251 results about "Back calculation" patented technology

A system and method for determining the temperature of a semiconductor wafer at the time of thermal contact of the semiconductor wafer with a temperature sensing element. According to the invention, a temperature profile of the temperature sensing element is recorded from the time of thermal contact up to the time of thermal equilibrium between the semiconductor wafer and the temperature sensing element and the temperature of the semiconductor wafer at the time of thermal contact is determined on the basis of a time period between the time of thermal contact and the time of thermal equilibrium and the temperature TG of the semiconductor wafer reached at the time tG of thermal equilibrium is determined by back calculation with the aid of an equation derived from Newton's law of cooling.

A system and method for determining the temperature of a semiconductor wafer at the time of thermal contact of the semiconductor wafer with a temperature sensing element. According to the invention, a temperature profile of the temperature sensing element is recorded from the time of thermal contact up to the time of thermal equilibrium between the semiconductor wafer and the temperature sensing element and the temperature of the semiconductor wafer at the time of thermal contact is determined on the basis of a time period between the time of thermal contact and the time of thermal equilibrium and the temperature TG of the semiconductor wafer reached at the time tG of thermal equilibrium is determined by back calculation with the aid of an equation derived from Newton's law of cooling.

The invention belongs to the technical field of visual detection, and provides a calibration method which is convenient and fast to use and accurate. The fast calibration method aims to make full use of characteristics of the multiple line structured light visual sensor in the aspect of three-dimensional shape measurement so as to be applied to on-line three-dimensional detection of industrial products. According to the technical scheme, the fast calibration method of the multiple line structured light visual sensor comprises the steps that a three-dimensional measurement model of the multiple line structured light visual sensor is built, wherein a measurement coordinate system is set in a world coordinate system where virtual space feature points are calibrated through the three-dimensional measurement model; according to the measurement model, due to the adoption of the Zhang calibration method that visual angles are not determined, a plurality of coplanar positions are placed through the visual sensor (namely, a plane target) to acquire virtual space feature point arrays so as to carry out camera calibration; three-dimensional point arrays on each light plane are acquired according to camera calibration results and the cross ration invariability theory to finish light plane calibration; the three-dimensional shape of a measured object is worked out in a back calculation mode through the measurement model. The fast calibration method of the multiple line structured light visual sensor is mainly used for visual detection.

The invention discloses a device for online monitoring lightning current and lightning voltage waveform parameters of an overhead transmission line, relating to the fields of lightning information measurement and lightening prevention design. The monitoring system is directly installed on a sub-wire of a single-phase wire or a split wire of the overhead transmission line, a voltage sensor and a current sensor respectively monitor a lightning current signal and a lightning voltage signal, and the two paths of signals are sent to a GPRS (General Packet Radio Service) network by a GPRS module through a signal acquisition and processing module. When the system runs normally, the induced voltage of a check coil is processed and then directly supplies power to the signal acquisition and processing module and the GPRS module and fully charges a lithium ion storage battery; and when the wire has the lightning current and the system is power failure, the lithium ion storage battery supplies power for converting an instant power failure protective capacitor to prevent data loss. The invention provides original data for lightning positioning of the transmission line and the lightning current and lightning voltage waveform parameters of a back calculation lightning stroke point and has the advantages of simple manufacturing process, small size and convenient installation.

A method and system for determining a temperature of a working gas passing through a passage to a turbine section of a gas turbine engine. The method includes identifying an acoustic frequency at a first location in the engine upstream from the turbine section, and using the acoustic frequency for determining a first temperature value at the first location that is directly proportional to the acoustic frequency and a calculated constant value. A second temperature of the working gas is determined at a second location in the engine and, using the second temperature, a back calculation is performed to determine a temperature value for the working gas at the first location. The first temperature value is compared to the back calculated temperature value to change the calculated constant value to a recalculated constant value. Subsequent first temperature values at the first location may be determined based on the recalculated constant value.

