density basis function and the expression are constructed under the cylindrical face; thirdly, a functional construction related to the current density is built, and the coefficient of the current density expression is obtained by solving the functional extreme value; fourthly, the solved coefficient of the current density expression is utilized for solving distribution of current density in the space, and the wiring shape of the gradient coil is obtained. The complex calculation of gradient coil designed through the network discrete method is avoided, and the high efficiency and precision are obtained.

The invention discloses an ECU (Electric Control Unit) security access processing method. An ECU is set into a factory mode in a production line, and is set into a non-factory mode when leaving the factory; when external equipment transmits an access request to the ECU, the ECU reads a value of a mode mark position, and the state of the ECU is judged; if the ECU is in the factory model, the ECU is automatically unlocked to accept the access operation of the external equipment; if the ECU is in the non-factory mode, the ECU carries out security verification operation to the external equipment, when the external equipment passes the security verification, the ECU is unlocked to accept the access operation of the external equipment; and otherwise, the ECU refuses the access operation of the external equipment. According to the embodiment of the invention, the detection steps on the production line of automobiles can be reduced, the production efficiency is improved, and the safety of the ECU can be ensured.

The invention relates to the technical field of processing of images, in particular to a real-time circular printing image defect detecting method. The method comprises the following steps: a standard template graph and an on-line real-time graph of a target to be detected, obtaining a circle point coordinate through an exponential changing and searching step-length method, calculating a circle center and semi-diameter of a circular area of each of the standard template graph and the on-line real-time graph according to a method for positioning a circle through three points, dividing each circular area into a large quantity of annular areas, counting characteristics of grey level histograms of the annular areas, and comparing the characteristics of the grey level histograms of the annular areas of the standard template graph and the on-line real-time graph to complete image defect detection. Through the adoption of the real-time circular printing image defect detecting method, the circular printing image defection speed is improved from two aspects; a practicable method is provided for on-line defect detection based on machine vision.

The present invention discloses a method for estimating the runoff in the non-data area based on the ensemble Kalman filter, and relates to the technical field of hydrological simulation. The method comprises: calculating an optimized parameter value of the whole basin; calculating the estimated flow and the distribution regularity of the point C in the exit point of the sub-basin A; calculating the ensemble perturbation value of the variable and the parameter; calculating the runoff forecast set of the exit point in the downstream sub basic B, i.e., the hydrological station D; carrying out Gaussian perturbation on the actually measured runoff value of the hydrological station D, and taking the obtained Gaussian perturbation value as the observation data set; integrating in the observation data set to carry out assimilate update; and finally obtaining an optimal estimate value of the runoff QC of the point C. According to the method disclosed by the present invention, the filtering assimilation process is simplified, and the assimilation precision is better improved; and the shortcomings of the conventional method for calculating the runoff in the non-data area are avoided.

The invention discloses a small gear rough cutting method for a spiral bevel gear. The method comprises the steps that the concave face and the convex face of a small gear are positioned through small gear finishing bevel gear cutter parameters, machine tool adjusting parameters and midpoint checking sizes to obtain a small gear finish cutting gear socket, and coordinates and normal vectors of key points of the two tooth faces are calculated; initial rough cutting machine tool adjusting parameters are given, an equation of the tooth faces on the two sides of the small gear rough cutting gear socket is built, and coordinates of the corresponding key points are obtained; after the midpoint of the small gear rough cutting hear socket is rotated to the midpoint of the small gear finish cutting gear socket, the key points corresponding to the two gear sockets are connected and projected on the normal line of the key points of the finish cutting gear socket to obtain finish cutting allowance; the square root of the difference between the optimized finish cutting allowance and the preset finish cutting allowance is made to be the minimum by optimizing roughing bevel gear cutter parameters or machine tool adjusting parameters. According to the method, the roughing bevel gear cutter specifications can be reduced, the roughing bevel gear cutter and cutter tooth utilization rate is increased, it is ensured that the finish cutting allowance of the tooth faces on the two sides is uniform as much as possible, the follow-up finishing frequency is lowered, and the machining efficiency of the bevel gear is improved.

