A space target detection imaging device under a complex starry sky background

Abstract

The invention provides a space target detection imaging device under a complex starry sky background, which includes a signal transmitter and a received signal decoupling component. It is characterized in that the signal transmitter is a laser light source, an achromatic wave plate, and a linear polarizer arranged coaxially in sequence. The received signal decoupling component includes a signal receiving control box and a computer. A reflector is arranged inside the signal receiving control box. A first beam splitter, a collimating lens group, and a second beam splitter are arranged coaxially in sequence on the right side of the reflector. A first CMOS camera is arranged at a position corresponding to the first beam splitter. A third beam splitter is arranged at a position parallel to the second beam splitter. A color wheel mechanism is arranged at a corresponding position on the right side of the second beam splitter and the third beam splitter. A second CMOS camera and a third CMOS camera are respectively arranged at corresponding positions on the right side of the color wheel mechanism and the second beam splitter and the third beam splitter. The second CMOS camera and the third CMOS camera are controlled by a CMOS driving circuit, and the CMOS driving circuit is connected to the computer. By organically combining the brightness information, spectral information, polarization information, and orbital information of light, the target recognition detection probability, success rate, and accuracy are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of target detection and recognition, and particularly to a spatial target detection imaging device under a complex starry sky background. Background Art

[0002] Regarding the target detection technology under a complex starry sky background, there are already many image processing processes to perform the task of target recognition from the perspective of software algorithms. However, if the accuracy is improved from the acquisition of data, that is, from the source of image data, it can bring greater success rate and accuracy to target recognition and detection.

[0003] Polarization detection technology and spectral imaging technology are new remote sensing detection technologies developed in recent years. Compared with traditional optical detection systems, the combination of polarization measurement and spectral measurement can provide more dimensional information of the detected object. Based on obtaining the polarization information of the detected object, the target of the detected object can be reconstructed and enhanced. Moreover, polarization detection has a unique advantage in distinguishing artificial targets from complex backgrounds.

[0004] However, the multi-dimensional data fusion brings problems such as large computational amount and long processing time. Summary of the Invention

[0005] The purpose of the present invention is to propose a spatial target detection imaging device under a complex starry sky background, which performs fusion calculation on the brightness, polarization, and spectral information of the target, effectively improving the probability, success rate, and accuracy of target recognition and detection.

[0006] To achieve the above purpose, the present invention proposes a spatial target detection imaging device under a complex starry sky background, including a signal transmitter and a received signal decoupling component. It is characterized in that the signal transmitter is a laser light source, an achromatic wave plate, and a linear polarizer arranged coaxially in sequence. The received signal decoupling component includes a signal receiving control box and a computer. A reflector is arranged inside the box of the signal receiving control box. A first beam splitter, a collimating lens group, and a second beam splitter are arranged coaxially in sequence on the right side of the reflector. A first CMOS camera is arranged at a position corresponding to the first beam splitter. A third beam splitter is arranged at a position parallel to the second beam splitter. A color wheel mechanism is arranged at a corresponding position on the right side of the second beam splitter and the third beam splitter. A second CMOS camera and a third CMOS camera are respectively arranged at corresponding positions on the right side of the color wheel mechanism and the second beam splitter and the third beam splitter. The second CMOS camera and the third CMOS camera are controlled by a CMOS driving circuit, and the CMOS driving circuit is connected to the computer.

[0007] Furthermore, a rotating shaft is arranged at the center of the back of the color wheel mechanism, and the color wheel mechanism is installed on a controllable motor through the rotating shaft.

[0008] Further, the color wheel mechanism is connected to the CMOS driving circuit through a synchronization circuit, and the synchronization circuit controls the synchronization between the shooting of the second CMOS camera and the third CMOS camera and the rotation of the color wheel mechanism.

[0009] Further, the bottoms of the reflecting mirror, the first beam splitter, the second beam splitter, and the third beam splitter are all adjustably mounted on the bottom of the box body of the signal receiving and control box through rotating shafts.

[0010] Further, the color wheel mechanism is an eight-channel color wheel mechanism.

[0011] Further, the present invention also proposes a method for identifying a space target of the space target detection and imaging device under the complex starry sky background, which is characterized by including the following steps:

[0012] Step A1: Actively emit laser through the signal transmitter to make linearly polarized light at a specified angle exit to the target.