The invention provides a real-time sensing system and method of the tunneling rack mass state of a TBM. The method and system is used for solving the problem that an adaptive tunneling scheme cannot be selected due to the fact that the tunneling tunnel rack mass state of the TBM cannot be effectively reflected in real time through existing means. By collecting hard-rock tunnel TBM tunneling parameters and corresponding rock mass state parameter data, a data storehouse module is built; step-by-step regression is adopted, a rock-machine relationship sensing model is built, and a clustering algorithm is adopted to achieve rock mass quality grading and form a calculation module kernel; in a model calculation module, by reading current tunneling parameters of the TBM in a tunneling project, back calculation is conducted to obtain rock mass state information, and the rock mass state information is displayed on a TBM upper computer visible interface in a real-time output and display module inreal time. By means of the system and method, the deficiencies of difficult obtaining of traditional rock mass condition parameters, backward means and unknown rock mass state in front of the tunnelface are made up for; energy consumption of the TBM is reduced, the equipment and personnel safety is guaranteed, the tunneling efficiency of the TBM is improved, and the safe efficient tunneling capability of the TBM is greatly improved.

The invention discloses an image characteristic registration based geometrical fine correction method for an aviation multispectral remote sensing image, comprising following steps of: 1) utilizing a Sobel edge detection operator to carry out edge extraction on both an aviation multispectral remote sensing image to be registered and a normal incidence high-resolution satellite image which is taken as a standard; 2) utilizing a Harris angular point detection method to detect angular points of the aviation multispectral remote sensing image to be registered and the normal incidence high-resolution satellite image which is taken as the standard; 3) carrying out rough correlation between the two images through a correlation method; 4) carrying out fine correlation between the angular points of the two images through calculating supporting strength; 5) carrying out back calculation to obtain a multinomial coefficient according to a multinomial model; and 6) carrying out gray level re-sampling by adopting a bilinear interpolation to obtain a registered image. According to the image characteristic registration based geometrical fine correction method for the aviation multispectral remote sensing image, disclosed by the invention, the geometrical correction problem of the aviation multispectral remote sensing image lacking of a ground reference point can be better solved and the geometrical correction precision of an aviation multispectral scanner is improved, so that an aviation remote sensing technology can be better applied to production livings of national economy.

A star map data based method for measurement of in-orbit precision of a star sensor. The single-frame static star map data obtained by the star sensor is utilized for measure the key performances such as in-orbit pointing accuracy and viewing angle of the star sensor; a quaternion estimation algorithm (QUEST), which converts the eigenvalue solution into the solution of a root of a fourth-order equation to speed up, is employed to obtain an optimal attitude matrix Aq; then according to the attitude matrix Aq and the identified n navigational stars, conducting back calculation to obtain the theoretical position of each navigational star on the image sensor, wherein the theoretical position is the theoretical point of intersection (xit, yit) of the primary optical axis of the star sensor and the interface; calculating the errors between the theoretical positions of the star points (xit, yit) and the measured actual positions of star points, wherein the actual star point position is the actual point of intersection (xi, yi) of the primary optical axis of the star sensor and the interface; and calculating a mean value of the errors and converting the mean value to an equivalent angle value, so as to obtain the precision of the star sensor. Compared to the commonly used in-orbit comparison method, the method provided by the invention eliminates the influence of the errors of time alignment and installation matrix between the star sensors, and reaches higher measurement accuracy.

The invention provides a training system of back propagation neural network DNN (Deep Neural Network). The training system of the back propagation neural network DNN comprises a first graphics processor assembly, a second graphics processor assembly and a controller assembly, wherein the first graphics processor assembly is used for performing DNN forward calculation and weight update calculation; the second graphics processor assembly is used for perform DNN forward calculation and DNN back calculation; the controller assembly is used for controlling the first graphics processor assembly and the second graphics processor assembly to perform Nth-layer DNN forward calculation respectively according to respective input data, and after the completion of the forward calculation, controlling the first graphics processor assembly to perform the weight update calculation and controlling the second graphics processor assembly to perform the DNN back calculation; and N is a positive integer. The training system provided by the invention has the advantages of high training speed and low data transmission cost, so that the training speed of the back propagation neural network DNN is promoted.

The invention belongs to a method for detecting reservoir fissure development direction in the field of petroleum seismic exploration and oil gas field development, particularly relating to an analytical expression of an included angle between an acquisition system coordinate system and a natural system coordinate system by utilizing multiple wave seismic data, an orthogonality principle formula and an Alford rotation formula; an angle analytical formula inverts the main direction of the fissure and analyzes the fissure development density so as to detect the reservoir fissure development. The angel has a specific physical significance, back calculation is simple and convenient, and an algorithm is not affected by fast and slow shear wave delay time. Even if the polarization of the fast and slow shear wave does not strictly satisfy the relationship of cross polarization, the angle of the inversion angle analytical expression is still within the acceptable error range. The precision of the inversion angle more approaches the real result along with the improvement of the signal to noise ratio of the multi-component data.