The invention discloses a global optimization control method of a ground source heat pump. Physical models of a ground source heat pump unit, a ground heat exchanger and a water pump are established and analyzed, an optimization algorithm is used for optimizing user side flow rate and ground side flow rate, and the optimized value is used as a set value to a bottom layer closed loop control to enable a ground source heat pump system to operate at an optimal state point. According to the global optimization control method of the ground source heat pump, a brand-new global control strategy of the ground source heat pump system is determined by modeling analysis of a compressor, an evaporator and a condenser of the ground source heat pump system and combining a particle swarm optimization algorithm, the influence of external interference and the control strategy on the system are fully considered, corresponding correction compensation measures are set, and strong engineering practical significance is achieved; in the actual operation process, the energy is saved by 14% compared with a constant flow operation strategy, and energy is saved by 6.7% compared with a traditional five-degreetemperature difference control strategy, and the global optimization control method of the ground source heat pump has important significance for saving the operation energy consumption of the groundsource heat pump system.

The invention discloses an image segmentation method based on a two-channel texture segmentation active contour model in the technical field of digital picture processing. The image segmentation method based on the two-channel texture segmentation active contour model comprises the steps that the gray level, the horizontal gradient field and the vertical gradient field of each pixel in an image are extracted; textural features corresponding to the gray level, the horizontal gradient field and the vertical gradient field of each pixel in the image are calculated; a gray feature channel and an edge feature channel are obtained according to the textural features; the two-channel texture segmentation active contour model is created; a texture segmentation model is minimized through evolvement of a horizontal set function to complete image segmentation. The image segmentation method improves algorithm efficiency, avoids incorrect segmentation caused by gray information, and improves accuracy of an algorithm.

The embodiment of the invention discloses a light field display control method and device, and light field display equipment, wherein the light field display control method comprises the following steps of: determining a part of depth distribution sub-area of a content at least according to depth distribution information of the content; and adjusting the focal length of a first sub-lens at least according to the display field depth range of the light field display equipment and the depth distribution sub-area, wherein the first sub-lens is a sub-lens influencing display of a first object in a sub-lens array of the light field display equipment; and the first object is a part that the content is positioned in the depth distribution sub-area. By means of the light field display control method and device, and the light field display equipment disclosed by the invention, the display quality of the object in the local depth distribution sub-area in the to-be-displayed or displayed content can be improved.

The invention discloses an electricity selling package recommendation method and device and a storage medium, and the method comprises the steps: classifying a user into a corresponding cluster according to the annual declaration quantity characteristics of the user, inputting the related characteristics of the user into a pre-established electricity selling package recommendation model corresponding to the cluster to obtain a package suitable for the user; wherein the electricity selling package recommendation model is pre-established through an improved XGBoost algorithm and a screened inputfeature set, according to the improved XGBoost algorithm, the optimal parameter combination of the XGBoost algorithm is obtained through a genetic algorithm, and then the improved XGBoost algorithm is obtained. According to the invention, the electricity selling package can be intelligently recommended.

The invention discloses an interactive method and an interactive system, and an acquisition method of a relative depth. The interactive method comprises the steps that a laser image containing an interactive object in target space is collected; the relative depth between the laser image and a reference laser image is computed, wherein the reference laser image is a pre-collected laser image which does not contain the interactive object in the target space; the relative depth between the laser image and the reference laser image is used to extract an image part of the interactive object from the laser image; and the image part of the interactive object is recognized, and a corresponding control instruction is generated according to a recognition result. In this way, the interactive method, the interactive system and the acquisition method of the relative depth disclosed by the invention can avoid cumbersome computation steps brought by computation of a direct depth and can increase the speed of depth acquisition and interaction.