[0013] Step A2: Transmit multiple frames of images captured by the first CMOS camera to a computer, perform target orbit calculation, and obtain orbit information.

[0014] Step A3: Rotate the color wheel mechanism and transmit the eight-channel image data captured by the second CMOS camera and the third CMOS camera to the computer.

[0015] Step A4: Use a feature preprocessing classifier to select the clearest image.

[0016] Step A5: Perform parameter decoupling to obtain the brightness, spectrum, and polarization parameters of the image.

[0017] Step A6: According to the energy weighting method, combine the information of brightness, spectrum, and polarization with the orbit information to form a target parameter set.

[0018] Step A7: Judge whether there are unclear or lacking parameters. If there are no lacking parameters, enter Step A8; if there are unclear or lacking parameters, first judge whether they can be obtained through recalculation. If they can, repeat Step A5. If they cannot be obtained through recalculation, output an error.

[0019] Step A8: Compare the target parameter set with the known data set. If known information is matched, it is confirmed that the captured object is a certain pre-coded target, and the matched target result is output. If no known information is matched, enter Step A9.

[0020] Step A9: Sort out the parameters, add a new number to the target, and add it to the database.

[0021] 11. Further, in step A3, the eight-channel images are respectively at polarization angles of 0°, 45°, 90°, and 135°, and the spectral bands are near ultraviolet (<400 nm), near infrared (>780 nm), visible light band (400 - 780 nm), and unconditional.

[0022] 12. Further, in step A5, the method for obtaining the spectral parameters includes the following steps:

[0023] Step G1: Simultaneously obtain continuous and narrow-band spectral band images within a spectral coverage range of at least 0.4 - 2.5 μm.

[0024] Step G2: Preprocess the images and perform relative reflectance image conversion.

[0025] Step G3: Extract spectral features to obtain the l(x, y, λ) model.

[0026] Further, in step G2, the relative reflectance conversion uses a conversion method based on a regression statistical equation: ρ b = ∑F σ ·R e，b = aG i，b + C;

[0027] ρ b is the relative reflectance; G i，b is the image gray value; R e，b is the spectral reflectance value at the e wavelength point within the band b range; F σ is the weighting coefficient related to the sensor spectral response function, ∑F σ = 1; a is the slope; C is the intercept.

[0028] Further, in step A5, the method for obtaining the polarization parameters includes the following steps:

[0029] Step P1: Extract polarization information and calculate the Stokes vector parameters S0, S1, S2, and S3.

[0030] Step P2: Further calculate the degree of linear polarization, polarization azimuth angle, and polarization ellipticity.

[0031] Compared with the prior art, the advantages of the present invention are as follows: 1. Aiming at the deficiencies and defects of the prior art, the present invention provides a target detection imaging system based on the combination of polarization spectral imaging and orbital information acquisition with high target detection probability and wide application range. 2. The two optical paths enter the high-sensitivity camera system through the symmetric dual channels in the eight-channel color wheel structure, enabling synchronous shooting at different polarization angles and different spectral band ranges, reducing the measurement time and improving the measurement efficiency. Description of the Drawings

[0032] Figure 1 This is a schematic structural diagram of the spatial target detection imaging device under a complex starry sky background in the present invention.

[0033] Figure 2 This is a schematic structural diagram of the eight-channel color wheel mechanism in the present invention.

[0034] Figure 3 This is a simplified flowchart of the spatial target recognition method based on the spatial target detection imaging device under the complex starry sky background in the present invention. Detailed implementation manners

[0035] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be further described below.

[0036] In the description of the present invention, it should be noted that for orientation terms, such as the terms "center", "horizontal", "vertical", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., the orientation and position relationships indicated are based on the orientation or position relationships shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and should not be construed as limiting the specific protection scope of the present invention.

[0037] In the present invention, unless otherwise clearly specified and defined, for the terms "assembled", "connected", "coupled", they should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection; it can be directly connected, or connected through an intermediate medium, and can be internally connected and communicated between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0038] In the following paragraphs, the present invention will be described more specifically by way of example with reference to the accompanying drawings. According to the following description, the advantages and features of the present invention will be clearer. It should be noted that the drawings are all in a very simplified form and use non-precise scales, only for the purpose of conveniently and clearly assisting in explaining the objectives of the embodiments of the present invention.