The invention relates to a method for estimating the temperature of near-surface air near from MODIS data, which can be used for remote sensing application departments, i.e. weather, environmental monitoring, land management, agricultural monitoring, disaster monitoring, and the like. The method comprises three steps: step1: utilizing the near-surface temperature and the emission rate of each pixel of a MODIS data product and the water vapour content value of atmosphere as priori knowledge and input parameters of atmosphere radiation transmission module simulation software MODTRAN4, carrying out forward direction simulation in different areas and seasons aiming at a 29th thermal infrared band, a 31st thermal infrared band and a 32nd thermal infrared band of each pixel of the obtained remote sensing data MODIS and establishing a training and testing database; step2: utilizing a neural network to repeatedly training and testing a training and testing dataset; step3: carrying out back calculation for practical image data of the MODIS to obtain the temperature distribution condition of the air near the earth's surface of an objective area of the earth's surface. The method can be usedfor weather prognosis, the environmental monitoring, the agricultural monitoring, the disaster monitoring, and the like.

The invention discloses an optical film ultra wide band optical constant testing method for calculating an optical constant of a whole spectrum range from a visible light band to an infrared band of an optical film material. Particularly, a spectral transmission factor of a substrate-film transparent zone and the spectral transmission factor of a non-transparent zone are compounded as a target spectrum in the wavelength range of 0.3 microns-20 microns, a vibrator model is adopted as a chromatic dispersion model of the optical constant, and the ultra wide band optical constant of the film is calculated through target spectrum data back calculation. The optical film ultra wide band optical constant testing method has universality for film materials.

The invention discloses a multi-layered thickness eddy current testing device based on a giant magnetoresistance sensor and an intelligent algorithm. The device generates actuating signals via a signal generator, and provides the actuating signals for built-in GMR and drive coil of magnetic steel after the signals are amplified by a power amplifier. The device detects the magnitude of the eddy current signals based on the built-in GMR and the drive coil of the magnetic steel, and the signals are sent to an amplifying circuit for being amplified after being tested by a detection circuit. The output of the amplifying circuit is collected to a computer by a data acquisition circuit based on a single chip microcomputer to be processed. The signal has a close relation with the thickness of the detected multi-layered conducting materials, carries out back calculation after adopting an improved BP network and can obtain the thickness of the conducting materials of each layer. The device is applicable to the thickness testing of multi-layered conducting structure in fields of machine, printed circuit board, aerospace, national defense and the like.

The invention provides a detection method and device of a track regularity state. The track irregularity is measured through a preset chord measuring method, and the chord measuring value corresponding to the preset measuring position is obtained; according to the incidence relation of the chord measuring value and the track irregularity, a mathematical model between the chord measuring value and the track irregularity is established; back calculation computing is carried out on the chord measuring value according to the mathematical model, and the track geometric shape and location are obtained; and filtering is carried out on the track geometry, and the track irregularity is output. The mathematical model between the track irregularity and the chord measuring value is established again, and the chord measuring value serves as a state quantity for the track irregularity. Back calculation is carried out on the track irregularity through the state quantity. By means of the method, a transfer function can be basically stabilized within 1+ / -0.2, the track irregularity can be measured more accurately, and the principle of least squares is involved in a one-chord and multi-point chord measuring method. The track irregularity is corrected through the multi-point measuring value, and the measuring precision and the measuring resolution ratio of the detection method and device are improved.

The invention discloses a method for preventing shield penetration of a building compact district from inducing ground subsidence by controlling ground loss. The method includes the following steps: firstly, building a set of concept system of ground subsidence 'ground loss' caused by the shield penetration the building compact district; secondly, building mathematical models of ground loss and ground subsidence and ground loss and construction parameters by utilizing the monitoring data of a testing section; thirdly, performing back calculation on a ground loss target according to a subsidence control target of the shield penetration of the building compact district; and finally, proposing construction control parameters of the ground subsidence induced by the shield penetration of the building compact district. The method integrates theoretical analysis, scientific experiments and numerical simulation, is strong in application, wide in monitoring and measuring rage, good in reliability and high in accuracy, and achieves quantified and refined control on the ground subsidence construction parameters of the shield penetration of the building compact district.

