A single-pixel dynamic target computational imaging method based on corrected Radon spectrum

Abstract

The present invention discloses a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, including a structured light modulation device, an optical signal detection system, and an image processing system, and specific method steps include: generating a two-dimensional image pattern; grouping the two-dimensional image pattern, and sending it to the structured light modulation device in groups in sequence; using the two-dimensional image pattern to modulate the light source in the spatial domain, and irradiating the dynamic target to be imaged with the modulated light; using the optical signal detection system to detect the reflected light intensity signal of the dynamic target, and calculating the projection distribution curve of the dynamic target at each angle according to the detection information; using the curve to calculate the center of mass position of the dynamic target, and translating the curve, and arranging the angular size after translation to obtain the corrected Radon spectrum. The present invention corrects the Radon spectrum by combining single-pixel imaging and Radon transform technology and calculating the center of mass position information of the dynamic target, thereby avoiding image blur and improving the imaging capability of the system for dynamic targets.

Description

Technical Field

[0001] The present invention relates to the field of imaging technology, and in particular to a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum. Background Art

[0002] In recent years, single-pixel imaging technology has been widely studied as a new type of computational imaging technology based on corrected Radon spectrum to implement single-pixel dynamic target computational imaging. Since single-pixel detectors can detect a large spectral range and have high sensitivity, they have been applied and developed in imaging fields such as far-infrared, terahertz, and x-rays. However, since single-pixel imaging obtains target information through continuous sampling to reconstruct an image, the target must be regarded as stationary during this continuous sampling process; however, for fast dynamic targets, it is difficult to obtain sufficient sampling within the time when the target can be regarded as stationary, which will lead to blurred reconstruction results.

[0003] The shortcoming of the existing technology is that single-pixel imaging of static targets has been applied in many fields, but the application of dynamic target imaging still needs to be studied and developed. Radon spectrum is composed of one-dimensional projections of the target in different directions, and it has been widely used in fields such as computed tomography. The traditional method uses single-pixel detection technology to obtain the Radon spectrum and directly uses it to reconstruct the target image. However, for fast dynamic targets, this direct reconstruction method will make the reconstructed image blurred. There is currently no relevant research report on how to use a single-pixel imaging system to obtain the Radon spectrum of a dynamic target and reconstruct a clear image. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art. To achieve the above purpose, a single-pixel dynamic target computational imaging implementation method based on corrected Radon spectrum is adopted to solve the problems raised in the above background technology.

[0005] A method for implementing single-pixel dynamic target computational imaging based on corrected Radon spectrum includes an image processing system, a structured light modulation device, and an optical signal detection system, wherein the specific steps of the implementation method include:

[0006] Generate a two-dimensional image pattern using an image processing system;

[0007] Grouping the two-dimensional image patterns, and then sequentially inputting the groups into a structured light modulation device;

[0008] Using the two-dimensional image pattern to modulate the light source in the spatial domain, and irradiating the dynamic target to be imaged with the modulated light;

[0009] The reflected light intensity signal from the target is detected by the optical signal detection system, and the projection distribution curve of the dynamic target at different angles is calculated according to the detected response value and modulation information;

[0010] The projection distribution curve is used to calculate the centroid position of the dynamic target, the projection distribution curve is translated according to the centroid position, and the translated projection distribution curve is arranged according to the angle to obtain a corrected Radon spectrum;

[0011] The corrected Radon spectrum is back-projected to reconstruct the image of the dynamic target to be imaged.

[0012] As a further technical solution of the present invention: the specific steps of generating a two-dimensional image pattern using an image processing system include:

[0013] Generate a two-dimensional matrix using an image processing system, represent the two-dimensional matrix as a two-dimensional function, use the row and column directions of the two-dimensional matrix as the horizontal and vertical coordinate directions of the coordinate system where the two-dimensional function is located, and the value of the two-dimensional function corresponds one-to-one to the element value of the two-dimensional matrix;

[0014] Each row of data of the two-dimensional function is subjected to a back-projection transformation at different angles to obtain a series of two-dimensional projection matrices.

