Moving target trajectory prediction device and method based on single-pixel imaging

Abstract

The invention discloses a moving target trajectory prediction device and method based on single-pixel imaging, and relates to the field of motion trajectory prediction.In the device, an optical collection component collects reflected light of a moving target to a spatial light modulator; the spatial light modulator performs spatial coding modulation on the received light, and reflects the modulated light to the first detection light path assembly and the second detection light path assembly respectively to realize single-pixel detection to obtain a corresponding first light intensity signal and a corresponding second light intensity signal; the trajectory prediction component performs differential processing on the first light intensity signal and the second light intensity signal, and realizes moving target positioning according to the light intensity signals subjected to differential processing by adopting a Fourier single-pixel positioning method to obtain a moving trajectory position; and inputting the motion track position into a preset track prediction model for track prediction to obtain a future track position of the motion target. According to the invention, single-pixel positioning can be carried out on the moving target, and accurate track prediction is realized.

Description

Technical Field

[0001] The present application relates to the field of motion trajectory prediction, and in particular to a device and method for predicting the trajectory of a moving target based on single-pixel imaging. Background Art

[0002] Single-pixel imaging is a new imaging method that measures the fluctuations of light field intensity and phase and reconstructs images using the correlation between the detected light field and the modulated light field. It has the advantages of anti-interference, ultra-sensitivity, and super-diffraction limit. The unique imaging method of single-pixel imaging makes most of its research objects static targets, and the research focus is also on how to improve the imaging quality and efficiency of static targets. However, there is still a lack of single-pixel detection, positioning, and trajectory prediction solutions for moving objects, but many practical applications are inseparable from the detection of moving objects. The relative motion between the object and the imaging system will inevitably affect the detection results. Positioning and trajectory prediction of moving targets are important components of the engineering and practical application of single-pixel imaging technology, and have great potential application value in laser radar, space remote sensing, security monitoring, and autonomous driving. How to perform single-pixel positioning and trajectory prediction of moving targets is a key issue that needs to be solved at present. Summary of the invention

[0003] The purpose of the present application is to provide a moving target trajectory prediction device and method based on single-pixel imaging, which can use single-pixel positioning for moving targets and achieve accurate trajectory prediction.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a moving target trajectory prediction device based on single-pixel imaging, comprising an optical collection component, a first detection optical path component, a second detection optical path component, a spatial light modulator, and a trajectory prediction component;

[0006] The optical collection component is used to collect the reflected light of the moving target to the spatial light modulator;

[0007] The spatial light modulator is used to perform spatial coding modulation on the received light, and reflect the modulated light to the first detection optical path component and the second detection optical path component respectively;

[0008] The first detection optical path component and the second detection optical path component are both used to implement single-pixel detection to obtain a corresponding first light intensity signal and a second light intensity signal;

[0009] The trajectory prediction component is used to: perform differential processing on the first light intensity signal and the second light intensity signal; use Fourier single pixel positioning method to locate the moving target according to the light intensity signal after differential processing to obtain the motion trajectory position; input the motion trajectory position into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

[0010] In a second aspect, the present application provides a moving target trajectory prediction method based on single-pixel imaging, comprising:

[0011] Build a moving target trajectory prediction device based on single-pixel imaging and set the moving target;

[0012] Collecting the reflected light of the moving target to a spatial light modulator through an optical collection component;

[0013] Performing spatial coding modulation on the received light by the spatial light modulator, and reflecting the modulated light to the first detection optical path component and the second detection optical path component respectively;

[0014] Implementing single-pixel detection through the first detection optical path component and the second detection optical path component to obtain a corresponding first light intensity signal and a second light intensity signal;

[0015] The first light intensity signal and the second light intensity signal are differentially processed by the trajectory prediction component, and then the moving target is positioned according to the light intensity signal after the differential processing by using the Fourier single pixel positioning method to obtain the moving trajectory position;

[0016] The trajectory prediction component inputs the motion trajectory position into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

[0017] According to the specific embodiments provided by the present application, the present application has the following technical effects: the present application provides a moving target trajectory prediction device and method based on single-pixel imaging, which realizes single-pixel detection through the setting of the first detection optical path component, the second detection optical path component and the trajectory prediction component, and realizes differential processing of the two light intensity signals obtained by detection, thereby obtaining the light intensity signal to be used subsequently. Then, the moving target in the scene is positioned by the Fourier single-pixel positioning method, and the motion preset trajectory prediction model is built by the LSTM network to perform trajectory prediction, and accurate trajectory prediction coordinates are obtained, that is, accurate trajectory prediction is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of the structure of a moving target trajectory prediction device based on single-pixel imaging provided in one embodiment of the present application.