The invention discloses a quantitative analysis method for evaluating a dynamic compaction reinforcement effect of foundation by utilizing Rayleigh surface waves. The method comprises compacting pretreatment and compacting post-treatment, namely before compacting, performing field Rayleigh surface wave test according to frequency dispersion curve back calculation, so as to obtain a shear wave velocity Vs at different depths before compacting, performing a compacting test on a soil sample at different depths in the field, and finishing an indoor shear wave velocity test, wherein the testing result comprises a relationship Vs-e between the shear wave velocity and a void ratio, and the maximum shear wave velocity Vsmax; and after compacting, performing field Rayleigh surface wave test according to frequency dispersion curve back calculation, so as to obtain a shear wave velocity Vs' at different depths after compacting, and finally, calculating the compaction degree delta according to the parameters, and realizing evaluation on the dynamic compaction reinforcement effect according to the compaction degree delta. Compared with the prior art, the method has the advantages that the quantization parameter, namely the compaction degree delta, of the dynamic compaction reinforcement effect of the foundation is evaluated due to introduction of quantitative analysis, an evaluation effect of the dynamic compaction reinforcement effect can be quantitatively analyzed, and the method is specific and accurate.

The invention discloses a multi-scale license plate precise locating method based on affine correction. Firstly, candidate license plate plaques are detected by using texture information and color features of license plates, then affine correction is carried out on each candidate license plate, an energy function is defined to accurately locate the positions of the license plates, noise is removed, and the positions of license plates in original images are obtained through back calculation. According to the multi-scale license plate precise locating method based on the affine correction, the license plates can be accurately located from multi-scale images in a complex environment, high robustness, accuracy and high-speed performance are achieved, and the omission rate of the license plates is close to zero. The affine correction is uniquely introduced, the license plate locating problem brought by view angles of the images is well solved, and a basis is established for the accurate locating of the license plates. Meanwhile, the energy function is originally defined to evaluate and accurately locate the license plates, more deep information beneficial to the locating of the license plates is used in the method, and the locating of the license plates is made to be more accurate, more stable and more reliable.

The invention provides a panel control measurement method based on sine and cosine theorem algorithms. The method disclosed by the invention comprises the following steps of: encrypting a new control point C between A and B points which cannot be mutually visualized; utilizing an angle observation set method of an electronic total station and measuring a reciprocating distance to measure a distance between AB and BC and an angle of an angle ACB; utilizing A and B coordinates of a coordinate algorithm to calculate a theoretical distance between the A and B points and utilizing the theoretical distance to identify the precision and the accuracy of measured data of the electronic total station; and finally, utilizing sine and cosine theorems and a back calculation of the coordinates to calculate a plane position of the new control point C. Therefore, the measurement precision is improved and the measurement work is ensured to be smoothly processed.

A method for creating multifunctional cementitious composites that provide load-bearing and self-sensing properties. The method involves dispersing conductive nanomaterials (e.g., multi-walled carbon nanotubes) into a polymer (e.g., latex) material from which a thin film is created and deposited (e.g., sprayed) onto aggregates, which after drying, can be incorporated with cementitious materials and desired liquids and cast, along with sufficient number of electrodes, into a form for curing. After curing, the resultant structure can be electrically tested through the electrodes, for structural characteristics, including determination of damage severity and location using back-calculation utilizing electrical resistance tomography (ERT), or electrical impedance tomography (EIT), to generate a spatial resistivity map (distribution).

The invention discloses a method for measuring the content of free gas in coal bed gas. By calculating the surface water saturation of a fresh coal core and performing back calculation to obtain water saturation in the stratum state, the volume of the free gas in a coal rock of unit mass is measured. Preferably, the method comprises the following steps of: collecting a sample, namely collecting a cylindrical coal sample which is not polluted by drilling fluid; measuring the volume Vtotal of the coal sample; measuring the volume Vwater1 of initial water; measuring the volume Vwater2 of saturated water; measuring an effective burden pressure delta p and a volume coefficient B; and according to parameter results in the preceding steps, calculating the content of the free gas. By adoption of the method, the content of the free gas can be accurately measured, so a total gas content test result is approximate to an underground true value, and great significance is provided for exploration and development of the coal bed gas.

