A method for pairing orthogonal pixels in polarization spectrum imaging system

Abstract

The present invention discloses a method for pairing orthogonal pixels of an image of a polarization spectral imaging system, comprising: firstly performing field of view calibration, then performing spectral calibration data processing in a pixel set corresponding to the field of view calibration, and then performing resampling of the field of view calibration data to complete field of view and spectral calibration, and finally performing pairing of orthogonal pixels in the S light area and the P light area of ​​the image, and introducing an integrating sphere during spectral calibration to cover the entire field of view of the system. Aiming at the characteristics of dual imaging areas and the existence of single field of view spectral line bending, the present invention optimizes the calibration process and data processing method, performs spectral calibration data processing in a pixel set corresponding to each field of view angle obtained after field of view calibration, and performs field of view resampling, effectively eliminating the fitting error introduced by the inconsistency between the fitting direction of spectral calibration data and the actual spectral line direction of the traditional calibration method; and has the advantage of significantly improving the calibration efficiency.

Description

Technical Field

[0001] The invention relates to an image processing method for a polarization spectrum imaging system, in particular to a method for pairing orthogonal pixels in an image according to space and spectrum information. Background Art

[0002] A polarization spectral imaging system uses spectral modulation to obtain target space, spectrum, and polarization information at one time with strong real-time performance. The target polarization information is modulated on two spectra with mutually perpendicular polarization directions, which significantly reduces the complexity of data processing for this differential detection method compared to traditional interferometric spectral modulation, and there is no frequency domain aliasing problem. Before using the spectral imaging system, spectral calibration and field of view calibration are required to give clear physical meanings such as wavelength and field of view angle to the image pixel coordinates.

[0003] The traditional field of view and spectrum calibration methods of spectral imaging systems, such as the "Full Field of View Spectral Calibration Device for Push-Broom Imaging Spectrometer" with application number 201310310741.5, have the following problems:

[0004] 1) The data fitting of spectral calibration is performed in the row or column direction, which is not applicable to the curved spectral lines caused by the dispersion effect, resulting in fitting errors;

[0005] 2) It does not involve pairing of orthogonal pixels with mutually perpendicular polarization directions and consistent spectral information;

[0006] 3) Spectral calibration is performed by scanning different fields of view, which is inefficient. Summary of the invention

[0007] In order to solve the problems existing in the above-mentioned prior art, the present invention provides a method for pairing orthogonal pixels of a polarization spectral imaging system image, so as to improve the spectral calibration accuracy and the full coverage efficiency of the field of view for the dual imaging areas and curved spectral line phenomena, thereby completing the pairing of pixels in the S light area and the P light area, and making data preparations for the subsequent calculation of physical quantities.

[0008] The technical solution adopted by the present invention to solve the technical problem is:

[0009] The present invention discloses an orthogonal pixel pairing method for a polarization spectrum imaging system, wherein the polarization spectrum imaging system comprises: a polarization modulation module, a telescope module and a spectrum imaging module;

[0010] The polarization modulation module includes: an achromatic 1 / 4 wave plate, a multi-stage phase retarder, and a polarization beam splitter; the spectral imaging module includes: a slit grating dispersion component and a planar array detector component; the slit grating dispersion component includes: a slit, a grating, and a reflector group; its characteristic is that the image orthogonal pixel pairing method is performed according to the following steps:

[0011] Step 1: Field of view calibration;

[0012] Step 1.1: Use parallel light source at the mth lighting angle θ m After illuminating the polarization spectrum imaging system, the parallel light source light sequentially passes through the achromatic 1 / 4 wave plate, the multi-stage phase retarder, and the polarization beam splitter of the polarization adjustment module, and is received by the telescope module and imaged on the slit of the spectrum imaging module. After the light beam passes through the slit and is dispersed by the grating, it is received by the photosensitive surface of the area array detector assembly and forms a field of view calibration image; and the field of view calibration image contains areas where the modulated spectra are orthogonal to each other, which are respectively recorded as the field of view calibration S light area and the field of view calibration P light area;

[0013] Step 1.2: For the first row of data in the field of view calibration image, find the jth row of the field of view calibration S light area. S,θm (1) The peak coordinates of the column {1, j S,θm (1)} and the field of view calibration P light area in the 1st row j P,θm (1) The peak coordinates of the column {1, j P,θm (1)}; where j s,θm (1) represents the first row of field of view calibration data of the field of view calibration S light area at the mth illumination angle θ m The corresponding peak column number, j P,θm (1) represents the first row of field of view calibration data of the field of view calibration P light area at the mth illumination angle θ m The corresponding peak column number;