[0039] Such as Figure 1As shown in the figure, the present invention provides a space target detection imaging device under a complex starry sky background, which includes a signal transmitter 1 and a received signal decoupling component 3. It is characterized in that the signal transmitter 1 is a laser light source 11, an achromatic wave plate 12, and a linear polarizer 13 arranged coaxially in sequence. The received signal decoupling component 3 includes a signal receiving control box and a computer 313. A reflector 301 is arranged inside the signal receiving control box. A first beam splitter 302, a collimating lens group 303, and a second beam splitter 304 are arranged coaxially in sequence on the right side of the reflector 301. A first CMOS camera 312 is arranged at a position corresponding to the first beam splitter 302. A third beam splitter 305 is arranged at a position parallel to the second beam splitter 304. A color wheel mechanism 306 is arranged at a corresponding position on the right side of the second beam splitter 304 and the third beam splitter 305. A second CMOS camera 308 and a third CMOS camera 309 are respectively arranged at corresponding positions on the right side of the color wheel mechanism 306 and the second beam splitter 304 and the third beam splitter 305. The second CMOS camera 308 and the third CMOS camera 309 are controlled by a CMOS driving circuit 311, and the CMOS driving circuit 311 is connected to the computer 313.

[0040] During operation, first, according to information such as a known navigation system, a laser is emitted by a signal transmitter 1 to a specified detection area. Among them, the laser light emitted by the laser light source 11 passes through an achromatic waveplate 12 and a linear polarizer 13 and then is emitted into free space 2. The laser passing through free space 2 is collected and analyzed by the signal receiving control box of the signal receiving decoupling component 3. The signal receiving control box is an opaque, stable, constant-temperature, and constant-humidity box. First, through a reflector 301, the light beam is adjusted to the best range that can be received by the system optical path, and the light is reflected to a first beam splitter 302. The first beam splitter 302 divides the optical path into two paths. One path is used for a first CMOS camera 312 to take multiple images, which are transmitted to a computer 313 for data preprocessing. The multiple pictures are superimposed to obtain a result, and the orbital information of the target is calculated through software algorithms. The orbital information is used to describe those target objects not recorded in the existing dataset. The other light beam is transmitted to a collimating lens group 303, and the collimating lens group 303 collimates the light beam so that the light beam is transmitted to a second beam splitter 304. The second beam splitter 304 divides the optical path into two paths again. One light beam passes through an eight-channel color wheel mechanism 306 and is imaged on a second CMOS camera 308. The other light beam is transmitted to a third beam splitter 305 and passes through the symmetric side channel of the eight-channel color wheel mechanism 306 under the reflection of the third beam splitter 305 and is imaged on a third CMOS camera 309. The second CMOS camera 308 and the third CMOS camera 309 are controlled by a CMOS driving circuit 311. The CMOS driving circuit 311 is connected to the computer 313. An operator can control the second CMOS camera 308 and the third CMOS camera 309 to take pictures at the computer 313 end. This device forms a dual optical path through the second beam splitter 304 and the third beam splitter 305. Through the dual optical path synchronous detection technology, combined with the sequential sampling technology, that is, the images in the two cameras are sequentially sampled and combined in the order of time, a frame rate higher than that which can be achieved by using only a single camera originally can be obtained. Theoretically, it is twice the frame rate of using only a single camera, so as to obtain a higher-quality and clearer shooting result.

[0041] A rotating shaft is provided at the center of the back of the color wheel mechanism 306. The color wheel mechanism 306 is installed on a controllable motor 307 through the rotating shaft. An operator can control the motor through the computer 313 during operation, so as to control the rotation speed and rotation direction of the color wheel mechanism 306. The controllability of the color wheel can solve the problems of inconvenient color wheel replacement or manual rotation, and improve the accuracy, thereby improving the overall detection and recognition efficiency and accuracy.

[0042] The color wheel mechanism 306 is connected to the CMOS drive circuit 311 through the synchronization circuit 310 to control the synchronization between the shooting of the second CMOS camera 308 and the third CMOS camera 309 and the rotation of the color wheel mechanism 306. If the synchronization cannot be guaranteed, then the two CMOS cameras may not be able to capture the images of the light beam passing through the color wheel mechanism 306 very accurately, and it may be necessary to capture a very large number of images. Through the CMOS drive circuit 311, the rotation of the color wheel mechanism 306 and the cooperation between the two CMOS cameras can be accurately controlled, ensuring that when the color wheel mechanism 306 rotates to the required angle, the CMOS cameras capture correspondingly, thereby improving the shooting efficiency and accuracy, and thus improving the efficiency of the entire detection and recognition.