[0015] As a further technical solution of the present invention: the transformation formula of the back projection transformation is:

[0016] p θ (x,y;L,M)=t(L,M)δ(xcosθ+ysinθ-L),

[0017] where t M (L) is the representation of the two-dimensional function when the ordinate is M rows, and x and y represent the projection direction and t respectively. M (L) is the horizontal and vertical coordinates of the plane coordinate system, δ(xcosθ+ysinθ-L) is the transformation function, which means that the value on the straight line xcosθ+ysinθ-L=0 is 1, otherwise it is 0, and the angle θ satisfies the relationship 0°≤θ<180°.

[0018] As a further technical solution of the present invention: the specific steps of grouping the two-dimensional image patterns and then inputting the groups into the structured light modulation device in sequence include:

[0019] The two-dimensional image patterns are grouped, and a projection pattern in a back-projection direction and a pattern perpendicular to the back-projection direction are first extracted to form a group of projection patterns, until all patterns are extracted and grouped;

[0020] Finally, they are sent to the structured light modulation device in groups.

[0021] As a further technical solution of the present invention: the light source can be ordinary lighting light or laser.

[0022] As a further technical solution of the present invention: the specific steps of using the optical signal detection system to detect the reflected light intensity signal from the target and calculating the projection distribution curve of the dynamic target at different angles according to the detected response value and modulation information include:

[0023] The optical signal detection system is used to detect the reflected light intensity signal from the target, and the detection response value is I θ (M), the response value I θ (M) can be expressed as:

[0024] I θ (M) = ∑ L t(L,M)F θ (L);

[0025] where F θ (L) is the one-dimensional projection distribution curve of the dynamic target at angle θ, and t(L,M) represents a row of data of the two-dimensional function;

[0026] The one-dimensional projection distribution curve F of the dynamic target at different angles is calculated according to the response value and each row of data of the two-dimensional Hadamard matrix. θ (L), the one-dimensional projection distribution curve F θ (L) can be expressed as:

[0027]

[0028] As a further technical solution of the present invention: the specific steps of calculating the centroid position of the dynamic target using the projection distribution curve include:

[0029] The two projection curves F obtained by using a set of projection patterns with mutually perpendicular projection directions θ (L) and F θ+90° (L) Calculate the center of mass position of the dynamic target.

[0030] As a further technical solution of the present invention, the specific steps of translating the projection distribution curve according to the centroid position and arranging the translated projection distribution curve according to the angle to obtain the corrected Radon spectrum include:

[0031] A rectangular coordinate system is established along these two projection directions, where the points where L is zero correspond to the origins of the horizontal and vertical coordinates, respectively. The expression of the reference centroid position in the coordinate system is:

[0032]

[0033]

[0034] As a further technical solution of the present invention: the step of calculating the reference centroid position includes:

[0035] Calculating the centroid positions of other groups of projection curves according to the calculation method of the reference centroid position;

[0036] Different projection curves directly calculate the centroid coordinates in different coordinate systems, and the calculated reference centroid position coordinates are transformed into other coordinate systems.

[0037] As a further technical solution of the present invention: when the angle between the coordinate system where the reference centroid position is located and other coordinate systems is θ', the expression for transforming the reference centroid position coordinate table to other coordinate systems is:

[0038] x' c =x c *cos(θ')-y c *sin(θ'),

[0039] y' c =x c *sin(θ')+y c *cos(θ'),

[0040] Translate the projection curves of the other groups until the centroid position coordinates are equal to (x' c ,y' c );

[0041] The translated projection distribution curves are arranged according to the angle to obtain the corrected Radon spectrum of the dynamic target.