[0020] Figure 2 A schematic diagram of trajectory prediction results provided in one embodiment of the present application.

[0021] Figure 3 A schematic diagram of trajectory prediction error results provided in one embodiment of the present application.

[0022] Figure numerals: 1 - moving target, 2 - optical collection component, 3 - optical focusing component, 4 - single pixel detector, 5 - spatial light modulator, 6 - trajectory prediction component. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] In order to make the purpose, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0025] In an exemplary embodiment, Figure 1 As shown, a moving target trajectory prediction device based on single-pixel imaging is provided, comprising an optical collection component 2, a first detection optical path component, a second detection optical path component, a spatial light modulator 5 and a trajectory prediction component 6. The first detection optical path component and the second detection optical path component both comprise an optical convergence component 3 and a single-pixel detector 4 which are arranged in sequence.

[0026] The optical collection component 2 is used to collect the reflected light of the moving target 1 to the spatial light modulator 5, for example, it can be collected on the micro-mirror surface of a digital micromirror array (DMD).

[0027] The spatial light modulator 5 is used to perform spatial coding modulation on the received light, and reflect the modulated light to the first detection optical path component and the second detection optical path component respectively; the first detection optical path component and the second detection optical path component are both used to realize single-pixel detection to obtain the corresponding first light intensity signal and second light intensity signal; specifically, the optical converging component 3 collects the light modulated by the spatial light modulator 5 to the single-pixel detector 4 to detect the light intensity signal, which is then used for subsequent trajectory prediction.

[0028] The trajectory prediction component 6 is used to: perform differential processing on the first light intensity signal and the second light intensity signal; use Fourier single pixel positioning method to locate the moving target according to the light intensity signal after differential processing to obtain the motion trajectory position; input the motion trajectory position into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

[0029] In an application example, the trajectory prediction component 6 is also used to: after obtaining the motion trajectory position, use the Kalman filter algorithm to perform denoising on the motion trajectory position, and then input it into a preset trajectory prediction model for trajectory prediction. That is, the present application uses the trajectory prediction component 6 to locate, denoise, and predict the trajectory of the moving target in the scene based on the light intensity signal after differential processing.

[0030] In an application example, the moving object 1 may be a moving object display.

[0031] In an application example, the spatial light modulator 5 is a digital micromirror array, and the digital micromirror array performs spatial coding modulation through a Fourier array. During the spatial coding modulation process, the spatial light modulator 5 reflects the modulated light to the first detection optical path component and the second detection optical path component along the ±R direction; wherein R represents the angle with the optical axis direction of the spatial light modulator 5, which can be 24°.

[0032] Based on the same inventive concept, the embodiment of the present application also provides a moving target trajectory prediction method based on single-pixel imaging. The implementation solution provided by the method to solve the problem is similar to the implementation solution recorded in the above-mentioned device, so the specific limitations of one or more method embodiments provided below can refer to the limitations of the device above, and will not be repeated here. The moving target trajectory prediction method based on single-pixel imaging includes the following steps 100-600.

[0033] Step 100: construct a moving target trajectory prediction device based on single pixel imaging and set the moving target, such as Figure 1 shown.

[0034] Step 200, collecting the reflected light of the moving target to a spatial light modulator through an optical collection component; specifically, the light containing the moving target information can be focused on a digital micromirror array through an imaging lens.

[0035] Step 300, spatially code modulate the received light through the spatial light modulator, and reflect the modulated light to the first detection optical path component and the second detection optical path component respectively; specifically, the digital micromirror array reflects the modulated light signal along ±24° during the spatial code modulation process, collects it on the single-pixel detector through the optical convergence module, and records the light intensity.

