Spatial imaging method of micromotion characteristic mode based on first-order field correlation and target micromotion model

Abstract

A method for spatial imaging of target micro-motion characteristic modes based on a first-order field correlation imaging system, comprising the following steps: calculating a complex-valued radiation field matrix of a target surface based on a complex-valued matrix of an emitted radiation field; obtaining a one-dimensional electrical signal using a coherent detector; performing time-frequency characteristic analysis and processing on the one-dimensional electrical signal based on the time-frequency vibration characteristics of the target; constructing complex-valued models of discrete targets and continuous targets based on target characteristics; performing first-order field correlation on the processed signal and complex-valued radiation fields of different target surfaces to obtain spatial distributions of vibration characteristics of different targets; the present invention utilizes the one-dimensional vibration characteristics of the target to be analyzed, utilizes the radiation field with spatial characteristics to generate spatiotemporal modulation with the vibrating target surface, and reconstructs the spatial distributions of the time-varying vibration characteristics of the discrete target and continuous target planes through a coherent correlation method.

Description

Technical Field

[0001] The present invention relates to a device and method for spatial imaging of target micro-motion characteristic modes, in particular to a method for spatial imaging of micro-motion characteristic modes based on first-order field correlation and target micro-motion model. Background Art

[0002] Micro-motion is a kind of motion feature that is widely present on the surface of the target. These vibration detection methods can be used in many scenarios, such as vital sign detection in earthquake relief, or vibration of large building structures. Usually vibrating objects contain two-dimensional information: time dimension change and spatial dimension distribution. The existing time information perception system can reconstruct the time-frequency variation characteristics of the vibrating object in the field of view by analyzing the micro-Doppler characteristics, while the spatial information cannot be effectively reconstructed. Although a scheme for detecting quasi-spatial vibration information using spatial array detectors has been proposed, it is still limited by imaging efficiency. A detection method that simultaneously perceives time and space dimension information is needed. By regulating the formation of space-varying and time-varying radiation patterns,

[0003] Intensity correlation imaging is to detect the echo signal through the radiation field modulated in time and space, and obtain the target image through computational reconstruction. The ghost imaging system breaks through the limitation of antenna aperture on imaging resolution, and has the advantages of forward-looking, staring, and quick-shot imaging. It is widely used in staring observations in key areas and unmanned systems, and has been verified in many bands, such as X-ray, microwave, and terahertz optical bands.

[0004] In the optical segment, second-order correlation is currently mainly used, and some first-order correlation systems can achieve phase imaging for static complex-valued targets. In static target correlation imaging, the target is modeled as a complex-valued reflectivity distribution, but the micro-moving target can be regarded as a time-varying complex-valued model in the detection system. Traditional static phase imaging cannot achieve accurate imaging of micro-moving targets. Summary of the invention

[0005] In order to overcome the deficiencies of the above-mentioned prior art, the present invention proposes a device and method for spatial imaging of micromotion characteristic modes based on first-order field correlation and target micromotion model. By establishing the micromotion target model as a dynamic complex reflectivity distribution, since the target micromotion can essentially be regarded as a two-dimensional temporal and spatial modulation of the radiation field, an attempt is made to recover from the echo phase change, which is expected to solve the problem of simultaneous temporal and spatial imaging of micromotion. Considering that light waves are essentially electromagnetic waves, this method can also be applied to other bands except for the differences caused by wavelength characteristics.

[0006] The basic idea of ​​the present invention is to use the first-order correlation of the radiation field to restore the static complex-valued plane characteristics of the target, analyze the echo based on the target micro-motion model, obtain the time-varying characteristics of the target complex-valued plane, and then use different types of target models as judgment conditions to compensate or eliminate the time-varying characteristics of the target complex-valued plane, making it equivalent to a static complex-valued plane, and then use the first-order field correlation characteristics of the radiation field to restore the spatial distribution of the target micro-motion mode.