[0043] The bottom of the reflector 301, the first beam splitter 302, the second beam splitter 304 and the third beam splitter 305 are all adjustably mounted on the bottom of the box body of the signal receiving and control box through rotating shafts. The purpose of setting the rotating shafts is to make it more convenient to adjust each mirror. By rotating the rotating shafts, the angles of each mirror can be controlled more conveniently, so as to adjust the appropriate reflection angle according to the incident angle of the light beam.

[0044] The color wheel mechanism 306 is an eight-channel color wheel mechanism 306. The eight channels are four polarization angles (0°, 45°, 90°, 135°) and four spectral bands (<400nm, >780nm, 400 - 780nm, unconditional). The two optical paths enter the high-sensitivity camera system through two symmetric channels in the eight-channel color wheel structure, enabling synchronous shooting with different polarization angles and different spectral band ranges, reducing the measurement time and improving the measurement efficiency.

[0045] A method for identifying space targets of a space target detection and imaging device under a complex starry sky background includes the following steps:

[0046] Step A1: Actively emit laser through the signal transmitter 1 to make linearly polarized light with a specified angle exit to the target, and adjust the parameters of the achromatic wave plate 12 to emit linearly polarized light with δ1 = 0°, δ2 = 45°, δ3 = 90°, δ4 = 135°. When the detection system starts to run, the first thing to do is to obtain images. Under the guidance of the navigation system, the signal transmitter 1 completes the task of actively emitting a laser beam, making linearly polarized light with a specified angle exit to the target. The laser beam serves to illuminate the target.

[0047] Step A2: Transmit multiple frames of images captured by the first CMOS camera 312 to the computer 313 for target orbit calculation. Obtain the orbit information. Through the reflection of the mirror 301 and the beam splitting of the first beam splitter 302, transmit multiple frames of images of the specified detection area captured by the first CMOS camera 312 to the computer 313, calculate the target orbit, and input the orbit information of the target to the measurement system. The orbit information is used for pairing the recognition results or supplementing information for unrecognized targets.

[0048] Step A3: Rotate the color wheel mechanism 306 and transmit the eight-channel image data captured by the second CMOS camera 308 and the third CMOS camera 309 to the computer 313. Another optical path of the first beam splitter 302 is transmitted to the collimating lens group 303 to collimate the measured laser beam. Then, through the beam splitting of the second beam splitter 304 and the third beam splitter 305, perform dual-path synchronous detection. The dual paths pass through the symmetric channels in the color wheel mechanism 306, and finally generate eight-channel images on the second CMOS camera 308 and the third CMOS camera 309: polarization angles = 0°, 45°, 90°, and 135°, and the spectral bands are near ultraviolet (<400 nm), near infrared (>780 nm), visible light band (400 - 780 nm), and unconditional, and transmit the image data to the computer 313.

[0049] Step A4: Use the feature preprocessing classifier to select the clearest image. Among many results, use the feature preprocessing classifier to select the optimal image. This method enables preliminary screening of the image input before entering the subsequent data fusion stage. The addition of the feature preprocessing classifier can improve the recognition accuracy and reduce the computational amount of the system. The feature preprocessing classifier belongs to a computer software that can analyze various different features of the image and classify them according to the requirements of the operator, thus making the operation of selecting images more convenient.

[0050] Step A5: Perform parameter decoupling to obtain the brightness, spectral, and polarization parameters of the image.

[0051] The Mueller matrix and Stokes parameters can be used to describe a specific optical system. Before the system conducts measurement activities, system calibration needs to be carried out first. That is, measure with polarized light of known parameters to obtain the instrument matrix of the system. For an optical system composed of multiple optical devices, the total Mueller matrix M of the system can be obtained by multiplying the Mueller matrices of each optical device. system

[0052] M system =M N ·M N-1 ·...·M2·M1

[0053] Then, use the measured outgoing light S out to inversely calculate the original Stokes parameters S of the incident light in , which is expressed by the formula as:

[0054] S out = M system · S in

[0055] In this paper, the idea of Stokes polarization imaging is adopted to extract the polarization information of an object, and the process is as follows:

[0056] In this paper, based on 12 and 13 as basic devices to form Stokes polarization imaging, the Stokes vector of the outgoing light can be expressed as:

[0057]

[0058] In the formula, represents the Mueller matrix of the polarizer, represents the Mueller matrix of the achromatic wave plate 12, and their specific forms are as follows:

[0059]

[0060]

[0061] The magnitude of the total light intensity I has the following mathematical relationship with the four parameters of the Stokes vector:

[0062]

[0063] Among them, m0, m1, m2, and m3 are the four elements in the first row of the system matrix, and their specific mathematical expressions are:

[0064] m0 = 1

[0065]

[0066]

[0067]

[0068] When using the Stokes vector method, it is necessary to construct at least four imaging equations with unknown parameters S0, S1, S2, and S3. The coefficients of the constructed equations are functions related to (φ, α, δ). φ is the angle between the transmission axis of the polarizer and the reference axis; α is the angle between the fast axis of the wave plate and the reference axis, and δ is the phase retardation of the wave plate. The constructed equations can be expressed in matrix form as:

[0069] M i · S in = I i

[0070] Then the Stokes vector S of the incident light in can be obtained by finding the inverse matrix of the coefficient matrix M:

[0071] S in =(M i ) -1 ·I i

[0072] Substitute four different sets of (φ, α) values, namely (0°, 0°), (0°, 45°), (45°, 135°), (90°, 0°), then S in can be expressed as:

[0073]

[0074] After obtaining S0, S1, S2, S3, normalization is usually performed to make S0 equal to 1, and the other three parameters are scaled by corresponding multiples. Next, calculate:

[0075]

[0076]

[0077]

[0078] The basis of spectral imaging is to simultaneously obtain a large number of continuous, narrow-band spectral band images within a spectral coverage of 0.4 - 2.5 μm or longer, where each pixel can form a reflectance spectral curve. In this article, the light beam carrying information passes through the filter channels of the color wheel mechanism 306, and the scientific-grade CMOS lens captures the relatively independent two-dimensional image information in different spectral bands. The original imaging spectral data contains not only the spectral radiation information of the spatial target, but also the influence of factors such as sensor drift, radiation transfer effects, and the atmosphere. Therefore, it is necessary to preprocess the image and perform relative reflectance image conversion. The relative reflectance image conversion uses a conversion method based on a regression statistical equation:

[0079] ρ b =∑F σ ·R e，b =aG i，b +C

[0080] In the formula, ρ b is the relative reflectance; G i，b is the image gray value; R e，b is the spectral reflectance value at the e wavelength point within the band b range; F σ is the weighting coefficient related to the sensor spectral response function, ∑F σ= 1; a is the slope; C is the intercept.

[0081] After the relative reflectance image conversion is completed, spectral feature extraction can be carried out. Spectral images can usually be represented by a spatial-spectral data cube. Each layer in the cube is a band, and the attribute values of each band for each pixel form a spectral vector. By analyzing the spectral vector, the spectral curve of the image can be extracted to achieve the purpose of classification and feature extraction. Spectral absorption feature parameters include: the wavelength position, width, depth, slope, and symmetry degree of the absorption peak. The peaks and valleys in the spectral curve can be used to express the absorption and reflection characteristics of the detection target. For the spectral curve, the binary coding method can be used to emphasize its shape characteristics. The binary coding uses a single threshold processing, that is, the attribute values of each band for each pixel are compared with a certain threshold. If it is less than the threshold, it is assigned "0", and if it is higher than the threshold, it is assigned "1". Since the imaging curves of the same type of target in spectral imaging are generally similar, but the positions of the characteristic bands may shift. The extended coding method can be used to reduce the influence of the characteristic band shift. That is, on the basis of binary coding, the values of "1" are also assigned to the positions before and after a certain position coded as "1", and then the feature matching analysis is carried out.

[0082] For the detection of space targets, their spectral imaging will have a certain degree of similarity. To solve the classification problem, a numerical index and a shape index are combined to comprehensively determine the differences between spectral curve vectors. This difference is based on the spectral difference curve. Let P and Q be a spectral curve vector respectively, and n be the spectral dimension. The subtraction operation is performed point by point on P and Q to obtain a new vector NS with the dimension still being n. NS is called the spectral difference curve of P and Q. When the similarity between P and Q is very large, NS is a straight line with a slope nearly 0; when the difference between P and Q is very large, NS is a curve with obvious fluctuations. The fluctuations can be represented by the curve information entropy NH:

[0083]

[0084] In the formula, n is the number of intervals where the curve is discretized; P i is to calculate the corresponding probability value after counting the number of points in each interval.