The invention provides a satellite passive positioning method based on GPS broadcast data. The method comprises: acquiring measuring data of a satellite; according to the measuring data, calculating a conversion matrix M from an earth-fixed coordinate system to a satellite local system; selecting an earth ellipsoidal model and obtaining corresponding earth geometrical parameters and an earth ellipsoidal surface equation; and according to the conversion matrix M, solving an object position equation, and solving object position coordinates by simultaneously solving the earth ellipsoidal surface equation. The measuring data comprises the GPS broadcast data including a satellite real-time position P<se><-> and a speed V<se><->, object measurement angles beta<x> and beta<y> and attitude angle data, i.e., a pitch angle phi, a yaw angle yaw angle gamma and a roll angle theta, wherein the GPS broadcast data is obtained through a satellite data bus, the beta<x> and the beta<y> are obtained through a load subsystem, and the attitude angle data is obtained through an attitude orbit control subsystem. According to the invention, the high-precision GPS broadcast data is taken as a system input quality, such that the precision of each parameter during a calculation process is ensured, accordingly, the precision of a positioning result is ensured, at the same time, an offline processing process is unnecessary, and the satellite autonomous calculating capability is enhanced.

The invention discloses a monotonic tensile parameter-based metal material fatigue crack growth model construction method. The method includes the following steps that: step 1, a linear cumulative damage-based crack growth rate model is obtained; step 2, low-cycle fatigue parameters are characterized with monotonic tensile parameters; and step 3, results obtained in the step 2 are introduced intothe linear cumulative damage-based crack growth rate model in the step 1, so that a required crack growth rate model is obtained. According to the monotonic tensile parameter-based metal material fatigue crack growth model construction method of the invention, relationships between the monotonic tensile parameters and the low-cycle fatigue parameters are established; and the model is characterizedwith the monotonic tensile parameters, high test costs and cumbersome calculation processes in a low-cycle fatigue test can be avoided, and the prediction of a crack growth rate is more accurate.

The invention discloses a GPS L2C signal generation method and device. The method includes: capturing the GPS L1CA signal and inserting a guide mark to generate a code NCO accumulation amount; based on the code NCO accumulation amount, generating E, P, L The local pseudo-code synchronization signal group and the capture pattern selection flag group of the three branches; based on the local pseudo-code synchronization signal group, use the CM shift register and the CL shift register to count the CM code and CL code in the L1CA signal Operation and shift operation to generate three groups of local CM codes and CL codes of E, P and L; based on the capture pattern selection flag group, perform time division multiplexing processing on the three groups of local CM codes and CL codes of E, P and L, To obtain three groups of local L2C codes of E, P and L; use the code cycle jump edge of the L1CA signal after a preset time to guide the three groups of local L2C codes of E, P and L based on the guidance flag, and generate the target L2C signal .

The invention discloses a discriminating method of a natural indigo dyed silk fabric. The method comprises the following steps: 1, dyeing a silk fabric by adopting a natural indigo dye to obtain the natural indigo dyed silk fabric; 2, extracting the extract liquid of the natural indigo dyed silk fabric; and 3, carrying out fluorescence test on the extract liquid obtained in step 2. Characterization test of the extract liquid of the natural indigo dyed silk fabric is carried out, so the purpose of rapid discrimination of the natural indigo dye and a synthesized indigo dye is achieved according to characteristic emission peaks of the extraction liquid in a three dimensional fluorescence spectrogram. The method has the advantages of simple operation, good practical use effect, and great significance to food exploitation, fabric dyeing, and discrimination and maintenance of ancient textiles.