[0042] Compared with the prior art, the present invention has the following technical effects:

[0043] By adopting the above-mentioned technical scheme, a two-dimensional image pattern is generated, and then the dynamic target is modulated and illuminated by the device, and the light signal obtained by detecting the light information detection system is used to obtain the projection distribution curve of the Radon spectrum at different angles, and then the center of mass position of the dynamic target is calculated, and then the projection distribution curve is corrected using the center of mass position to obtain the corrected Radon spectrum and used for the reconstruction of the dynamic target. This method can reduce the image blur caused by motion and improve the imaging effect of the single-pixel imaging system on the dynamic target. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings:

[0045] Figure 1 A schematic diagram of an application system for a computational imaging implementation method of some embodiments disclosed in this application;

[0046] Figure 2A flowchart of the steps of the computational imaging implementation method of some embodiments disclosed in this application;

[0047] Figure 3 A flowchart of generating a two-dimensional image pattern for a computational imaging implementation method of some embodiments disclosed in the present application;

[0048] Figure 4 Schematic diagram of two-dimensional image patterns at projection angles of 0 degrees and 90 degrees for some embodiments disclosed in this application;

[0049] Figure 5 A flowchart of correcting Radon spectrum of a computational imaging implementation method disclosed in some embodiments of the present application;

[0050] Figure 6 Schematic diagram of object patterns of dynamic targets converted according to some embodiments disclosed in the present application.

[0051] In the figure: 110, structured light modulation device; 120, optical signal detection system; 130, image processing system. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] like Figure 1 As shown, the diagram is a schematic diagram of a system application for realizing a method for realizing single-pixel dynamic target computational imaging in an embodiment. The system includes a structured light modulation device 110 for modulating structured light, a light signal detection system 120, and an image processing system 130 for reconstructing an image.

[0054] In a specific embodiment, the structured light modulation device 110 uses the input two-dimensional pattern to modulate the light source in the spatial domain to generate spatial structured light and is used to illuminate the dynamic target. The structured light modulation device 110 may be composed of a light source, a beam expander, a digital micromirror array, and a projection lens. The light source in the structured light modulation device 110 is an LED white light lamp or a laser generating device. The optical signal detection system 120 is composed of a photodetector and a data acquisition card. The photodetector is used to detect the reflected light from the dynamic target, and then the photodetector converts the detected light signal into an electrical signal. The data acquisition card is responsible for collecting the electrical signal and sending it to the image processing system 130.

[0055] The image processing system 130 is used to control the structured light generating device, generate the modulation pattern and reconstruct the target image. The image processing system 130 can be any computing device with computing and control capabilities, such as a computer, a mobile phone, etc.

[0056] Please refer to Figure 2 In an embodiment of the present invention, a method for implementing single-pixel dynamic target computational imaging based on corrected Radon spectrum is provided, and the specific steps of the method include:

[0057] Step S201, generating a two-dimensional image pattern using an image processing system;

[0058] Step S202, grouping the two-dimensional image patterns, and then sequentially sending the groups to a device capable of modulating structured light;

[0059] Step S203, using the two-dimensional image pattern to modulate the light source in the spatial domain, and irradiating the dynamic target to be imaged with the modulated light;

[0060] Step S204, using the optical signal detection system to detect the reflected light intensity signal from the target, and calculating the projection distribution curve of the dynamic target at different angles according to the detected response value;

[0061] Step S205, using the projection distribution curve to calculate the centroid position of the dynamic target, translating the projection distribution curve according to the centroid position, and arranging the translated projection distribution curve according to the angle to obtain a corrected Radon spectrum;

[0062] Step S206: Perform a back-projection transformation on the corrected Radon spectrum to reconstruct an image of the dynamic target.

[0063] like Figure 3 As shown, it is a flow chart of a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum to generate a two-dimensional image pattern in an embodiment of the present invention. In the embodiment of the present invention, step S201 is to generate a two-dimensional image pattern by using an image processing system, which specifically includes:

[0064] Step S301, using an image processing system to generate a two-dimensional matrix, the matrix is ​​in Hadamard form or random form, the random form can be generated using software such as the random function in Matlab, the Hadamard matrix is ​​an n-order square matrix consisting of +1 and -1 elements and satisfies Hn*Hn'=nI (where Hn' is the transposed matrix of Hn, and I is the unit square matrix), and can be generated using the hadamard function in Matlab;

[0065] Step S302, expressing the two-dimensional matrix as a two-dimensional function, with the row and column directions of the two-dimensional matrix serving as the horizontal and vertical coordinate directions of the coordinate system where the two-dimensional function is located, and the values ​​of the two-dimensional function corresponding to the element values ​​of the two-dimensional matrix one by one;