[0014] Step 1.3: According to the process of step 1.2, the remaining rows of the field of view calibration data in the field of view calibration image are processed to obtain the field of view calibration data of each row of the field of view calibration S light area at the mth illumination angle θ m The corresponding peak coordinate set And the field of view calibration data of each row of the field of view calibration P light area at the mth illumination angle θ m The corresponding peak coordinate set i represents the number of any row of data of the field of view calibration image, and I represents the total number of rows of the field of view calibration image;

[0015] Step 1.4: After changing the illumination angle, repeat the process of steps 1.1 to 1.3 to obtain the pixel coordinate set corresponding to the field calibration S light area of ​​the polarization spectrum imaging system at each standard illumination angle. And the pixel coordinate set corresponding to the field calibration P light area at each standard illumination angle M represents the total number of standard lighting angles;

[0016] Step 1.5: Based on the pixel coordinate set The peak column number set corresponding to the i-th row With the standard lighting angle set Θ std ={θ1,θ2,...,θ m ,...,θ M}, and obtain the field of view calibration equation θ′ of the field of view calibration S light area S,i =a i,S j′ i，S +b i,S , where j′ i,S ,θ′ S,i represents any peak column number in the i-th row of the field calibration S light area and its corresponding illumination angle, a i,S and b i,S Respectively represent the field of view calibration equation θ′ of the field of view calibration S light area S,i The slope and intercept of

[0017] According to the pixel coordinate set The peak column number set corresponding to the i-th row With the standard lighting angle set Θ std ={θ1,θ2,...,θ m ,...,θ M}, and obtain the field of view calibration equation θ′ of the field of view calibration P light area P,i =a i,P j′ i,P +b i,P , where j′ i,S ,θ′ S,i represents the column number of any peak value in the i-th row of the field calibration P light area and its corresponding illumination angle, a i,P and b i,P Respectively represent the field of view calibration equation θ′ of the field of view calibration P light area P,i The slope and intercept of the field of view calibration S light area corresponding to each row are obtained. SA ={θ′ S,1 ,θ′ S,2 ,...,θ′ S,i ,...,θ′ S,I} and the field of view calibration equation set F for the light area field of view calibration P PA ={θ′ P,1 ,θ′ P,2 ,...,θ′ P,i ,...,θ′ P,I};

[0018] Step 1.6: Set T sets of fields of view of interest represents the tth field of view of interest; and Substitute into F SA and F PA , thus obtaining The corresponding field of view calibration S light area coordinate set and the field of view calibration P light area pixel coordinate set Then, the pixel coordinate set corresponding to the field of view calibration S light area in each field of view angle set of interest is obtained. and the pixel coordinate set corresponding to the field calibration P light area in each of the field angle sets of interest in, Indicates the column number corresponding to the tth interesting field of view angle in the i-th row of the field of view calibration S light area, Indicates the column number corresponding to the tth interesting field of view angle in the i-th row of the field of view calibration P light area;

[0019] Step 2: Spectral calibration;

[0020] Step 2.1: Let the N standard wavelengths provided by the narrowband light source be denoted as Λ std ={λ std,1 ,λ std,2 ,...,λ std,n ,...,λ std,N}, where λ std,n is the nth standard wavelength; the narrow-band light source is introduced into the integrating sphere and illuminates the polarization spectrum imaging system to cover the full field of view of the polarization spectrum imaging system, and is received by the photosensitive surface of the planar array detector assembly (3.2) to form a spectrum calibration image, and the spectrum calibration image contains two regions with modulated spectra orthogonal to each other, which are respectively recorded as the spectrum calibration S light region and the spectrum calibration P light region;

[0021] Step 2.2: For the pixel coordinate set The i-th row and the t-th field of interest Coordinates Column number Round down to get the column number after downward processing Round up to get the column number after upward processing The rounded coordinates and The corresponding data in the spectral calibration image are recorded as and Coordinates obtained by linear interpolation The corresponding sub-pixel data in the spectral calibration image Thus, it is concluded The corresponding sub-pixel data set Then, the spectral calibration S light area is obtained in each row and each set of the field of view angles of interest. The corresponding pixel coordinate set The corresponding sub-pixel data set

[0022] Similarly, we get the pixel coordinate set The tth field of interest angle of the ith row of Coordinates The corresponding sub-pixel data in the spectral calibration image Thus, it is concluded The corresponding sub-pixel data set Then, the spectral calibration S light area is obtained in each row and each set of the field of view angles of interest. The corresponding pixel coordinate set The corresponding sub-pixel data set

[0023] Let coordinates express or The i-th row and the t-th field of interest in location.

