60 results about "Polarization scattering matrix" patented technology

The present invention discloses a polarization SAR ground object classification method based on a self-step learning convolutional neural network, in order to mainly solve the problems that the priorart has low accuracy in classifying complex ground object scenes and is heavily affected by noise. The implementation scheme comprises: 1, obtaining a polarization scattering matrix S and a pseudo color RGB image under the Pauli basis from original complete polarization SAR data; 2, constructing a three-dimensional matrix to form a sample set for each pixel, and constructing a training sample setand a test sample set; 3, constructing a convolutional neural network and training the convolutional neural network based on self-step learning to accelerate network convergence and improve the generalization ability of the network; and 4, classifying the test samples by using the trained convolutional neural network to obtain a final complete polarization SAR ground object classification result.According to the method disclosed by the present invention, accuracy for classifying the target ground objects of complex ground object scenes in the polarization SAR image is improved, and the methodcan be used for feature classification and target recognition.

The invention provides a simple and practical indoor polarization scattering measurement system, which is used for obtaining high-polarization-purity scattering data based on a vector network analyzer. The indoor polarization scattering measurement system is formed by the vector network analyzer, antennas, a main control computer, a microwave shielding darkroom and a turntable control cabinet. By filling a part of absorbing materials between the antennas, influence of mutual coupling between the antennas is reduced physically;and after converting broadband step frequency domain data to a time domain, near-end antenna mutual coupling signals are removed by utilizing selectivity of a time domain wave door. High-purity target polarization scattering characteristic data is obtained through a method of setting appropriate parameters and the like for the step frequency signals, and polarization data purity is higher than 45 dB; and the polarization scattering measurement system is simplified, and a lot of production cost is reduced.

The invention discloses a polarized SAR image classification method based on DCCGAN. The method comprises steps of (1) inputting an image; (2) preprocessing data; (3) normalizing a feature matrix andextracting blocks; (4) constructing a data set; (5) constructing a DCCGAN model; (6) training the DCCGAN model; (7) constructing and initializing a discriminant classification network model; (8) training the discriminant classification network model; and (9) predicting classification. The method is not required to decompose the polarized target of the polarized SAR image, can extract the featuresdirectly from a polarization scattering matrix, makes full use of the rich information of the polarized SAR image, and improves the classification precision of the polarized SAR image.

The invention provides a radar target back scattering simulation method influenced by Faraday rotation changes. The method comprises the steps of obtaining four complete polarization SAR compound images of an identical scene after various error influences are calibrated; obtaining a polarization scattering matrix S of a target; calculating a Faraday rotating angle omega caused by an ionized layer TEC value of each impulse transmission moment; calculating a polarization scattering matrix O influenced by the Faraday rotation changes; obtaining a back scattering coefficient influenced by the Faraday rotation changes; putting the back scattering coefficient influenced by the Faraday rotation changes into an SAR echo expression to carry out imaging processing and corresponding analysis. Through the adoption of the method, any influence on the back scattering coefficient and SAR imaging by the Faraday rotation changes can be simulated. In addition, radar target back scattering coefficients influenced by the Faraday rotation changes can be simulated easily and effectively based on existing SAR images.

The invention discloses a microwave photonic radar fully polarimetric radar detection method. The method includes the following steps that: 1, up-conversion processing is performed on optical carriers, complex baseband signals and radio frequency local oscillation signals in an optical domain, so that two orthogonal paths of radio frequency signals, and one path of optical reference signals are generated; 2, the two orthogonal paths of radio frequency signals are respectively sent to a horizontal polarization antenna and a vertical polarization antenna, and the horizontal polarization antennaand the vertical polarization antenna radiate two orthogonal paths of polarized electromagnetic waves, so that the polarized electromagnetic waves irradiate a target, and the horizontal polarization antenna and the vertical polarization antenna receive the echoes of the target, so that two paths of echo signals are obtained; and 3, the two paths of echo signals and the optical reference signals are subjected to deramp and polarization response separation processing in the optical domain, so that four paths of intermediate-frequency analog electrical signals corresponding to four elements in atarget polarization scattering matrix are obtained; and 4, the four paths of intermediate-frequency analog electrical signals are processed, so that target detection information can be obtained. The invention also discloses a microwave photonic radar fully polarimetric radar. With the microwave photonic radar fully polarimetric radar detection method and the microwave photonic fully polarimetric radar of the invention adopted, a problem that the bandwidth of a fully polarimetric radar is limited, the problem of the structural complexity of the fully polarimetric radar and a problem that it isdifficult for the fully polarimetric radar to realize instantaneous measurement can be solved.