[0036] Step 400: Implement single-pixel detection through the first detection optical path component and the second detection optical path component to obtain corresponding first light intensity signals and second light intensity signals.

[0037] Step 500, the first light intensity signal and the second light intensity signal are differentially processed by the trajectory prediction component, and then the Fourier single pixel positioning method is used to locate the moving target according to the light intensity signal after the differential processing to obtain the motion trajectory position.

[0038] Among them, Fourier single-pixel imaging was originally proposed for image acquisition. Its characteristic is that the spatial light modulation is performed using the Fourier basis (i.e., sinusoidal intensity) pattern, so that the spatial information of the object is encoded into a one-dimensional temporal light signal. Then, by measuring the intensity of the generated light signal with a single-pixel detector, Fourier single-pixel imaging can restore the Fourier spectrum of the target image, and the target image can be obtained by performing an inverse Fourier transform on the restored Fourier spectrum. Each Fourier basis image P(x, y) consists of its spatial frequency pair (f x ,f y ) and the initial phase Characterization can be expressed as the following function:

[0039]

[0040] Where (x, y) is the two-dimensional coordinate in the spatial domain, A is the average intensity of the pattern, and B is the contrast. The Fourier basis pattern is used to modulate the illumination light field or the detection light field, and the obtained light intensity D is equal to the inner product of the target object image I and the Fourier basis P, which can be represented by the following function:

[0041]

[0042] The integral in the formula shows that the Fourier transform is a global to point transform. Specifically, each coefficient in the Fourier domain is contributed by all points in the spatial domain. Therefore, any change in the spatial domain affects all coefficients in the Fourier domain. Using this property, the presence or motion of objects in the scene can be detected by monitoring the changes in one or several Fourier coefficients in the Fourier domain. It can be done by using Fourier single pixel imaging, because Fourier single pixel imaging obtains Fourier coefficients by using Fourier basis speckle for spatial light modulation.

[0043] By modulating the illumination light field with a Fourier basis speckle, changes in single pixel measurements can be observed when an object enters the scene or begins to move in the scene. In this application, two Fourier coefficients are used for target positioning. Since only two Fourier coefficients need to be measured instead of the complete Fourier spectrum, this application is efficient in data acquisition and is conducive to the detection of fast moving targets. In the process of moving target tracking, the linear phase shift characteristic of the Fourier transform is utilized. An image with a displacement of (x0, y0) in the spatial domain has a phase shift of (-2πf x x0,-2πf y y0):

[0044]

[0045] Among them, (f x ,f y ) represents the spatial frequency coordinate in Fourier domain, Represents the Fourier spectrum of the moving target image I(x,y), F -1 In order to obtain the displacement of the target object, this application uses background subtraction, which can be achieved by using replace to complete. are coefficients obtained from a background frame, which refers to a frame obtained before the moving object enters the scene or starts to move.

[0046] The displacement of an object can be expressed by the phase term Get. Since there are two unknowns (x0 and y0) in the formula, a system of equations consisting of two equations can be established to solve the two unknowns. This linear system of equations requires two Fourier coefficients. The acquisition of Fourier coefficients can refer to the n-step phase shift method. Specifically, this application selects the three-step phase shift method because this method has high data acquisition efficiency and is highly robust to noise and light fluctuations. In order to use the three-step phase shift method to obtain the Fourier coefficients Three Fourier-based speckles are required. The three Fourier-based speckles have the same spatial frequency pair (f x ,f y ), but the initial phase is different, that is, The Fourier coefficients are obtained through three corresponding single-pixel detection values.

[0047]

[0048] In this method, one frame is defined as one positioning of a moving target object. Since target positioning requires the use of two Fourier coefficients to locate the coordinates in two directions respectively, and each coefficient uses three patterns, only six Fourier basis speckles are required for one frame. The target can be positioned by the corresponding six single-pixel detection values. However, the Fourier basis pattern is grayscale, and the digital micromirror device DMD can only generate binary patterns. Therefore, the present application uses binary Fourier basis speckles for spatial light modulation, and uses the Floyd-Steinberg dithering method to binarize the positioning speckles. The Fourier positioning speckles after Floyd-Steinberg dithering are shown in the figure. The positioning speckles after dithering are composed of 0 and 1, corresponding to the state "0" and state "1" of the DMD respectively.