[0085] Construct the similarity measure SDE:

[0086] SDE = NH × |1 - ρ| + LD

[0087] In the formula, ρ is the spectral correlation coefficient; LD is the Lance distance. The Lance distance is a dimensionless standardized value that can effectively suppress the influence of noise.

[0088] There are n types of targets to be classified and recognized. The classification process using spectral information is as follows: First, determine the reference spectra of each category on the image; Second, calculate NH for each pixel and use it as the weight coefficient of SDE to construct SDE; Third, compare the SDE calculated for this pixel with the SDE distances of each category to determine the size, and classify this pixel into the SDE with the smallest distance difference to complete the classification.

[0089] Step A6: Combine the information of brightness, spectrum, polarization with the orbital information according to the energy weighting method to form a target parameter set.

[0090] For a target point, its observation data consists of 7 independent variables: (S0, S1, S2, S3, λ, x, y).

[0091] The fusion of polarization information includes two processes: S0, S1, S2, S3 are fused to obtain Y λ , DOP and AOP are fused to obtain P λ ; then Y λ and P λ are fused to obtain YP λ . The formulas used are as follows:

[0092]

[0093] Among them, F D，λ (n) 2 is the gray value of pixel n in the λ band, D ∈ {S0, S1, S2, S3}.

[0094]

[0095]

[0096] Among them, M′×N′ is a small area selected from the image DOP λ , and μ is the mean value of this area.

[0097]

[0098] After these steps, the fusion results in K bands can be obtained, and K feature images are obtained, constituting the fused data cube (YP, λ, x, y). The performance of the parameter set integrating polarization, spectrum, brightness and orbital information is better than that of the parameter set only containing brightness alone. Under the weak light detection of the starry sky background, the polarization information of the object can better feedback the information of its surface characteristics. Combined with the characteristics of the target composition material provided by the spectrum and the spatial information expressed by the orbit, the target can be recognized more accurately.

[0099] The target detection imaging system that combines polarization spectroscopy imaging with orbital information acquisition organically combines the brightness information, spectral information, polarization information, and orbital information of light. Among them, the orbital information lays the foundation for the tracking of the target and is also a parameter category for target recognition. The brightness information reflects the detection distance, target shape, and target size, etc.; the spectral information reflects the material composition and surface morphology of the space target, etc.; the polarization information reflects the material, roughness, and contrast with the background of the target. The combined application of the four-dimensional information of brightness, polarization, spectrum, and orbital information helps to improve the target detection probability.

[0100] Step A7: Determine whether there are any unclear or lacking parameters. If there are no lacking parameters, proceed to Step A8; if there are unclear or lacking parameters, first determine whether they can be obtained through recalculation. If they can, repeat Step A5; if they cannot be obtained through recalculation, output an error.

[0101] Step A8: Compare the target parameter set with the known data set. If known information is matched, confirm that the captured object is a certain pre-coded target, and output the matched target result. If no known information is matched, proceed to Step A9.

[0102] Step A9: Sort out the parameters, add a new number to the target, and add it to the database.

[0103] The above is only the preferred embodiment of the present invention and does not impose any limitation on the present invention. Any person skilled in the art, without departing from the technical solution of the present invention, makes any form of equivalent replacement or modification and other changes to the technical solution and technical content disclosed by the present invention, which are all within the content of the technical solution of the present invention and still fall within the protection scope of the present invention.

Claims

1. A space target detection imaging device under a complex starry sky background, comprising a signal transmitter and a received signal decoupling component, characterized in that, The signal transmitter is a laser light source, an achromatic wave plate, and a linear polarizer arranged coaxially in sequence. The received signal decoupling component includes a signal receiving control box and a computer. A reflector is arranged inside the box of the signal receiving control box. A first beam splitter, a collimating lens group, and a second beam splitter are arranged coaxially in sequence on the right side of the reflector. A first CMOS camera is arranged at a position corresponding to the first beam splitter. A third beam splitter is arranged at a position parallel to the second beam splitter. A color wheel mechanism is arranged at a corresponding position on the right side of the second beam splitter and the third beam splitter. A second CMOS camera and a third CMOS camera are respectively arranged at corresponding positions on the right side of the color wheel mechanism and the second beam splitter and the third beam splitter. The second CMOS camera and the third CMOS camera are controlled by a CMOS driving circuit, and the CMOS driving circuit is connected to the computer; multiple frames of images captured by the first CMOS camera are transmitted to the computer for target orbit calculation to obtain orbit information. The color wheel mechanism is an eight-channel color wheel mechanism. The color wheel mechanism is rotated, and eight-channel image data captured by the second CMOS camera and the third CMOS camera are transmitted to the computer; the eight-channel images are respectively polarized angles = 0°, 45°, 90°, and 135°, and the spectral bands are near ultraviolet <400nm, near infrared >780nm, visible light band 400 - 780nm, and unconditional.