[0007] The technical solution of the present invention is as follows:

[0008] A method for spatial imaging of target micro-motion characteristic modes based on first-order field association, which is characterized by including:

[0009] Step 1: Obtain the radiation speckle field E modulated by the spatial modulator s (x s ,t), and calculate the complex-valued reference radiation speckle field of the target surface;

[0010] S1.1. Record the speckle field generated by the main controller and the complex value matrix of the radiation field modulated by the radiation source;

[0011] S1.2. Establish a radiation field transfer function including wavelength and propagation distance parameters, and calculate the complex reference radiation speckle field of the target surface based on the transfer function;

[0012] Step 2: Obtain the local oscillator radiation field E LO (t), the local oscillator radiation field is obtained by inputting a radiation source modulated by a time signal into a frequency shifter;

[0013] Step 3: The radiation speckle field E s (x s ,t) After free propagation in space, it reaches the target plane. The radiation speckle field undergoes two-dimensional temporal and spatial modulation with the complex-valued time-varying target. After being reflected by the complex reflectivity target, it is received by the receiving system to obtain the radiation field at the receiving end;

[0014] Step 4: After the local oscillator radiation field and the receiving end radiation field are coherent in the coherent detector, the signal spectrum peak is moved away from the zero-frequency noise, and a one-dimensional electrical signal i(t) containing target micro-motion information is output, and the formula is as follows:

[0015]

[0016] Where η is the receiving efficiency of the receiver;

[0017] Step 5: Perform Fourier transform on the one-dimensional electrical signal i(t), obtain target micro-motion characteristic parameters in the time-frequency spectrum, including initial phase, amplitude and frequency, and construct a virtual modulation model;

[0018] Step 6: Reconstruct spatial images of different target models:

[0019] For discrete vibrating targets: use the virtual modulation model and the complex-valued reference radiation speckle field generated in step 1 to perform two-dimensional temporal and spatial modulation, and perform first-order field correlation with the one-dimensional electrical signal obtained in step 4, and finally match the reflectivity image of the target micro-motion feature object in the field of view;

[0020] For continuous vibration targets: use the number of micro-motion modes existing in the field of view obtained in the time-frequency spectrum, and convert the number of sampling points contained in one cycle of different micro-motion modes according to the vibration periods corresponding to the different vibration modes displayed in the time-frequency spectrum. For extracting different micro-motion modes, set different sampling intervals to sample the one-dimensional electrical signal obtained in step 4. If it is necessary to obtain the motion changes of all micro-motion modes in one cycle, that is, the complex-valued time-varying characteristics of the entire target plane, then set the value of the sampling interval to the lowest common multiple of the number of samples in all micro-motion cycles; if it is necessary to obtain the spatial distribution of a single micro-motion mode, then it is necessary To find the vibration period of the corresponding mode in the time-frequency spectrum obtained in step 4, convert the number of sampling points contained in the period, set the value of the sampling interval to the number of sampling points corresponding to the period of the micro-motion mode, and sample the one-dimensional electrical signal obtained in step 4 at the same time. Perform first-order field correlation on the sampled one-dimensional electrical signal and the corresponding complex-valued reference radiation field obtained in step 1, and under the ensemble average, the matching expectation value of other modes is 0, and the spatial distribution of a single mode will be completely extracted under the first-order correlation. Finally, the result of spatial image reconstruction is the spatial distribution of the expected target mode.

[0021] Compared with the prior art, the technical effects of the present invention are as follows:

[0022] The spatial phase imaging method of the micro-motion modality provided by the present invention not only detects the time-dimensional variation characteristics within the detection field of view, but also can reconstruct the spatial distribution corresponding to the time-dimensional micro-motion characteristics using the first-order field correlation. Considering that the traditional first-order field correlation has a phase reconstruction algorithm for the static complex-valued plane, compared to the traditional phase reconstruction algorithm, this technology proposes a new extraction method for the time-varying complex-valued characteristics of the vibrating target, thereby reconstructing the plane state at different times within the micro-motion cycle without being affected by the blurred imaging results caused by the time-varying characteristics. At the same time, this technology can be applied to detection systems of different bands, and can reconstruct micro-motions of different types of scenes according to the wavelength of the radiation field. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a structural schematic diagram of an embodiment of an imaging device of first-order field correlation of the present invention.

[0024] 1- Radiation source 2- Waveform generator 3- Time modulator 4- Spatial modulator 5- Transmitting array 6- Time-varying complex-valued target 7- Single point receiver 8- Frequency shifter 9- Coherent detector 10- Main controller / computer.

[0025] Figure 2 It is a flowchart of the interval sampling method for continuous target modality imaging.

[0026] Figure 3 It is a flowchart of the virtual modulation method for discrete target modal imaging. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] The present invention is aimed at discrete models. Since there is no spatial correlation, a time-frequency characteristic curve is virtually modulated into each resolution unit in the calculation reference radiation field, thereby screening discrete targets in the field of view that meet the vibration mode. Time-frequency analysis of the echo signal at the receiving end can obtain the time-frequency characteristics of the micro-motion in the target field of view. The virtually modulated calculation reference arm is essentially a conjugate match for the entire target surface, thereby eliminating the time-varying characteristics of the target, thereby screening out scattering points that meet the modified vibration characteristics. For continuous targets, different sampling intervals are set for the one-dimensional echo signal to obtain a new sampling sequence. In the new sampling sequence, the target plane can be equivalent to a stationary complex-valued plane. At the same time, by setting different initial sampling points, the vibration changes of the continuous target plane over time can be obtained.