[0049] Assuming that there is no target moving object at the beginning, the single-pixel measurement value should be constant in theory, but it actually varies within a very small range. This change may be caused by noise and ambient light. Once a moving object enters the scene or starts to move, it will cause a large change in the single-pixel measurement. Once a moving object is detected, the object will be located. Based on this, the Fourier single-pixel positioning method is used to locate the moving target according to the light intensity signal after differential processing to obtain the position of the motion trajectory, including: according to the light intensity signal after differential processing, the Fourier coefficient is calculated using the following formula, and the displacement is calculated by the Fourier coefficient:

[0050]

[0051] Among them, f x When it is 0, the Fourier coefficient is obtained f y When it is 0, the Fourier coefficient is obtained D represents the light intensity signal after differential processing, the subscript of D represents the initial phase of the Fourier basis, which are 0, 2π / 3, and 4π / 3 respectively; j represents the imaginary number symbol, (x0, y0) represents the displacement of the moving target image in the spatial domain, and arg{} is the argument of the complex number.

[0052] In an application example, after step 500, the moving target trajectory prediction method based on single-pixel imaging further includes: performing denoising processing on the moving trajectory position by using a Kalman filter algorithm through the trajectory prediction component, and then executing subsequent step 600.

[0053] Among them, the essence of Kalman filtering is a recursive variance minimum estimation method. It can grasp the change law of the estimated quantity by solving and analyzing the system state quantity, and then process the observation data at the current moment and the state update value at the previous moment by constructing the mathematical model of the Kalman filtering algorithm to obtain the state update value at the current moment. The processing flow of the Kalman filtering algorithm is divided into two parts: prediction and estimation. Through multiple cycles and iterations, the error is continuously corrected to finally obtain the optimal estimation result. The basic equation of Kalman filtering is as follows:

[0054] 1. Time update:

[0055] Compute the state one-step forecast:

[0056] One-step forecast mean square error:

[0057] 2. Measurement update:

[0058] Filter gain:

[0059] State Estimation:

[0060] The estimated mean square error is: P k =(IK k H k ) k / k-1 .

[0061] in: One-step prediction for the state; is the state estimation; P k / k-1 is the one-step prediction error variance matrix; K k is the filter gain matrix; P k is the estimated error variance matrix. and P0, through the five basic equations of Kalman filtering, according to t k Measurement of time k , we can recursively calculate the state estimate at time k When performing state estimation, we need to use measurement z k Correction, K k That is, the filter gain matrix representing the correction gain. And determine K k The process indicator is to minimize the variance of the estimation, so the Kalman filter is a linear minimum variance estimation.

[0062] Taking into account the influence of noise in the process of determining the center of mass by the single-pixel center of mass detection method, this application uses a Kalman filter algorithm to denoise the obtained center of mass coordinates, thereby improving the accuracy of center of mass positioning and thereby improving the accuracy of subsequent trajectory prediction.

[0063] Step 600: input the motion trajectory position into a preset trajectory prediction model through the trajectory prediction component to perform trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

[0064] In an application example, the moving target trajectory prediction method based on single-pixel imaging also includes:

[0065] When the reflected light of the moving target within a preset time length is collected to the spatial light modulator by the optical collection component, the corresponding multiple motion trajectory positions are obtained by the trajectory prediction component (for example: each moment within the preset time length corresponds to a motion trajectory position); the multiple motion trajectory positions are input into a preset trajectory prediction model by the trajectory prediction component for trajectory prediction to obtain the future trajectory position of the moving target; the future trajectory position is a trajectory position at a future moment or multiple trajectory positions at future moments.

[0066] LSTM (Long Short-Term Memory) is a deep learning model for processing sequence data. Its characteristic is that it can control the flow of information through its internal logic gates, so as to better process long sequence data. LSTM introduces three logic gates, namely input gate, output gate and forget gate. These logic gates can autonomously select which information needs to be passed on and which information needs to be forgotten, so as to effectively process long sequence data.