[0066] Step S303, performing a back projection transformation on each row of data of the two-dimensional function at different angles to obtain a series of two-dimensional projection matrices, wherein the back projection transformation formula is:

[0067] p θ (x,y;L,M)=t(L,M)δ(xcosθ+ysinθ-L),

[0068] where t M (L) is the representation of the two-dimensional function when the ordinate is M rows, and x and y represent the projection direction and t respectively. M (L) is the horizontal and vertical coordinates of the plane coordinate system, δ(xcosθ+ysinθ-L) is the transformation function, which means that the value on the straight line xcosθ+ysinθ-L=0 is 1, otherwise it is 0, and the angle θ satisfies the relationship 0°≤θ<180°, that is, the angle division setting range is 0°≤θ<180°.

[0069] In the embodiment of the present invention, taking a two-dimensional Hadamard matrix of size 128×128 as an example, in each row of data t extracted M (L) consists of two values: -1 and +1. Since the structured light generator cannot generate negative values, the negative values ​​are set to zero instead.

[0070] Under the condition that the projection angle is 0°≤θ<180° and the interval is 3°, the row data of the two-dimensional Hadamard matrix is ​​back-projected and transformed. The transformation area is limited to a circular domain with a diameter of 128 pixels. Finally, 60*128 128*128 pixel two-dimensional image patterns are obtained, and each angle contains 128 transformation patterns. The two-dimensional image pattern can be transformed according to the above formula (1) as follows:

[0071]

[0072] Among them, x and y represent the projection direction and t respectively. M (L, M) is the horizontal and vertical coordinates of the plane where the two-dimensional Hadamard matrix is ​​located. M represents the serial number of the corresponding row of the two-dimensional Hadamard matrix. C(R) represents that the function is 1 when the diameter R is less than or equal to 128, otherwise it is 0. δ(xcosθ+ysinθ-L) is the transformation function, which represents that the value on the straight line xcosθ+ysinθ-L=0 is 1, otherwise it is 0. Each angle contains M two-dimensional image patterns. Figure 4As shown, the diagram shows examples of two-dimensional image patterns at projection angles of 0 degrees and 90 degrees provided by an embodiment of the present invention.

[0073] In the embodiment of the present invention, step S202, that is, grouping the two-dimensional image patterns, and then sequentially sending the groups to a device capable of modulating structured light, is specifically:

[0074] The two-dimensional image patterns are grouped. First, a projection pattern in a rear projection direction and a pattern perpendicular to the rear projection direction are extracted to form a group of projection patterns. All patterns are extracted and grouped accordingly, and then sent in groups to a device capable of modulating structured light.

[0075] In the embodiment of the present invention, taking the projection direction of 60 as an example, in the projection angle range of 0°≤θ<180°, under the condition that the angle interval corresponding to the 60 projection directions is 3°, 0° and 90° will be extracted to form a group of projection patterns, 3° and 93° will be extracted to form a group of projection patterns, and so on to 87° and 177°. Among all angles, the directions that must be greater than or equal to 90 degrees and the projection directions that are less than or equal to 90 degrees must be equal in number.

[0076] In the embodiment of the present invention, step S203 is to use the two-dimensional image pattern to perform spatial modulation on the light source, and use the modulated light to illuminate the dynamic target to be imaged. The light source can be ordinary illumination light or laser.

[0077] In the embodiment of the present invention, step S204, i.e., using the optical signal detection system to detect the reflected light intensity signal from the target, and calculating the projection distribution curve of the dynamic target at different angles according to the detected response value, specifically includes:

[0078] Step S401, receiving an optical signal to detect the response value I of the system θ The detected response value is obtained by converting the light intensity signal reflected by the optical signal detection system after receiving the modulated light irradiating the target. The response value I θ It can be expressed as:

[0079] I θ (M) = ∑ L t(L,M)F θ (L)........................................(3),

[0080] Among them, F θ (L) is the one-dimensional projection distribution curve of the target at angle θ.