[0024] Step 2.3: With {1, 2, …, i, …, I} as the horizontal coordinates, middle The corresponding column of data Perform curve fitting to obtain the row numbers corresponding to N data peaks in, Indicates the row number corresponding to the nth data peak at the tth field of view of interest in the spectral calibration S light area;

[0025] With Λ std is the horizontal axis, Perform linear fitting to obtain the S light area The corresponding spectral calibration equation in, express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent the S light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each field of view of interest in

[0026] Similarly, we get the P light area The corresponding spectral calibration equation in, express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent the P light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each of the fields of interest in

[0027] Step 2.4: Set the wavelength of interest represents the wavelength of interest tt; and Substitute into F SB and F PB , thus obtaining The corresponding spectral calibration S light area coordinate set And the spectral calibration P light area coordinate set Then, the pixel coordinate set corresponding to the spectral calibration S light area in each of the interested field angle sets is obtained. and the pixel coordinate set corresponding to each of the viewing angle sets of interest in the spectral calibration P light area in, Indicates the row number corresponding to the tth interesting field angle and the ttth interesting wavelength in the spectral calibration S light region, Indicates the row number corresponding to the tth interesting field angle and the ttth interesting wavelength in the spectral calibration P light area;

[0028] Step 3: Resampling of spatial dimension:

[0029] Step 3.1: The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation to obtain The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling:

[0030] It can be concluded that the S light area corresponds to The set of resampled coordinates of all spatial dimensions

[0031] by The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation and get The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling: So we get the P light area corresponding to The set of resampled coordinates of all spatial dimensions

[0032] Step 4: Pixel Pairing:

[0033] For any target observation image, respectively in ΨΨ S ,ΨΨ P Find out The corresponding pixel coordinates and And obtain the pixel coordinates in the arbitrary target observation image by two-dimensional interpolation Corresponding data and pixel coordinates Corresponding data Thus completing Corresponding orthogonal pixel pairing, thus completing Orthogonal pixel pairing.

[0034] The image orthogonal pixel pairing method of the polarization spectrum imaging system of the present invention is also characterized in that the two-dimensional interpolation method in step 4 is:

[0035] Step 4.1, coordinate The row and column numbers are rounded down and up respectively, and then combined to get four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates The corresponding data is recorded as The row and column numbers are rounded up to get the coordinates The corresponding data is recorded as The row number is rounded down and the column number is rounded up to get the coordinates The corresponding data is recorded as The row number is rounded up and the column number is rounded down to get the coordinates The corresponding data is recorded as

[0036] Step 4.2: Use formula (1) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value

[0037]

[0038] Step 4.2: Substitute the result of formula (1) into formula (2) and The column data is linearly interpolated in the row direction to obtain The corresponding

[0039]

[0040] Step 4.3, same as step 4.1, change the coordinates The row and column numbers are rounded down and up respectively, and then combined to find the four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates The corresponding data is recorded as The row and column numbers are rounded up to get the coordinates The corresponding data is recorded as The row number is rounded down and the column number is rounded up to get the coordinates The corresponding data is recorded as The row number is rounded up and the column number is rounded down to get the coordinates The corresponding data is recorded as

[0041] Step 4.4: Use formula (3) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value

[0042]

[0043] Step 4.5: Substitute the result of formula (3) into formula (4) and The column data is linearly interpolated in the row direction to obtain The corresponding

[0044]

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] Aiming at the characteristics that an image of a polarization spectral imaging system has an S light region and a P light region, i.e., a dual imaging region, and there is bending of a single field of view spectral line, the present invention optimizes the calibration process and the data processing method, processes the spectral calibration data in a pixel set corresponding to each field of view angle obtained after field of view calibration, and performs field of view resampling, thereby effectively eliminating the fitting error introduced by the inconsistency between the fitting direction of the spectral calibration data and the actual spectral line direction in the traditional calibration method; the spectral calibration adopts an integrating sphere to form an extended uniform light source, thereby directly covering the full field of view of a polarization spectral imaging system involved in the present invention, and the full field of view spectral calibration data can be obtained by a single observation, which has the advantage of significantly improving the calibration efficiency compared with the traditional spectral calibration method of performing full field of view scanning by a mechanical device. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a schematic diagram of a polarization spectrum imaging system of the present invention;