[0067] Specifically, each time step in the LSTM has a hidden state h t and a cell state c t . LSTM can remove or add "cell state" information through the "gate" structure. Through the sigmoid layer, a probability value between 0 and 1 is output, which describes how much each part can pass through. 0 means "variables are not allowed to pass", and 1 means "variables are allowed to pass". The core of LSTM is the cell state, which is represented by a horizontal line that passes through the cell. It runs through the entire cell and ensures that information remains unchanged as it flows through it. LSTM is controlled by three gates, namely the forget gate, input gate, and output gate.

[0068] The forget gate is processed by the sigmoid unit, which checks the hidden state h at the previous moment. t-1 and the current input information x t , outputs a value between 0 and 1, indicating whether to keep or discard the cell state c t-1 Which information in the forget gate is 0, which means all information is discarded, and 1 means all information is retained. The formula of the forget gate is as follows: t =σ(W f ·[h t-1,x t ]+b f ). Among them, f t represents the output of the forget gate, h t-1 represents the hidden state at the previous moment, x t represents the input information at the current moment, and σ(···) represents the sigmoid unit.

[0069] The input gate determines what new information is added to the cell state. First, h t-1 and x t Update the information through the sigmoid unit. Then, h t-1 and x t After the tanh unit, a new candidate cell state is created for subsequent updating of cell information. The output gate formula is as follows: t =σ(W i ·[h t-1 ,x t ]+b i );

[0070]

[0071] Then update the cell information. The update rule is that the forget gate first selects a part of the past cell information to forget, and then the input gate selects a part of the candidate cell information to obtain the new cell information. The formula is as follows:

[0072] After the cell state is updated, h is required t-1 and x t To determine the characteristics of the output hidden state, we need to pass the input through the sigmoid unit of the output gate to obtain the judgment condition, and then pass the cell state through the tanh unit to get a value between -1 and 1, and multiply this value by the judgment condition of the output gate to get the output of the hidden state. The output gate formula is as follows: t =σ(W o ·[h t-1 ,x t ]+b o );h t =o t tanh(C t ).

[0073] In summary, after differential processing of the light intensity signals detected by two single-pixel detectors, the present application locates the moving target in the scene through the Fourier single-pixel positioning method. Considering the noise in the positioning process of the single-pixel centroid detection method and the error influence of the binary speckle, the Kalman filter algorithm is used to denoise the obtained centroid coordinates to improve the accuracy of centroid positioning, thereby improving the accuracy of subsequent trajectory prediction. Compared with the hidden unit of RNN, the internal structure of the hidden unit of LSTM is more complex. In the process of information flowing along the network, LSTM can selectively add or reduce information by increasing intervention. LSTM network can solve the problem of long-term dependence and has long-term memory ability for historical information. It is more suitable for application in the problem of moving target trajectory prediction. Therefore, this application uses LSTM network to build a moving target trajectory prediction model, uses the historical coordinates after Kalman filter denoising for trajectory prediction, and outputs more accurate trajectory prediction coordinates based on the comprehensive Kalman filter prediction results. Figure 2 It is a schematic diagram of trajectory prediction results, where the blue curve represents the historical motion trajectory position (i.e., the positioning result obtained by single-pixel positioning) and the orange coordinates represent the future trajectory position (i.e., the trajectory prediction result). Figure 3 This is a schematic diagram of the trajectory prediction error results, where the horizontal axis represents the number of positioning frames and the vertical axis represents the number of trajectory error pixels.

[0074] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0075] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A moving target trajectory prediction device based on single-pixel imaging, characterized in that: The moving target trajectory prediction device based on single-pixel imaging includes an optical collection component, a first detection optical path component, a second detection optical path component, a spatial light modulator and a trajectory prediction component; The optical collection component is used to collect the reflected light of the moving target to the spatial light modulator; The spatial light modulator is used to perform spatial coding modulation on the received light, and reflect the modulated light to the first detection optical path component and the second detection optical path component respectively; The first detection optical path component and the second detection optical path component are both used to implement single-pixel detection to obtain a corresponding first light intensity signal and a second light intensity signal; The trajectory prediction component is used to: perform differential processing on the first light intensity signal and the second light intensity signal; The Fourier single-pixel positioning method is used to locate the moving target according to the light intensity signal after differential processing to obtain the motion trajectory position; the motion trajectory position is input into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

2. The moving target trajectory prediction device based on single pixel imaging according to claim 1, characterized in that: The first detection optical path component and the second detection optical path component both include an optical converging component and a single pixel detector arranged in sequence; When working, the optical converging component converges and collects the light modulated by the spatial light modulator to the single-pixel detector to detect the light intensity signal.