[0081] Step S402: Calculate the one-dimensional projection distribution curve F of the dynamic target at different angles according to the response value and each row of data of the two-dimensional matrix.θ (L), one-dimensional projection distribution curve F θ (L) can be expressed as:

[0082]

[0083] In the embodiment of the present invention, taking the two-dimensional matrix as a Hadamard matrix as an example, the corresponding decomposed received response values ​​are respectively expressed as:

[0084]

[0085] Where f(x,y) is the expression of the target image in two-dimensional space, its value represents the pixel value of a single pixel of the target image, and x and y correspond to the position coordinates of each pixel respectively.

[0086] Further calculation of the response value I θ

[0087]

[0088] According to the target Radon transform properties:

[0089] F θ (L) = ∑ x,y f(x,y)δ(xcosθ+ysinθ-L)........................(7)

[0090] and then:

[0091]

[0092] The one-dimensional projection distribution curve F of the target at angle θ can be obtained θ (L).

[0093] like Figure 5 As shown, it is a flowchart of correcting Radon spectrum of a method for implementing single-pixel dynamic target computational imaging based on corrected Radon spectrum in an embodiment of the present invention. In the embodiment of the present invention, step S205 uses the projection distribution curve to calculate the centroid position of the dynamic target, translates the projection distribution curve according to the centroid position, and arranges the translated projection distribution curve according to the angle size to obtain the corrected Radon spectrum, which specifically includes:

[0094] Step S501, using a set of projection patterns with mutually perpendicular projection directions to obtain a projection distribution curve F θ (L) and F θ+90° (L) Calculate the centroid position and establish a rectangular coordinate system along the two projection directions. The expression of the centroid position in this coordinate system is:

[0095]

[0096]

[0097] Step S502, based on this, calculate the centroid position of other groups of projection curves. Different projection curves directly calculate the centroid coordinates in different coordinate systems. The calculated reference centroid position coordinates are transformed into other coordinate systems. When the angle between the coordinate system where the reference centroid position is located and the other coordinate system is θ', the expression for transforming the reference centroid position coordinate table into other coordinate systems is:

[0098] x' c =x c *cos(θ')-y c *sin(θ')........(10),

[0099] y' c =x c *sin(θ')+y c *cos(θ').......(11),

[0100] Step S503, translate the projection curves of other groups until the centroid position coordinates are equal to (x' c , y' c );

[0101] Step S504 arranges these translated projection distribution curves according to angles to obtain the corrected Radon spectrum F(L, θ) of the target.

[0102] In the embodiment of the present invention, step S206 is to perform a filtered back projection transformation on the corrected Radon spectrum to reconstruct the image of the dynamic target. In combination with the above embodiment provided by the present invention, the filtered back projection transformation can be implemented by the iradon() function in Matlab, and finally a two-dimensional matrix of the target is obtained, which can be transformed to obtain the pattern of the target, such as Figure 6 As shown, the object pattern of a dynamic target is shown.

[0103] Beneficial effects:

[0104] The present invention mainly proposes to obtain the Radon spectrum of the dynamic target using the single-pixel imaging technology in the light field based on the Radon mode, and correct the Radon spectrum, and finally use the corrected Radon spectrum to perform back-projection operation to obtain the image of the moving target. This solves the problem that the dynamic target image result is extremely blurred when the dynamic target is reconstructed because it is difficult to obtain sufficient sampling within the time when the fast-moving dynamic target is regarded as a stationary state.

[0105] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents, and all should be included within the scope of protection of the present invention.