[0048] Figure 2 is a schematic diagram of the polarization spectrum imaging method of the present invention;

[0049] Figure 3 It is a schematic diagram of field of view calibration data fitting in step 1.2 of field of view calibration data processing of the present invention;

[0050] Figure 4 It is a schematic diagram of calibration data arrangement and preparation in step 2.2 of spectrum calibration data processing of the present invention;

[0051] Figure 5 is a schematic diagram of spectrum calibration data fitting in step 2.3 of spectrum calibration data processing of the present invention;

[0052] Figure 6 This is a schematic diagram of the effect of field resampling in step 3 of the present invention;

[0053] Figure 7 It is a schematic diagram of the two-dimensional interpolation method used in the pixel pairing of step 4 of the present invention.

[0054] The numbers in the figure are: 1. modulation module, 2. telescope module, 3. spectral imaging module, 1.1. achromatic 1 / 4 wave plate, 1.2. multi-stage phase retarder, 1.3. polarization beam splitter, 3.1. slit grating dispersion component, 3.2. array detector component, 3.1.1. slit, 3.1.2. grating, 3.1.3~3.1.5. reflector group, 3.2.1 and 3.2.2 are S light area and P light area. DETAILED DESCRIPTION

[0055] In this embodiment, Figure 1 As shown, the polarization spectrum imaging system includes: a polarization modulation module 1, a telescope module 2 and a spectrum imaging module 3;

[0056] The polarization modulation module 1 includes: an achromatic 1 / 4 wave plate 1.1, a multi-stage phase retarder 1.2, and a polarization beam splitter 1.3; the spectral imaging module 3 includes: a slit grating dispersion component 3.1, and a planar array detector component 3.2; the slit grating dispersion component 3.1 includes: a slit 3.1.1, a grating 3.1.2, and a reflector group.

[0057] like Figure 2 As shown, the same incident angle in two areas of the target image received by the array detector assembly 3.2 θ , same wavelength λ The corresponding pixels are calculated one by one, and the spectrum, polarization and one-dimensional spatial information of the target can be obtained simultaneously to form a one-dimensional information array. Each information unit contains the spectrum and polarization information of the corresponding spatial unit (i.e., the incident angle θ). Finally, the polarization spectrum information of the two-dimensional space is obtained through push scanning.

[0058] In a specific implementation, the purpose of orthogonal pixel pairing of polarization spectral imaging system images is to clarify the pixel coordinates of the S light area and the P light area corresponding to each specified wavelength point and field of view, and then pair the orthogonal pixels of the S light area and the P light area corresponding to the same wavelength and field of view, so as to prepare data for subsequent analytical calculations. A method for orthogonal pixel pairing of polarization spectral imaging system images is performed in the following steps:

[0059] Step 1: Field of view calibration;

[0060] Step 1.1: Use parallel light source at the mth lighting angle θ m After illuminating the polarization spectrum imaging system, the parallel light source light passes through the achromatic 1 / 4 wave plate 1.1, the multi-stage phase retarder 1.2, and the polarization beam splitter 1.3 of the polarization adjustment module 1 in sequence, and is received by the telescope module 2 and imaged on the slit 3.1.1 of the spectrum imaging module 3. After the light beam passes through the slit 3.1.1 and is dispersed by the grating 3.1.2, it is received by the photosensitive surface of the array detector assembly 3.2 and forms a field of view calibration image; and the field of view calibration image contains areas where the modulated spectra are orthogonal to each other, which are respectively recorded as the field of view calibration S light area and the field of view calibration P light area;

[0061] Step 1.2: For the first row of data in the field of view calibration image, find the jth row of the field of view calibration S light area. S,θm (1) The peak coordinates of the column {1, j S,θm (1)} and the field of view calibration P light area in the first row j P,θm (1) The peak coordinates of the column {1, j P,θm (1)}; where j s,θm (1) represents the first row of field of view calibration data of the field of view calibration S light area at the mth illumination angle θ mThe corresponding peak column number, j P,θm (1) represents the first row of field calibration data of the field calibration P light area at the mth illumination angle θ m The corresponding peak column number;