The invention discloses a resistivity anisotropy recognition method in a horizontal well stratum environment. Firstly, an azimuth natural GR and an azimuth resistivity logging are obtained through logging, and then a stratigraphic dip and a stratigraphic azimuth angle are extracted. Secondly, the instrument azimuthal resistivity logging response value at multiple spacings and different frequencies in a horizontal well are acquired through logging. Through the environmental correction and automatic layering process, a false layer is removed, and standardized logging data and a stratum layering interface are obtained. Thirdly, according to a stratigraphic dip extracting result, the layering interface, an apparent resistivity measured value and adjoining well data, a horizontal well stratum model is constructed. Finally, the stratum model is subjected to back calculation, and then an anisotropy coefficient and a stratum resistivity inversion result are outputted. According to the method, the real resistivity information and the anisotropy coefficient information of the stratum in the horizontal well environment are extracted based on azimuthal resistivity logging data while drilling, so that the existing logging evaluation data of the horizontal well are perfected. During the comprehensive interpretation and evaluation process of the logging data of the horizontal well, the reservoir oil saturation and the like can be accurately calculated based on the stratum real resistivity and the anisotropy coefficient. As a result, the method provides reliable parameters for reservoir evaluation.

The invention provides a seismic reservoir prediction method by cross analysis based on target analysis. The method comprises the following steps of: carrying out attribute calculation and back calculation on original seismic data so as to obtain seism preparation data for seism reservoir prediction and preparation data for well measurement; carrying out the cross analysis on well measuring data, thus determining whether a reservoir section on a cross plot of the well measuring data has obvious characteristics; and if so, defining a rational seismic reservoir prediction result according to the scope of parameters obtained from the cross analysis on the well measuring data and the existing data.

The invention relates to a method for measuring load thresholds of geotechnical packing accumulated deformation states under cyclic loading. According to the method, brick walls and sand bags are adopted to construct a filling model, so as to form a testing system along with loading equipment and a displacement sensor, and a cyclic loading test is performed on the filling model, so as to obtain a changing curve of accumulated deformation with loading times; then the changing curve of segmented single loading average deformation with the loading times is obtained through calculation, and a negative power function as shown in the specification is adopted to fit, so as to obtain a relation curve of a power exponent Pi value with the change of load levels; a function relation between the Pi value and the load levels is obtained by virtue of cubic polynomial regression, and the corresponding load levels when P is equal to 2, 1 and 0 are obtained through back calculation, so as to be used as the load thresholds delta t1, delta t2 and delta t3 for distinguishing different evolution states of the accumulated deformation, wherein an experimental basis is respectively provided for design and maintenance of ballastless and ballast track bed structures of railways by the delta t1 and delta t2. The method has the advantages of convenience in model manufacturing, low cost, simplicity in operation and clear distinguishing criterion, and the obtained load thresholds are accurate and reliable.

The invention relates to a control system and method for permanent magnet synchronous electric spindle driving. The system carries out vector control to a sensor-less permanent magnet synchronous motor through the adoption of a position observation module. The position observation module comprises a motor model module, a counter electromotive force estimation module and an angle back calculation device, wherein the counter electromotive force estimation module is realized based on an internal model control principle comprising an electric angular velocity summit. The control system and method for the permanent magnet synchronous electric spindle driving can accurately observe the real angle of a rotor, so that the operation performance of the permanent magnet synchronous electric spindle can be improved.

The invention provides a multi-mode GNSS single-frequency cycle slip detecting and repairing method supporting RTPPP and RTK. The method comprises: an inter-epoch difference is constructed; a time-dependent ionized layer error, a tropospheric error and clock correction delay errors between different systems are eliminated; overall least squares estimation is carried out; if a position correction number covariance DX^X^ is larger than or equal to threshold 1, five satellites are selected circularly and a cycle-slip-free combination with the highest reliability is found out to obtain position and clock difference drift correction number; a new residual vector is obtained by back calculation, wherein the residual vector is an inter-epoch cycle slip floating point solution; and the cycle slipfloating point solution is processed by rounding off in an acceptable range to determine whether the cycle slip can be restored or be a gross error. According to the invention, three different empirical thresholds are set based on different detection quantity meanings, so that the probability of missed determination and false determination during the cycle slip restoring process can be reduced.

The invention discloses a source-tracing method of atmospheric pollutants for an industrial park. The method comprises the following steps: collecting observation data by monitoring points and judgingwhether the observation data is out of limit or not; screening areas according to observation data out of limits; and obtaining observation data obtained by screening and collected in the areas and performing back calculations on a model according to a source position estimation model; obtaining a result and verifying the result. The source-tracing method has the following advantages: concentration data collected by a monitoring station around a diffusion source park is utilized to estimate source intensity magnitude and concrete positions of diffusion sources; and additionally, the method can be suitable for scenarios with variable source intensity and wind field data.