[0029] like Figure 1 As shown, the detection system includes a transmitting system, a receiving system, a photoelectric conversion system and a main controller 10. The radiation field generated by the radiation source is modulated in the time dimension by the time modulator 3, and the generated radiation field with time-dimensional variation characteristics is divided into two paths through a beam splitter. One path enters the frequency shifter 8, and the radiation field output after frequency shift is recorded as the local oscillator radiation field E LO (t), and the other path enters the spatial modulator 4, which adds a specific spatial modulation of the speckle pattern to the radiation source with time dimension variation, and outputs a radiation field with two-dimensional characteristics of time and space, which is recorded as E s (x s ,t).

[0030] The radiation field E emitted by the transmitting system has two-dimensional characteristics of time and space s (xs ,t) field propagates freely in space to reach the target plane. The radiation field undergoes spatiotemporal two-dimensional modulation with the complex-valued time-varying target. After being reflected by the complex reflectivity target, the radiation field signal at the receiving end is received by the receiving system 7. The receiving system is composed of a single-point detector, and the radiation field received by it is recorded as the signal field as E xr (t). This signal field is the one-dimensional electrical signal i(t) screened out after the local oscillator radiation field is coherent in the coherent detector. Its mathematical expression can be written as:

[0031]

[0032] Where η is the receiving efficiency of the receiver. The motion parameters and position information of the target are obtained by Fourier transforming the one-dimensional electrical signal i(t). Assuming that the point detector is located at the center of the radiation source, and the radiation source surface and the target surface are parallel to each other, the expression of i(t) can be written as

[0033]

[0034] Where P(x o ,t) represents the radiation field speckle distribution on the target surface, represents the real reflectivity of the target plane, λ c is the carrier wavelength, f IF represents the offset frequency of the frequency shifter, z0 represents the distance from the center of the vibration plane to the single-point receiver, v represents the translational velocity, W(x o ,t) represents the distribution function of the lateral propagation of the target plane micro-motion mode, α, β are the azimuth and elevation angles of the radar respectively, α Q ,β Q represents the azimuth and elevation angles of the vibration direction of the vibration scattering center in the reference coordinate system, then A(α,β,α Q ,β Q ) represents the position relationship function of the two planes under the influence of azimuth and elevation angles of the source and target surfaces, and its specific expression is:

[0035] A(α,β,α Q ,β Q ) = cosβcosβ Q cos(α-α Q )+sinβsinβ Q

[0036] The phase difference caused by the longitudinal distance can be reflected in the propagation function between the two longitudinal planes (light source plane and target plane) The expression can be written as

[0037]

[0038] where k λ is the wave vector. For translation information, the target is considered as a whole without axis rotation. The scattering points of the whole target maintain the same translation state. For micro-motion information, in the standing wave system, each scattering point maintains the complex function W(x o ,t) constraint. And ω(t) represents the background noise. For the target model, it is simplified according to the following conditions

[0039] When the radar elevation angle and azimuth angle are set to 0

[0040] α=0,β=0

[0041] When the vibration azimuth angle and pitch angle relative to the radar of the target center point Q are also 0

[0042] α Q =0,β Q =0

[0043] This article focuses on the micro-motion mode of the target and the influence of the spatial main vibration model on the micro-motion characteristics, so the translation of the target is set to 0.

[0044] v=0

[0045] To simplify the scenario, we only discuss two-dimensional targets and do not discuss target depth issues.

[0046] z=0

[0047] Ideally, the noise floor is not considered

[0048] ω(t)=0

[0049] Then the echo signal can be simplified as:

[0050]

[0051] Then, by performing a fast Fourier transform on the one-dimensional electrical signal, we can obtain the distance information z of the target. o , since the detection system is a reflective detection system, z o It can also represent the distance from the transmitter to the target surface. Then the propagation function h(x s ,x o ) can be obtained by calculation, and its mathematical expression is

[0052]

[0053] Then the radiation field on the target surface can be regarded as the radiation source field E s (x s ,t) through the propagation function h(x s ,x o) after propagation, the radiation field of the target surface obtained by calculation can be used as the reference radiation field, denoted by E c (x c ,t).