The invention discloses an insect characteristic parameter inversion method based on the characteristic values of a polarization power matrix. The research on the polarization invariants of a polarization scattering matrix shows that the two characteristic values lambda 1 and lambda 2 (lambda 1 >= lambda 2) of a Graves power matrix have a strong correlation with the body length and body weight oflarge insects and small insects. On the basis, a method for inverting the body length and body weight of insects using lambda 1 of large insects and lambda 2 of small insects is presented, and a method for distinguishing between large insects and small insects is given. The measurement of body length parameters of insects is realized for the first time. The error of body weight inverted using themethod of the invention is lower than the error of body weight inverted using the traditional method. The ability of VLR (Vertical Looking Radar) to measure the biological parameters of insects is significantly enhanced. The ability to identify the species of migratory insects will be greatly improved. The method is of great significance to the research on the air ecological system and the management of insect pests.

The invention belongs to technical field of signal characteristic control and specifically relates to a method for accurately acquiring near field echo. The method comprises following steps of: a step 1, establishing a target coordinate system O-XtYtZt, an antenna coordinate system O-XaYaZa, and a scattering unit coordinate system O-XmYmZm; a step 2, acquiring a field incident on a mth scattering unit from the surface of a scattering body; a step 3, acquiring scattered fields of the scattering units; a step 4, acquiring a scattered field of the mth scattering unit received by a receiving antenna; a step 5, superposing the scattered fields of each scattering unit received by the receiving antenna in order to acquire a total scattered field; and a step 6, calculating the total scattered field so as to acquire near field echo intensity and power characteristics. The method establishes a complex target near field scattering model with a PO+PTD+GP method and introduces a concept of near field polarization scattering matrix so as to express scattering properties with different mechanisms by using a unified expression, thereby simplifying a calculating process.

The invention discloses a Pauli decomposition and depth residual network-based polarimetric SAR image classification method. The method mainly solves the problems in the prior art that the classification accuracy is low and the neural network cannot be increased more deeply. According to the technical scheme of the invention, the method comprises the steps of inputting a to-be-classified polarimetric SAR image, subjecting a polarization scattering matrix S to Pauli decomposition, and forming a pixel-based feature matrix F; representing an original element value by 28*28 blocks around each element in the feature matrix F and forming an image block-based feature matrix; constructing a training data set D; subjecting the image to super-pixel treatment after the Pauli decomposition treatment and forming a data set T1; constructing a classification model based on a depth residual network; training the classification model by using the training data set so as to obtain a well trained model; inputting the data set T1 into the well trained model to classify the data set T1 and then obtaining a predictive label matrix T2 of the entire image; removing the pixels of the training data set from the matrix T2 and then calculating the accuracy. The method of the invention adopts the depth residual network, so that network layers are increased. Meanwhile, the image is processed by adopting super pixels, so that the features of the image are effectively learnt. The classification accuracy of polarimetric SAR images is improved. The method can be used for target recognition.

The invention discloses a polarization calibration method for a fully polarized multi-input multi-output synthetic aperture radar (FP-MIMO-SAR). The method adopts a rotatable dual antenna active polarization calibrator (RODAPARC) and a standard metal cylinder as a polarization calibration body, and polarization calibration is achieved by the following two steps that firstly an RODAPARC dual-site polarization scattering matrix (PSM) and an active polarization calibrator full polarization (PARC-FP) radar polarization calibration model are utilized to conduct active polarization calibration on Nsets of transmit-receive channels in the FP-MIMO-SAR and four kinds of polarization combinations in each channel to obtain N sets of polarization calibration parameters; then the standard metal cylindrical calibration body is used for passive polarization calibration to optimize the N sets of polarization calibration parameters, and radiometric calibration, crosstalk and channel imbalance calibration of a radar system are achieved.