Claims

1. A method for implementing single-pixel dynamic target computational imaging based on corrected Radon spectrum, which is implemented using the following components, including an image processing system, a structured light modulation device, and an optical signal detection system, characterized in that: The specific steps of the implementation method include: The two-dimensional image pattern is generated by using an image processing system, and the specific steps include: Generate a two-dimensional matrix using an image processing system, represent the two-dimensional matrix as a two-dimensional function, use the row and column directions of the two-dimensional matrix as the horizontal and vertical coordinate directions of the coordinate system where the two-dimensional function is located, and the value of the two-dimensional function corresponds one-to-one to the element value of the two-dimensional matrix; Performing a back-projection transformation on each row of data of the two-dimensional function at different angles to obtain a series of two-dimensional projection matrices; Grouping the two-dimensional image patterns, and then sequentially inputting the groups into a structured light modulation device; Using the two-dimensional image pattern to modulate the light source in the spatial domain, and irradiating the dynamic target to be imaged with the modulated light; The reflected light intensity signal from the target is detected by the optical signal detection system, and the projection distribution curve of the dynamic target at different angles is calculated according to the detected response value and modulation information; The projection distribution curve is used to calculate the centroid position of the dynamic target, the projection distribution curve is translated according to the centroid position, and the translated projection distribution curve is arranged according to the angle to obtain a corrected Radon spectrum; Performing back-projection transformation on the corrected Radon spectrum to reconstruct the image of the dynamic target to be imaged; The transformation formula of the back projection transformation is: p θ (x,y;L,M)=t(L,M)δ(xcosθ+ysinθ-L), Where t(L,M) is the representation of the two-dimensional function when the ordinate is M rows, x and y represent the projection direction and the horizontal and vertical coordinates of the plane coordinate system where t(L,M) is located, respectively, and δ(xcosθ+ysinθ-L) is a transformation function, which means that when the x and y values ​​satisfy the straight line xcosθ+y sinθ-L=0, the value of δ(x cosθ+y sinθ-L) is 1, otherwise the value of δ(x cosθ+y sinθ-L) is 0, and the angle θ satisfies the relationship .

2. According to claim 1, a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The specific steps of grouping the two-dimensional image patterns and then inputting the groups into the structured light modulation device in sequence include: The two-dimensional image patterns are grouped, and a projection pattern in a back-projection direction and a pattern perpendicular to the back-projection direction are first extracted to form a group of projection patterns, until all patterns are extracted and grouped; Finally, they are sent to the structured light modulation device in groups.

3. According to claim 1, a method for implementing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The light source may be ordinary illumination light or laser.

4. According to claim 1, a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The specific steps of using the optical signal detection system to detect the reflected light intensity signal from the target and calculating the projection distribution curve of the dynamic target at different angles according to the detected response value and modulation information include: The optical signal detection system is used to detect the reflected light intensity signal from the target, and the detection response value is I θ (M), the response value I θ (M) can be expressed as: I θ (M)=∑ L t(L,M)F θ (L); where F θ (L) is the one-dimensional projection distribution curve of the dynamic target at angle θ, and t(L,M) represents a row of data of the two-dimensional function; The one-dimensional projection distribution curve F of the dynamic target at different angles is calculated according to the response value and each row of data of the two-dimensional Hadamard matrix. θ (L), the one-dimensional projection distribution curve F θ (L) can be expressed as:

5. According to claim 1, a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The specific steps of calculating the center of mass position of the dynamic target using the projection distribution curve include: The two projection curves F obtained by using a set of projection patterns with mutually perpendicular projection directions θ (L) and F θ+90° (L) Calculate the center of mass position of the dynamic target.

6. According to claim 5, a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The specific steps of translating the projection distribution curve according to the centroid position and arranging the translated projection distribution curve according to the angle to obtain the corrected Radon spectrum include: A rectangular coordinate system is established along these two projection directions, where the points where L is zero correspond to the origins of the horizontal and vertical coordinates, respectively. The expression of the reference centroid position in the coordinate system is:

7. According to claim 6, a method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum, characterized in that: The step of calculating the reference centroid position comprises: Calculating the centroid positions of other groups of projection curves according to the calculation method of the reference centroid position; Different projection curves directly calculate the centroid coordinates in different coordinate systems, and the calculated reference centroid position coordinates are transformed into other coordinate systems.

8. The method for realizing single-pixel dynamic target computational imaging based on corrected Radon spectrum according to claim 7, characterized in that: When the angle between the coordinate system where the reference centroid position is located and other coordinate systems is θ', the expression for transforming the reference centroid position coordinate table to other coordinate systems is: x' c =x c *cos(θ')-y c *sin(θ'), and' c =x c *sin(θ')+y c *cos(θ'), Translate the projection curves of the other groups until the centroid position coordinates are equal to (x' c ,y' c ); The translated projection distribution curves are arranged according to the angle to obtain the corrected Radon spectrum of the dynamic target.

Images