[0062] Step 1.3: According to the process of step 1.2, the remaining rows of field of view calibration data in the field of view calibration image are processed to obtain the field of view calibration data of each row in the field of view calibration S light area at the mth illumination angle θ m The corresponding peak coordinate set And the field of view calibration data of each row of the field of view calibration P light area at the mth illumination angle θ m The corresponding peak coordinate set i represents the number of any row of data in the field of view calibration image, and I represents the total number of rows in the field of view calibration image. Figure 3 As shown;

[0063] Step 1.4: After changing the illumination angle, repeat the process of steps 1.1 to 1.3 to obtain the pixel coordinate set corresponding to the field calibration S light area of ​​the polarization spectrum imaging system at each standard illumination angle. And the pixel coordinate set corresponding to the field calibration P light area at each standard lighting angle M represents the total number of standard lighting angles;

[0064] Step 1.5: Based on the pixel coordinate set The peak column number set corresponding to the i-th row With the standard lighting angle set Θ std ={θ1,θ2,...,θ m ,...,θ M}, and obtain the field of view calibration equation θ′ for the field of view calibration S light area S,i =a i,S j′ i,S +b i,S , where j′ i,S ,θ′ S,i represents the column number of any peak value in the i-th row of the field calibration S light area and its corresponding illumination angle, a i,S and b i,S They represent the field of view calibration equation θ′ for the field of view calibration S light area. S,i The slope and intercept of

[0065] According to the pixel coordinate set The peak column number set corresponding to the i-th row With the standard lighting angle set Θ std ={θ1,θ2,...,θ m ,...,θ M}, and get the field of view calibration equation θ′ for the field of view calibration P light area P,i =a i,P j′ i,P +b i,P , where j′ i,S ,θ′ S,i Indicates the column number of any peak value in the i-th row of the field calibration P light area and its corresponding illumination angle, a i,P and b i,P They represent the field of view calibration equation θ′ for the field of view calibration P light area respectively. P,i The slope and intercept of the field of view calibration S light area corresponding to each row are obtained. SA ={θ′ S,1 ,θ′ S,2 ,...,θ′ S,i ,...,θ′ S,I} and the field of view calibration equation set F for the light area field of view calibration P PA ={θ′ P,1 ,θ′ P,2 ,...,θ′ P,i ,...,θ′ P,I};

[0066] Step 1.6: Set T sets of fields of view of interest represents the tth field of view of interest; and Substitute into F SA and F PA , thus obtaining The corresponding field of view calibration S light area coordinate set And the field of view calibration P light area pixel coordinate set Then, the pixel coordinate set corresponding to the field of view calibration S light area in each field of view angle set of interest is obtained. The pixel coordinate set corresponding to the field of view calibration P light area at each field of view angle set of interest in, Indicates the column number corresponding to the tth interesting field of view angle in the i-th row of the field of view calibration S light area, Indicates the column number corresponding to the tth field of interest in the i-th row of the field of view calibration P light area.

[0067] Step 2: Spectral calibration;

[0068] Step 2.1: Let the N standard wavelengths provided by the narrowband light source be denoted as Λ std ={λ std,1 ,λ std,2 ,...,λ std,n ,...,λ std,N}, where λstd,n is the nth standard wavelength; a narrow-band light source is introduced into the integrating sphere and illuminates the polarization spectrum imaging system to cover the full field of view of the polarization spectrum imaging system, and is received by the photosensitive surface of the planar array detector assembly (3.2) to form a spectrum calibration image, and the spectrum calibration image contains two regions with modulated spectra orthogonal to each other, which are respectively recorded as the spectrum calibration S light region and the spectrum calibration P light region;

[0069] Step 2.2: For the pixel coordinate set The i-th row and the t-th field of interest Coordinates Column number Round down to get the column number after downward processing Round up to get the column number after upward processing The rounded coordinates and The corresponding data in the spectral calibration image are recorded as and Coordinates obtained by linear interpolation The corresponding sub-pixel data in the spectral calibration image Thus, it is concluded The corresponding sub-pixel data set Then we can get the spectrum calibration S light area in each row and each field of view of interest. The corresponding pixel coordinate set The corresponding sub-pixel data set

[0070] Similarly, we get the pixel coordinate set The tth field of interest angle of the ith row of Coordinates The corresponding sub-pixel data in the spectral calibration image Thus, it is concluded The corresponding sub-pixel data set Then we can get the spectrum calibration S light area in each row and each field of view of interest. The corresponding pixel coordinate set The corresponding sub-pixel data set

[0071] Let coordinates express or The i-th row and the t-th field of interest in location, such as Figure 4 Step 2.3: With {1, 2, …, i, …, I} as the horizontal coordinates, middle The corresponding column of data Perform curve fitting, such as Figure 5 As shown, the row numbers corresponding to N data peaks are obtained. in, Indicates the row number corresponding to the nth data peak at the tth field of view of interest in the spectral calibration S light area.