The invention discloses a polarization feature extraction method based on channel fusion, which is used for radar polarization feature extraction. The invention solves the problem that the inaccuracyof the polarization scattering matrix in the traditional polarization feature extraction method causes the inaccuracy of the polarization feature extinction, thereby affecting the polarization classification and recognition. The processing flow of the fast implementation method is channel fusion + feature extraction. The polarization feature extraction method based on channel fusion does not needto estimate the polarization scattering matrix, avoids the influence of the accuracy of the polarization scattering matrix on the polarization feature extraction. The method does not require matrix operation, and the calculation amount is also greatly reduced, which is suitable for engineering implementation.

The invention discloses a multi-polarization HRRP scattering center extraction method based on multi-vector ANM. By adopting the scheme of the invention, the robust and highly-accurate scattering center extraction can be achieved. The scheme of the invention utilizes the characteristics that the scattering centers of each channel of a multi-polarization high resolution range profile (HRRP) have the same position and different maximum peak values, performs scattering center estimation by using a multi-vector atomic norm minimization model, obtains a polarization scattering matrix of scatteringpoints while accurately obtaining the position information, and provides more comprehensive information for target recognition. The scheme of the invention fully utilizes the common information between the multi-polarization channels, and effectively improves the estimation accuracy; and compared with the traditional P-MUSIC and P-ESPIRIT methods, the scheme of the invention can adapt to the situation of poor signal-to-noise ratios and has better robustness to noise.

The invention belongs to the field of full-polarized SAR (Synthetic Aperture Radar) imaging remote sensing technology, and relates to a full-polarized SAR building falling loss rate estimation method based on rotating domain polarization coherence characteristic. The method comprises the following steps: 1, rotating a polarization scattering matrix around a full-polarized radar line of sight; 2, calculating a polarization coherence characteristic sequence in a rotating domain; 3, calculating a polarization coherence fluctuant degree; 4, calculating a building polarization falling loss factor; and 5, estimating a building falling loss rate. The method is simple in implementation, has good robustness on different building goals, is also very convenient to implement, and can be directly used for processing multi-time-phase full-polarized SAR images acquired by various full-polarized SAR systems. The method has an important reference value in the application fields of SAR building falling loss estimation, disaster reduction and prevention and the like.

The invention discloses a variable polarization monopulse radar target angle estimation method, system and device and a medium. The method comprises the steps of acquiring target echo signals of four channels of a variable polarization monopulse radar; performing two kinds of nonlinear digital beam synthesis according to the target echo signals to obtain a first horizontal polarization component, a first vertical polarization component, a second horizontal polarization component and a second vertical polarization component of each channel, and generating a first high-resolution range profile and a second high-resolution range profile of each channel; determining a sum-difference signal pair of the four polarization channels according to the first high-resolution range profile and the second high-resolution range profile; and determining a target polarization scattering matrix according to the sum-difference signal pair, and estimating a target azimuth angle and a target pitch angle of each polarization channel. The method improves the accuracy of variable polarization monopulse radar target polarization parameter measurement and angle estimation, and can be widely applied to the technical field of radars.

The invention discloses a transmitting and receiving combined polarization optimization method under a time-sharing complete polarization radar system. The method comprises the steps of obtaining a target Kennaugh polarization scattering matrix and a clutter Kennaugh polarization scattering matrix by using obtained target prior polarization information; constructing a transmitting-receiving combined polarization optimization model by using the target Kennaugh polarization scattering matrix and the clutter Kennaugh polarization scattering matrix; obtaining an optimal receiving polarization Jones vector and an optimal transmitting polarization Jones vector by using the receiving polarization Stokes vector and the transmitting polarization Stokes vector based on the transmitting-receiving combined polarization optimization model; according to the optimal transmitting polarization Jones vector, determining time-sharing transmitting horizontal polarization pulses and time-sharing transmitting vertical polarization pulses; and performing signal processing on the time-sharing transmitting horizontal polarization pulses and the time-sharing transmitting vertical polarization pulses in sequence by using the optimal receiving polarization Jones vector to obtain echo power. According to the optimization method, clutter and interference suppression is realized through a polarization optimization method and the like under a time-sharing complete polarization radar system.