[0054] At this time, the one-dimensional electrical signal and the reference radiation field are first-order correlated, which is essentially the first-order field correlation between the received signal radiation field and the reference radiation field. When the target model is a static complex value model, that is, the target mathematical model is Then the first-order field correlation function can directly obtain the static target complex value plane, and the expression is:

[0055]

[0056] However, since the micro-motion target is a time-varying complex-valued target, directly performing first-order field association will blur the plane image of the micro-motion target. Therefore, in order to eliminate the time-varying characteristics of the target, the present invention first proposes two different types of target micro-motion models. According to the ratio of the speckle field size and the target size in the detection far field, the target is divided into a discrete target model and a continuous target model. The discrete target should be smaller than the size of a single speckle in the radiation field of the target surface. At this time, the target is regarded as a particle, and its micro-motion characteristics are the micro-motion characteristics of the entire object. For discrete targets, assuming that there are k discrete points in the field of view, the surface integral can be equivalent to the superposition of each discrete point. For the kth scattering point, its complex-valued reflectivity function can be written as

[0057]

[0058] The Delta function represents the spatial position of the kth scattering point, η(t) represents the time dimension change of the target micromotion, and the periodic motion η(t) can be expanded by Fourier series, so for the entire target surface, we have

[0059]

[0060] according to Figure 2 The imaging strategy shown in the figure is to find the vibration frequency, amplitude and initial phase of a certain vibration mode in the time-frequency spectrum. x,n ,z x,n ,φ x,n , then its corresponding spatial distribution should be In order to filter out the target image, the phase vibration characteristics of the target are essentially eliminated so that the target reflectivity image will not be blurred during the ensemble averaging process. At this time, the reference arm virtual modulation function It can be written as

[0061]

[0062] Then when k = x, the first-order correlation function is

[0063]

[0064] Under the ensemble average, the projection residuals of other non-orthogonal micro-motion modes on the target mode in the above formula will approach zero infinitely, and the final result of the first-order field correlation function can be written as:

[0065]

[0066] The mode distribution of the vibration target explained in the discrete target can be reconstructed in terms of spatial distribution.

[0067] For multiple discrete targets, they are regarded as different scattering points without spatial correlation in the field of view; at the same time, a continuous target will be illuminated by multiple specks in the target radiation field during the detection process, so each scattering point in the field of view should be regarded as multiple scattering points with spatial correlation. Similarly, for a continuous target, and the lateral propagation conforms to the standing wave propagation, assuming that the two-dimensional reflection coefficient in the entire field of view is Assume that there are m main vibration modes in space. Due to the standing wave system, each main vibration mode corresponds to a periodic vibration η m (t),

[0068]

[0069] Similarly, for the periodic vibration η m (t) can be expanded into an n-order Fourier series

[0070]

[0071] Since η m (t) Characterize the target periodic motion to eliminate the spatial vibration characteristics, then according to Figure 3 The strategy shown in the first row of , when it is necessary to study the intra-periodic changes of the m main vibration modes of the entire target plane, that is, to study the complex value time-varying characteristics of the entire target plane, set the sampling interval N t is the lowest common multiple of the number of samples in all micro-motion mode cycles, then the one-dimensional electrical signal obtained by sampling at the new sampling interval can be recorded as i w (t), and the speckle pattern of the target surface corresponding to each sampling point in the same time axis is recorded as P m,n (x c ,t), and the corresponding reference radiation field is recorded as E w (x c ,t). Since the new one-dimensional signal obtained by interval sampling makes the target phase term not vibrate with time, it can be expressed in mathematical language as

[0072] η m (t) = η m (t+nN t )

[0073] Where n is the sampling point sequence number, assuming that the photoelectric receiver output signal contains a total of N s sampling points, then there are

[0074]

[0075] When the first sampling point is t0, the first-order field correlation is equivalent to the phase imaging of the static complex-valued target at time t0. At this time, the complex-valued phase image of m main vibration modes at time t0 can be obtained by the first-order field correlation, and its mathematical expression can be written as

[0076]

[0077] At this time, the reconstructed image is the image of the m main vibration modes in the target surface at time t0. Different starting sampling points t0 are selected, and t0∈[1,N t ], the periodic changes of the target plane within a period can be obtained.