The invention discloses a method for comprehensively evaluating the combustible dust explosion hazard level by utilizing multi-element dust explosion parameters. The method adopts fuzzy mathematics theory and analytic hierarchy process (AHP) in the system evaluation method, Using representative conventional explosives, According to the minimum ignition energy of dust cloud, the minimum ignition temperature of dust cloud, the minimum ignition temperature of dust layer, the lower limit of explosion, the maximum explosion pressure, the maximum explosion pressure rise rate and the characteristic parameters of explosion index, the membership function is established, and the evaluation model is established to evaluate the explosion hazard of different dust. The comprehensive evaluation method ofthe combustible dust explosion hazard grade established by the invention is simple, Easy to popularize, avoid tedious calculation and test, combine the existing data to obtain the corresponding comprehensive index value, easy to form a common evaluation standard, the result is intuitive, rapid identification of combustible dust dust explosion hazards, can provide guidance for the enterprise design safety production process.

The invention discloses a method for determining an IV characteristic fitting curve of a crystalline silicon photovoltaic cell. With the crystalline silicon photovoltaic cell as an object, the invention provides data in a data manual of the photovoltaic cell without solving a transcendental equation and any measured data. Disclosed is a method for simply fitting a left side and a right side of themaximum power point of the IV characteristic curve of the crystalline silicon photovoltaic cell by the two second-order Bezier functions, the IV characteristic fitting curve of the crystalline silicon photovoltaic cell meeting the engineering precision requirement is given by finding out the linear law between the control point position of a Bezier function curve and a filling factor of the crystalline silicon photovoltaic cell.

A method for calculating the ultimate bearing capacity of a strip foundation under an inclined load is disclosed. By constructing a failure model of a multi-rigid-block foundation under an inclined load, the formula for calculating the ultimate bearing capacity of strip foundation under inclined load is established, and the problem of solving the ultimate bearing capacity of strip foundation is equivalent to a multi-dimensional function extreme value problem with nonlinear constraints. The method takes the advantage of the improved radial motion algorithm to solve the extremum problem of multidimensional functions; the failure slip line of foundation is searched for and the corresponding ultimate bearing capacity of foundation is calculated. Through continuous comparison and progress, thefailure slip line in the domain that makes the minimum bearing capacity of foundation is found, and the corresponding minimum bearing capacity is the ultimate bearing capacity of foundation. Comparedwith the prior art, the invention avoids the tedious calculation of the traditional theoretical method and the limitation of the calculation precision, can conveniently and quickly find out the foundation failure slip surface under the action of the inclined load, and greatly improves the work efficiency.

The embodiment of the invention discloses a CSI-based human body activity data detection method and apparatus and an electronic device. The CSI-based human body activity data detection method comprises the steps of obtaining the original CSI data of multiple paths of received signals received by a Wi-Fi signal receiving device, wherein the received signals at least comprise the reflected signals generated by the human body activities in the propagation process of the transmitted signals transmitted by a Wi-Fi signal transmitting device, and each path of received signals comprises N subcarriers; acquiring the first target CSI data of a first path of received signals and a second path of received signals based on the original CSI data of the multiple paths of received signals; preprocessingthe first target CSI data of the two paths of received signals to obtain the second target CSI data of the two paths of received signals; and based on the second-order differential phase of the secondtarget CSI data of the two paths of received signals, determining the starting time and/or the ending time of the human body activity. Therefore, a tedious calculation process is avoided, and a detection result is rapidly obtained.

The invention discloses a typhoon path prediction display system for power grid, in accordance with that present invention, includes typhoon data input module, storage module, parameter retrieval module, data analysis module, a current typhoon model construction module and a data transmission module, a parameter obtaining module of the invention obtains the data of the current typhoon and the historical typhoon through a storage module, after the data analysis module matches the parameters of the data of the current typhoon and the historical typhoon, calculation of matching ratio and matchingdegree, The current typhoon model building module fits the historical typhoon track map and the current typhoon track according to the calculation results, and obtains a new typhoon track map by deviation correction, avoiding tedious calculation process, and can be completed only by simple calculation, which significantly shortens the calculation time and the use of large-scale calculation instruments, thereby reducing the use cost.