2. The space target detection and imaging device under a complex starry sky background according to claim 1, wherein A rotating shaft is arranged at the center of the back of the color wheel mechanism, and the color wheel mechanism is installed on a controllable motor through the rotating shaft.

3. The space target detection imaging device under a complex starry sky background according to claim 2, characterized in that, The color wheel mechanism is connected to the CMOS driving circuit through a synchronization circuit, and the synchronization circuit controls the synchronization of the shooting of the second CMOS camera and the third CMOS camera and the rotation of the color wheel mechanism.

4. The space target detection imaging device under a complex starry sky background according to claim 1, characterized in that, The reflector, the first beam splitter, the second beam splitter, and the third beam splitter are all adjustably installed at the bottom of the box of the signal receiving control box through rotating shafts.

5. A method for identifying a space target of a space target detection imaging device under a complex starry sky background according to claim 1, characterized in that, It includes the following steps: Step A1: Actively emit laser through the signal transmitter to make linearly polarized light at a specified angle exit to the target; Step A2: Transmit multiple frames of images captured by the first CMOS camera to the computer for target orbit calculation to obtain orbit information; Step A3: Rotate the color wheel mechanism and transmit eight-channel image data captured by the second CMOS camera and the third CMOS camera to the computer; Step A4: Use a feature preprocessing classifier to select the clearest image; Step A5: Perform parameter decoupling to obtain the brightness, spectrum, and polarization parameters of the image; Step A6: According to the energy weighting method, combine the information of brightness, spectrum, and polarization with the orbit information to form a target parameter set; Step A7: Judge whether there are unclear or missing parameters. If there are no missing parameters, enter Step A8; if there are unclear or missing parameters, first judge whether they can be obtained through recalculation. If they can, repeat Step A5. If they cannot be obtained through recalculation, output an error; Step A8: Compare the target parameter set with the known data set. If known information is matched, it is confirmed that a certain pre-coded target is captured, and the matched target result is output. If no known information is matched, go to Step A9; Step A9: Sort out the parameters, add a new number to the target, and add it to the database.

6. The method for identifying a space target of the space target detection imaging device based on a complex starry sky background according to claim 5, characterized in that, In Step A3, the eight-channel images are respectively with polarization angles of 0°, 45°, 90° and 135°, and the spectral bands are near ultraviolet (<400 nm), near infrared (>780 nm), visible light band (400 - 780 nm) and unconditional.

7. The method for identifying a space target of the space target detection imaging device based on a complex starry sky background according to claim 5, characterized in that, In Step A5, the method for obtaining the spectral parameters includes the following steps: Step G1: Simultaneously obtain continuous, narrow-band spectral band images within the spectral coverage range of at least 0.4 - 2.5 μm; Step G2: Preprocess the images and perform relative reflectance image conversion; Step G3: Extract spectral features to obtain the l(x, y, λ) model.

8. The method for identifying a space target of the space target detection and imaging device based on a complex starry sky background according to claim 7, characterized in that, In step G2, the relative reflectance conversion adopts a conversion method based on a regression statistical equation: ρ b =∑F σ ·R e,b =aG i,b +C; ρ b is the relative reflectance; G i,b is the image gray value; R e,b is the spectral reflectance value at the e-wavelength point within the b-band; F σ is the weighting coefficient related to the sensor spectral response function, ∑F σ = 1; a is the slope; C is the intercept.

9. The method for identifying a space target of the space target detection and imaging device based on a complex starry sky background according to claim 5, wherein In Step A5, the method for obtaining the polarization parameters includes the following steps: Step P1: Extract polarization information and calculate Stokes vector parameters S0, S1, S2, S3; Step P2: Further calculate the degree of linear polarization, polarization azimuth angle, and polarization ellipticity.

Images