Provided is a vector control apparatus for an induction motor which is capable of obtaining stable rotation in a short period of time for adjustment even at the time of low-speed operation by increasing the SN ratio of an output voltage to suppress the arithmetic error in the primary angular frequency arithmetic value based on a magnetic flux command value calculated from a circuit constant and an actual measurement of the induction motor even in the case where the rotation velocity of the induction motor is lower. The vector control apparatus includes: magnetic flux command value arithmetic means (11) for selecting and outputting the magnetic flux command value on the basis of a torque command value Tmr that is input from the external, the circuit constant of the induction motor, and the actual measurement related to the induction motor; and vector control means (10) for controlling the inverter device (3) on the basis of the magnetic flux command value that is output from the magnetic flux command value arithmetic means (11) and the circuit constant of the induction motor (8). The magnetic flux command value arithmetic means (11) includes a magnetic flux command value back calculation arithmetic section (21) for calculating a magnetic flux command arithmetic value F2RB so that a voltage that is applied to the induction motor 8 becomes equal to or higher than a setting value corresponding to an unstable area.

Provided is a vector control apparatus for an induction motor which is capable of obtaining stable rotation in a short period of time for adjustment even at the time of low-speed operation by increasing the SN ratio of an output voltage to suppress the arithmetic error in the primary angular frequency arithmetic value based on a magnetic flux command value calculated from a circuit constant and an actual measurement of the induction motor even in the case where the rotation velocity of the induction motor is lower. The vector control apparatus includes a magnetic flux command value arithmetic section for selecting and outputting the magnetic flux command value on the basis of a torque command value that is externally input, the circuit constant of the induction motor, and the actual measurement related to the induction motor; and a vector controller for controlling an inverter device on the basis of the magnetic flux command value that is output from the magnetic flux command value arithmetic section and the circuit constant of the induction motor. The magnetic flux command value arithmetic section includes a magnetic flux command value back calculation arithmetic section for calculating a magnetic flux command arithmetic value so that a voltage that is applied to the induction motor becomes equal to or higher than a setting value corresponding to an unstable area.

The invention discloses a method for predicting deformation of a rock slope of a reservoir bank in a storage period. The method includes the following steps: establishing a reservoir bank slope deformation regression analysis predicting model comprising various deformation components, and carrying out regression analysis on slope deformation monitoring data; carrying out back calculation and analysis on rock-mass mechanical parameters and thermal parameters of the reservoir bank by using a numerical simulation and analysis method on the basis of the predicting model; calculating and analyzing deformation of the slope of the bank under the current water level by using the numerical simulation and analysis method on the basis of the rock-mass mechanical parameters and the thermal parameters, which are obtained by back calculation, of the slope of the reservoir bank, evaluating according to the deformation monitoring data, and adjusting parameters according to an evaluating result to obtain the adjusted rock-mass mechanical parameters and the thermal parameters; predicting deformation of the rock slope under a next storage level according to the adjusted rock-mass mechanical parameters and the thermal parameters to obtain a deformation prediction result of the rock slope of the reservoir bank under the next storage level; and inspecting a deformation predicting result. On the basis of the mixed predicting model, the predicting result is accurate and comprehensive.

The invention discloses a narrow-view-field double-camera image fusion method based on a large virtual camera. The narrow-view-field double-camera image fusion method comprises the steps of constructing the large virtual camera according to two single cameras, and constructing geometric imaging parameters of the large virtual camera according to geometric imaging parameters of the two single cameras; constructing corresponding geometric imaging models according to the geometric imaging parameters of the two cameras and the large virtual cameras; calculating and outputting a rational polynomial model coefficient corresponding to the large virtual camera; respectively performing indirect-method geometric correction on images of the two single cameras according to a coordinate forward calculation process and a coordinate back calculation process based on the geometric imaging models to obtain two images under an image coordinate system of the large virtual camera, and obtaining a fused image of the large virtual camera. According to the narrow-view-field double-camera image fusion method based on the large virtual camera, the concept of the large virtual camera is ingeniously used; high-precision fusion of the images of the two cameras in a narrow view field is realized, and the rational polynomial model coefficient corresponding to the large virtual camera is supplied; furthermore, the processing procedure is fully automatic, and manual intervention is not needed; the narrow-view-field double-camera image fusion method is applicable to a ground preprocessing procedure.