[0072] With Λ std is the horizontal axis, Perform linear fitting to obtain the S light area The corresponding spectral calibration equation in, express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent S light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each field of view of interest in

[0073] Similarly, we get the P light area The corresponding spectral calibration equation in, express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent P light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each field of view of interest in

[0074] Step 2.4: Set the wavelength of interest represents the wavelength of interest tt; and Substitute into F SB and F PB , thus obtaining The corresponding spectral calibration S light area coordinate set And the spectral calibration P light area coordinate set Then, the pixel coordinate set corresponding to the spectral calibration S light area in each field of view angle set of interest is obtained. And the pixel coordinate set corresponding to the spectral calibration P light area at each field of view angle set of interest in, Indicates the row number corresponding to the tth interesting field angle and the ttth interesting wavelength in the spectral calibration S light region, Indicates the row number corresponding to the tth field of view of interest and the ttth wavelength of interest in the spectral calibration P light area.

[0075] Step 3: Resampling of spatial dimension:

[0076] Step 3.1: The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation to obtain The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling:

[0077] It can be concluded that the S light area corresponds to The set of resampled coordinates of all spatial dimensions like Figure 6 As shown;

[0078] by The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation to obtain The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling: So we get the P light area corresponding to The set of resampled coordinates of all spatial dimensions

[0079] Step 4: Pixel Pairing:

[0080] For any target observation image, respectively in ΨΨ S ,ΨΨ P Find out The corresponding pixel coordinates and And obtain the pixel coordinates of any target observation image through two-dimensional interpolation Corresponding data and pixel coordinates Corresponding data Thus completing Corresponding orthogonal pixel pairing, thus completing Orthogonal pixel pairing.

[0081] In specific implementation, the two-dimensional interpolation method is:

[0082] Step 4.1, coordinate The row and column numbers are rounded down and up respectively, and then combined to get four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates The corresponding data is recorded as The row and column numbers are rounded up to get the coordinates The corresponding data is recorded as The row number is rounded down and the column number is rounded up to get the coordinates The corresponding data is recorded as The row number is rounded up and the column number is rounded down to get the coordinates The corresponding data is recorded as like Figure 7 As shown;

[0083] Step 4.2: Use formula (1) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value

[0084]

[0085] Step 4.2: Substitute the result of formula (1) into formula (2) and The column data is linearly interpolated in the row direction to obtain The corresponding

[0086]

[0087] Step 4.3, same as step 4.1, change the coordinates The row and column numbers are rounded down and up respectively, and then combined to find the four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates The corresponding data is recorded as The row and column numbers are rounded up to get the coordinates The corresponding data is recorded as The row number is rounded down and the column number is rounded up to get the coordinates The corresponding data is recorded as The row number is rounded up and the column number is rounded down to get the coordinates The corresponding data is recorded as

[0088] Step 4.4: Use formula (3) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value

[0089]

[0090] Step 4.5: Substitute the result of formula (3) into formula (4) and The column data is linearly interpolated in the row direction to obtain The corresponding

[0091]