The present invention discloses an underground linear target trend measurement method and system based on polarization ground penetration radar. The method comprises following steps: collecting a radar reflection signal of four channels of HH, HV, VH and VV by using a polarization ground penetration radar system, and extracting a polarization scattering matrix; performing Alford rotation on the polarization scattering matrix, and extracting a target response under the HH polarization channel after Alford rotation; conducting derivation to a target response under the HH polarization channel, and finding an angle value when the target response under the HH polarization channel reaches a maximum; and combining a transformation relationship between the target response under the HV polarization channel and a target angle, to expand a range of the angle value, to obtain the underground linear target trend. The method and system adopts the Alford rotation to measure the angle of the linear target, can extract target reflection at any angle between a transmitter and a receiver, and can accurately measure the underground linear target trend and save measurement calculation time.

The invention discloses a polarimetric SAR image change detection method based on a depth confidence network, and the method comprises the steps: inputting two to-be-detected polarimetric SAR images of the same region and different time phases; carrying out the registering of the polarimetric SAR images at two time phases; carrying out the despeckling of the registered images; carrying out the initial artificial marking; respectively solving polarization coherent matrixes TA and TB through the polarization scattering matrixes of the two polarimetric SAR images; respectively extracting diagonal elements of the matrixes, and carrying out the cascading to form a feature matrix F based on pixels; obtaining a feature matrix F1 after normalization; segmenting each element in the feature matrix F1 into a block, and forming a feature matrix F2 based on the image blocks; obtaining the feature matrix D1 of a training data set D and a feature matrix T1 of a test data set T according to F2; constructing a detection model based on a depth confidence network; carrying out the training of the detection model through the constructed data set; carrying out the detection of the to-be-detected images through the trained detection model. The method improves the detection precision.

The invention relates to an attitude-measurable and adjustable polarized radar passive calibrator and a radar calibration method. The polarized radar passive calibrator comprises an air maneuvering platform, control equipment of the air maneuvering platform, a dihedral angle and MEMS inertial navigation equipment, wherein the dihedral angle and the MEMS inertial navigation equipment are installed on the air maneuvering platform. The position and attitude of a dihedral angle on an air maneuvering platform are adjusted through control equipment, relative attitude measurement data of the dihedral angle relative to a polarized radar are obtained through MEMS inertial navigation, and the polarized radar is calibrated by the control equipment based on a polarized radar passive calibration technology according to the relative attitude measurement data. The polarization scattering matrix of the radar calibrator can be accurately obtained, ground clutter interference can be avoided by adjusting the position of the calibrator, and the radar calibrator can be flexibly applied to polarization radar calibration in different environments.

The invention discloses a simultaneous polarization measurement method based on an orthogonal intermittent sampling modulation radar signal. The method comprises the steps that 1, a signal source generates linear frequency modulation signals and carries out radio frequency modulation; 2, AD sampling is carried out on the received two paths of polarized echo signals, then the signals are separated through an intermittent sampling emission time sequence, and four paths of polarized echo signals are obtained; and 3, pulse compression processing is carried out on the four paths of signals by using a matched filter of the complete linear frequency modulation signal. And 4, windowing of a pulse compression result is performed, and false peaks are removed to obtain polarization scattering matrix estimation of the target. According to the method disclosed by the invention, the emission of two paths of signals can be completed by using the same signal source in a switch switching mode, so that the hardware cost of the system is greatly saved; the two paths of intermittent sampling emission time sequence signals utilize diversity on a signal time domain to realize ideal signal isolation similar to time-sharing polarization measurement, and coherence of target polarization characteristic measurement can be ensured.