3. The moving target trajectory prediction device based on single pixel imaging according to claim 1, characterized in that: The spatial light modulator is a digital micromirror array, and the digital micromirror array performs spatial coding modulation through a Fourier array.

4. The moving target trajectory prediction device based on single pixel imaging according to claim 1, characterized in that: During the spatial coding modulation process, the spatial light modulator reflects the modulated light along the ±R directions to the first detection optical path component and the second detection optical path component respectively; Wherein, R represents the angle with the optical axis direction of the spatial light modulator.

5. The moving target trajectory prediction device based on single pixel imaging according to claim 1, characterized in that: The trajectory prediction component is also used to: after obtaining the motion trajectory position, use a Kalman filter algorithm to perform denoising on the motion trajectory position, and then input it into a preset trajectory prediction model.

6. A moving target trajectory prediction method based on single pixel imaging, characterized in that: The moving target trajectory prediction method based on single pixel imaging comprises: Build a moving target trajectory prediction device based on single-pixel imaging as described in any one of claims 1 to 5, and set a moving target; Collecting the reflected light of the moving target to a spatial light modulator through an optical collection component; Performing spatial coding modulation on the received light by the spatial light modulator, and reflecting the modulated light to the first detection optical path component and the second detection optical path component respectively; Implementing single-pixel detection through the first detection optical path component and the second detection optical path component to obtain a corresponding first light intensity signal and a second light intensity signal; The first light intensity signal and the second light intensity signal are differentially processed by the trajectory prediction component, and then the moving target is positioned according to the light intensity signal after the differential processing by using the Fourier single pixel positioning method to obtain the moving trajectory position; The trajectory prediction component inputs the motion trajectory position into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the preset trajectory prediction model is obtained by training a long short-term memory network.

7. The moving target trajectory prediction method based on single pixel imaging according to claim 6 is characterized in that: After obtaining the motion trajectory position, the moving target trajectory prediction method based on single-pixel imaging further includes: performing denoising processing on the motion trajectory position by using a Kalman filter algorithm through the trajectory prediction component.

8. The moving target trajectory prediction method based on single pixel imaging according to claim 6, characterized in that: The moving target trajectory prediction method based on single pixel imaging also includes: When the reflected light of the moving target within a preset time period is collected to the spatial light modulator by the optical collection component, a corresponding plurality of moving track positions are obtained by the track prediction component; The trajectory prediction component inputs multiple motion trajectory positions into a preset trajectory prediction model for trajectory prediction to obtain the future trajectory position of the moving target; the future trajectory position is a trajectory position at a future moment or multiple trajectory positions at future moments.

9. The moving target trajectory prediction method based on single pixel imaging according to claim 6, characterized in that: The Fourier single pixel positioning method is used to locate the moving target based on the light intensity signal after differential processing to obtain the motion trajectory position, including: According to the light intensity signal after differential processing, the Fourier coefficient is calculated using the following formula, and the displacement is calculated using the Fourier coefficient: in, The Fourier spectrum of the moving target image I(x,y), (f x ,f y ) represents the spatial frequency coordinate in the Fourier domain, f x When it is 0, the Fourier coefficient is obtained f y When it is 0, the Fourier coefficient is obtained D represents the light intensity signal after differential processing, the subscript of D represents the initial phase of the Fourier basis, which are 0, 2π / 3, and 4π / 3 respectively; j represents the imaginary number symbol, (x0, y0) represents the displacement of the moving target image in the spatial domain, arg{} is the argument of the complex number, are coefficients obtained from a background frame, which refers to a frame obtained before the moving object enters the scene or starts to move.

Images