Claims

1. A method for pairing orthogonal pixels of a polarization spectral imaging system, the polarization spectral imaging system comprising: A polarization modulation module (1), a telescope module (2) and a spectral imaging module (3); The polarization modulation module (1) comprises: an achromatic quarter wave plate (1.1), a multi-stage phase retarder (1.2), and a polarization beam splitter (1.3); the spectral imaging module (3) comprises: a slit grating dispersion component (3.1), and a planar array detector component (3.2); the slit grating dispersion component (3.1) comprises: a slit (3.1.1), a grating (3.1.2), and a reflector group; and the image orthogonal pixel pairing method is performed according to the following steps: Step 1: Field of view calibration; Step 1.1: Use parallel light source at the mth lighting angle After illuminating the polarization spectrum imaging system, the parallel light source light sequentially passes through the achromatic quarter wave plate (1.1), the multi-stage phase retarder (1.2), and the polarization beam splitter (1.3) of the polarization modulation module (1), is received by the telescope module (2), and is imaged on the slit (3.1.1) of the spectrum imaging module (3); the light beam passing through the slit (3.1.1) is dispersed by the grating (3.1.2), and is received by the photosensitive surface of the planar array detector assembly (3.2) to form a field of view calibration image; and the field of view calibration image contains regions where the modulation spectra are orthogonal to each other, which are respectively recorded as the field of view calibration S light region and the field of view calibration P light region; Step 1.2: For the first row of data in the field of view calibration image, find the jth row of the field of view calibration S light area. S,θm (1) The peak coordinates of the column {1, j S,θm (1)} and the field of view calibration P light area in the first row j P,θm (1) The peak coordinates of the column {1, j P,θm (1)}; where j s,θm (1) represents the first row of field of view calibration data of the field of view calibration S light area at the mth illumination angle The corresponding peak column number, j P,θm (1) represents the first row of field of view calibration data of the field of view calibration P light area at the mth illumination angle The corresponding peak column number; Step 1.3: According to the process of step 1.2, the remaining rows of the field of view calibration data in the field of view calibration image are processed to obtain the field of view calibration data of each row of the field of view calibration S light area at the mth illumination angle. The corresponding peak coordinate set , and the field of view calibration data of each row of the field of view calibration P light area at the mth illumination angle The corresponding peak coordinate set ; i represents the number of any row of data of the field of view calibration image, and I represents the total number of rows of the field of view calibration image; Step 1.4: After changing the illumination angle, repeat the process of steps 1.1 to 1.3 to obtain the pixel coordinate set corresponding to the field calibration S light area of ​​the polarization spectrum imaging system at each standard illumination angle. , and the pixel coordinate set corresponding to the field calibration P light area at each standard illumination angle , M represents the total number of standard lighting angles; Step 1.5: Based on the pixel coordinate set The peak column number set corresponding to the i-th row With standard lighting angle set , the field of view calibration equation of the field of view calibration S light area is obtained ,in, , represents any peak column number in the i-th row of the field calibration S light area and its corresponding illumination angle, and Respectively represent the field of view calibration equation of the field of view calibration S light area The slope and intercept of According to the pixel coordinate set The peak column number set corresponding to the i-th row With standard lighting angle set , get the field of view calibration equation of the field of view calibration P light area ,in, , represents the column number of any peak value in the i-th row of the field calibration P light area and its corresponding illumination angle, and Respectively represent the field of view calibration equation of the field of view calibration P light area The slope and intercept of the field of view calibration S light area corresponding to each row are obtained. And the field of view calibration P light area field of view calibration equation set ; Step 1.6: Set T sets of fields of view of interest , represents the tth field of view of interest; and Substitute and , thus obtaining The corresponding field of view calibration S light area coordinate set and the field of view calibration P light area pixel coordinate set ; Then obtain the pixel coordinate set corresponding to the field of view calibration S light area in each of the field of view angle sets of interest and the pixel coordinate set corresponding to the field calibration P light area in each of the field angle sets of interest ;in, Indicates the column number corresponding to the tth interesting field of view angle in the i-th row of the field of view calibration S light area, Indicates the column number corresponding to the tth interesting field of view angle in the i-th row of the field of view calibration P light area; Step 2: Spectral calibration; Step 2.1: Let the N standard wavelengths provided by the narrowband light source be ,in, is the nth standard wavelength; the narrow-band light source is introduced into the integrating sphere and illuminates the polarization spectrum imaging system to cover the full field of view of the polarization spectrum imaging system, and is received by the photosensitive surface of the planar array detector assembly (3.2) to form a spectrum calibration image, and the spectrum calibration image contains two regions with modulated spectra orthogonal to each other, which are respectively recorded as the spectrum calibration S light region and the spectrum calibration P light region; Step 2.2: For the pixel coordinate set The i-th row and the t-th field of interest Coordinates , column number Round down to get the column number after downward processing , Round up to get the column number after upward processing , the rounded coordinates and The corresponding data in the spectral calibration image are recorded as and , the coordinates are obtained by linear interpolation The corresponding sub-pixel data in the spectral calibration image ; thus deriving The corresponding sub-pixel data set , and then obtain the spectral calibration S light area in each row and each of the interested field angle sets The corresponding pixel coordinate set The corresponding sub-pixel data set ; Similarly, we get the pixel coordinate set The tth field of interest angle of the ith row of Coordinates The corresponding sub-pixel data in the spectral calibration image , thus deriving The corresponding sub-pixel data set , and then obtain the spectral calibration P light area in each row and each of the interested field angle sets The corresponding pixel coordinate set The corresponding sub-pixel data set ; Let coordinates express or The i-th row and the t-th field of interest in location; Step 2.3: With {1, 2, …, i, …, I} as the horizontal coordinates, middle The corresponding column of data Perform curve fitting to obtain the row numbers corresponding to N data peaks ;in, Indicates the row number corresponding to the nth data peak at the tth field of view of interest in the spectral calibration S light area; by is the horizontal axis, Perform linear fitting to obtain the S light area The corresponding spectral calibration equation ,in, , express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent the S light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each field of view of interest in ; Similarly, we get the P light area The corresponding spectral calibration equation ,in, , express middle Any row number of a column of data and its corresponding wavelength, and Respectively represent the P light area The corresponding spectral calibration equation The slope and intercept of The set of spectral calibration equations corresponding to each of the fields of interest in ; Step 2.4: Set the wavelength of interest , represents the wavelength of interest tt; and Substitute and , thus obtaining The corresponding spectral calibration S light area coordinate set And the spectral calibration P light area coordinate set , and then obtain the pixel coordinate set corresponding to the spectral calibration S light area in each of the interested field angle sets and the pixel coordinate set corresponding to each of the viewing angle sets of interest in the spectral calibration P light area ;in, Indicates the row number corresponding to the tth interesting field angle and the ttth interesting wavelength in the spectral calibration S light region, Indicates the row number corresponding to the tth interesting field angle and the ttth interesting wavelength in the spectral calibration P light area; Step 3: Resampling of spatial dimension: Step 3.1: The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve ;by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation to obtain The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling: ; It can be concluded that the S light area corresponds to The set of resampled coordinates of all spatial dimensions ;by The row number is the horizontal axis and the column number is the vertical axis. Perform curve fitting to obtain the fitting curve ;by The coordinate line number in As a benchmark, the fitting curve Perform spline interpolation and get The corresponding field of view calibration S light area coordinate set is the coordinate set after spatial dimension resampling: ; Thus, the P light area corresponds to The set of resampled coordinates of all spatial dimensions ; Step 4: Pixel Pairing: For any target observation image, , Find out , The corresponding pixel coordinates and , and obtain the pixel coordinates of the arbitrary target observation image through two-dimensional interpolation Corresponding data and pixel coordinates Corresponding data ; thus completing , Corresponding orthogonal pixel pairing, thus completing , Orthogonal pixel pairing.