[0078] If you need to obtain the spatial distribution of a single micro-motion mode, such as Figure 2 As shown in the second row of , it is necessary to find the vibration period of the corresponding mode in the time-frequency spectrum obtained by time-frequency analysis of the one-dimensional electrical signal, convert the number of sampling points contained in the period, and convert the sampling interval N p,q The value of is set to the number of sampling points corresponding to the period of the micro-motion mode, and the one-dimensional electrical signal is sampled at the sampling interval. The one-dimensional electrical signal obtained by sampling at the new sampling interval can be recorded as i p,q (t), and the target surface speckle pattern corresponding to each sampling point in the same time axis is recorded as P p,q (x o ,t), and the corresponding reference radiation field is recorded as E p,q (x c ,t). It is similar to the first-order field correlation of discrete targets. For the micro-motion mode when m=p,n=q, according to the coherent correlation theory, the spatial distribution of this type of micro-motion mode can be completely preserved. For the residual of other non-orthogonal coherent modes, under the ensemble average of infinite sampling, its expected value is infinitely close to 0 and can be completely eliminated.

[0079] The above is a description of the specific embodiments of the present invention. It should be understood that the present invention is not limited to a specific system, and its imaging principle can be applied to imaging in a variety of fields, such as optical systems, microwave systems, terahertz systems or X-ray systems. Those skilled in the art can make various deformations or modifications within the scope of the claims, which does not affect the essence of the present invention.

Claims

1. A method for spatial imaging of target micro-motion characteristic modes based on first-order field association, characterized in that: include: Step 1: Obtain the radiation speckle field E modulated by the spatial modulator s (x s ,t), and calculate the complex-valued reference radiation speckle field of the target surface; S1.

1. Record the speckle field generated by the main controller and the complex value matrix of the radiation field modulated by the radiation source; S1.

2. Establish a radiation field transfer function including wavelength and propagation distance parameters, and calculate the complex reference radiation speckle field of the target surface based on the transfer function; Step 2: Obtain the local oscillator radiation field E LO (t), the local oscillator radiation field is obtained by inputting a radiation source modulated by a time signal into a frequency shifter; Step 3: The radiation speckle field E s (x s ,t) After free propagation in space, it reaches the target plane. The radiation speckle field undergoes two-dimensional temporal and spatial modulation with the complex-valued time-varying target. After being reflected by the complex reflectivity target, it is received by the receiving system to obtain the radiation field at the receiving end; Step 4: After the local oscillator radiation field and the receiving end radiation field are coherent in the coherent detector, the signal spectrum peak is moved away from the zero-frequency noise, and a one-dimensional electrical signal i(t) containing target micro-motion information is output, and the formula is as follows: Where η is the receiving efficiency of the receiver; Step 5: Perform Fourier transform on the one-dimensional electrical signal i(t), obtain target micro-motion characteristic parameters in the time-frequency spectrum, including initial phase, amplitude and frequency, and construct a virtual modulation model; (a) For discrete vibration targets, a conjugate matching function is generated based on the micro-motion characteristic parameters to eliminate the time-varying characteristics; (b) For continuous vibration targets, the sampling interval is set according to the micro-motion mode period to be equivalent to the static complex value plane; Step 6: Reconstruct spatial images of different target models: For discrete vibrating targets: use the virtual modulation model and the complex-valued reference radiation speckle field generated in step 1 to perform two-dimensional temporal and spatial modulation, and perform first-order field correlation with the one-dimensional electrical signal obtained in step 4, and finally match the reflectivity image of the target micro-motion feature object in the field of view; For continuous vibration targets: use the number of micro-motion modes existing in the field of view obtained in the time-frequency spectrum, and convert the number of sampling points contained in one cycle of different micro-motion modes according to the vibration periods corresponding to the different vibration modes displayed in the time-frequency spectrum. For extracting different micro-motion modes, set different sampling intervals to sample the one-dimensional electrical signal obtained in step 4. If it is necessary to obtain the motion changes of all micro-motion modes in one cycle, that is, the complex-valued time-varying characteristics of the entire target plane, then set the value of the sampling interval to the lowest common multiple of the number of samples in all micro-motion cycles; if it is necessary to obtain the spatial distribution of a single micro-motion mode, then it is necessary To find the vibration period of the corresponding mode in the time-frequency spectrum obtained in step 4, convert the number of sampling points contained in the period, set the value of the sampling interval to the number of sampling points corresponding to the period of the micro-motion mode, and sample the one-dimensional electrical signal obtained in step 4 at the same time. Perform first-order field correlation on the sampled one-dimensional electrical signal and the corresponding complex-valued reference radiation field obtained in step 1, and under the ensemble average, the matching expectation value of other modes is 0, and the spatial distribution of a single mode will be completely extracted under the first-order correlation. Finally, the result of spatial image reconstruction is the spatial distribution of the expected target mode.

Images