2. The method for orthogonal pixel pairing of polarization spectrum imaging system according to claim 1, characterized in that: The two-dimensional interpolation method in step 4 is: Step 4.1, coordinate The row and column numbers are rounded down and up respectively, and then combined to get four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates , the corresponding data is recorded as ; The row and column numbers are rounded up to get the coordinates , the corresponding data is recorded as ; Round down the row number and round up the column number to get the coordinates , the corresponding data is recorded as ; Round up the row number and round down the column number to get the coordinates , the corresponding data is recorded as ; Step 4.2: Use formula (1) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value : (1) Step 4.2: Substitute the result of formula (1) into formula (2) and The column data is linearly interpolated in the row direction to obtain The corresponding : (2) Step 4.3, same as step 4.1, change the coordinates The row and column numbers are rounded down and up respectively, and then combined to find the four enclosing coordinates Image pixels: row and column numbers are rounded down to get coordinates , the corresponding data is recorded as ; The row and column numbers are rounded up to get the coordinates , the corresponding data is recorded as ; Round down the row number and round up the column number to get the coordinates , the corresponding data is recorded as ; Round up the row number and round down the column number to get the coordinates , the corresponding data is recorded as ; Step 4.4: Use formula (3) to Line and The row data is linearly interpolated in the column direction to obtain The corresponding value and The corresponding value : (3) Step 4.5: Substitute the result of formula (3) into formula (4) and The column data is linearly interpolated in the row direction to obtain The corresponding : (4)。

Images