CN102410844B - Method and device for correcting non-uniformity of image of high-dynamic star sensor - Google Patents

Abstract

The invention discloses a non-uniform correction method for a high dynamic star sensor image. The method includes: after determining that the star map to be corrected is different from the gain and/or integration time of two calibration images, according to the imaging model of ICCD, the Perform gain condition transformation and gain response nonuniformity correction on the two calibration images, and/or integration time condition transformation and integration time response nonuniformity correction; use the corrected calibration image to perform nonuniformity correction on the star map to be corrected Uniformity correction. The invention also discloses a device for correcting image non-uniformity of a high-dynamic star sensor. By adopting the method and device of the invention, the non-uniformity of the star image can be effectively suppressed, thereby effectively improving the measurement accuracy of the star sensor.

Description

A high dynamic star sensor image non-uniform correction method and device

technical field

The invention relates to spacecraft attitude measurement technology, in particular to a method and device for correcting image non-uniformity of a high-dynamic star sensor.

Background technique

The star sensor is a high-precision, high-reliability attitude measurement component widely used in today's aerospace vehicles. The operation of high-dynamic carriers puts forward very high requirements on the dynamic performance of star sensors. Therefore, how to improve the dynamic performance of star sensors has become a new research hotspot.

Under high dynamic conditions, the dispersion of star point energy will cause the detection sensitivity of the star sensor to drop sharply, and the number of observed stars in the field of view will be significantly reduced, and the normal attitude output cannot be guaranteed. Therefore, in order to improve the detection sensitivity of the star sensor, there have been low-light detection devices such as enhanced charge-coupled devices (ICCD), electron multiplier charge-coupled devices (EMCCD), and electron bombardment charge-coupled devices (EBCCD) as image sensors. High dynamic star sensor. Due to the differences in pixel multiplication factor, quantum efficiency, and effective photosensitive area of various image sensors, the non-uniformity of ICCD, EMCCD, and EBCCD has a greater impact on imaging quality. As a high-precision attitude measurement component, the star sensor needs to achieve the sub-pixel calculation accuracy of the star point centroid, and the non-uniformity of ICCD, EMCCD, EBCCD and other low-light detection devices will have a greater impact on the star point centroid calculation accuracy. , thus affecting the accuracy of star sensor attitude measurement. Therefore, the non-uniform correction of the star sensor image is the key link to improve the reliability and measurement accuracy of the high dynamic star sensor.

Non-uniformity correction methods are mainly divided into two types: correction methods based on reference radiation sources and correction methods based on scene statistics. Among them, the calibration method based on the reference radiation source is relatively mature. At present, the reference radiation source calibration methods used in engineering practice mainly include: the two-point temperature calibration method proposed by R.W.Helfrich et al. and the multi-point temperature calibration method proposed by A.F.Miltion et al. Law. Among them, the advantage of the two-point temperature calibration method is that the algorithm is simple, but its dynamic range is narrow, and there is a large correction error when the output response of low-light detection devices such as ICCD, EMCCD, and EBCCD is seriously nonlinear; The temperature calibration method has high correction accuracy, but with the increase of calibration points, the calculation amount of correction increases obviously. In order to make up for the above shortcomings, an improved two-point temperature calibration method based on the nonlinear response model of the detector is proposed. This method has a wide dynamic range and a relatively ideal calibration accuracy, and the algorithm is simple.

However, in the practical application of high dynamic star sensors, it is necessary to adjust the two parameters of low-light detection device gain and integration time to control the imaging quality. The gain and integration time conditions of the calibration image are different, and the existing non-uniformity correction methods do not compensate for the different shooting conditions of the calibration image and the image to be corrected. The measurement accuracy of the sensor.

Contents of the invention

In view of this, the main purpose of the present invention is to provide a high dynamic star sensor image non-uniformity correction method and device, which can effectively suppress the non-uniformity of the star image, thereby improving the measurement accuracy of the star sensor.

In order to achieve the above object, technical solution of the present invention is achieved in that way:

The invention provides a method for correcting image non-uniformity of a high dynamic star sensor, the method comprising:

After determining that the star map to be corrected is different from the gain and/or integration time of the two calibration images, according to the ICCD imaging model, the gain condition transformation and gain response non-uniformity correction are performed on the two calibration images, and/or Or integration time transformation and integration time response non-uniformity correction;

Using the corrected calibration image to perform non-uniformity correction on the star map to be corrected;

The imaging model of the ICCD is: s=f(Φ) exp(c g+d) (a t+b); wherein, S represents the ICCD response, and Φ represents the irradiance received by the star sensor , t represents the exposure time of ICCD, g represents the gain of ICCD, and a, b, c, d are constants.

In the above scheme, after it is determined that the star map to be corrected is different from the gains of the two calibration images, the gain condition transformation and gain response non-uniformity correction are performed on the two calibration images, as follows:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ 或者， $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j);$ 其中，表示增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_n的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据，表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据， ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_n}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_n}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right];$ 其中，M、N表示所述增益响应非均匀校正定标图像的像素数，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度值。Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g1+d)+d^{\′}(i,j)]}{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\&Center\; Dot;S_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein, S _Φ (i, j) represents the gray value of the (i, j)th pixel of each calibration image, The gain response non-uniformity correction coefficient includes: c'(i, j) and d'(i, j), c'(i, j) represents the correction gain of the gain of the (i, j)th pixel, d' (i, j) represents the correction offset of the gain of the (i, j)th pixel, g1 represents the gain of the star map to be corrected, g2 represents the gain of the calibration image,

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j),$ or, $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\&Center\; Dot;Y_{g_{\mathrm{no}}}(i,j);$ in, Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with gain g _m ,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with a gain of g _n ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is _gm , Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n , ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_{\mathrm{no}}}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right];$ Wherein, M and N represent the number of pixels of the gain response non-uniform correction calibration image,

Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with gain g _m ,

Indicates the gray value of the (i, j)th pixel of the gain-response non-uniform correction calibration image with gain g _n .

In the above scheme, when it is determined that the integration time of the star map to be corrected is different from that of the two calibration images, the conversion of the integration time conditions and the non-uniformity correction of the integration time response are performed on the two calibration images, as follows:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

其中，表示积分时间为t_e的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，表示积分时间为t_f的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

其中，M、N表示所述积分响应非均匀校正定标图像的像素数。Scanning each image obtained by correcting the gain response non-uniformity of the calibration image line by line from top to bottom, and then according to the formula: ${S^p}_{\mathrm{\Φ}}(i,j)=\frac{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\·(a\·t2+b)+b^{\′}(i,j)}\·S_{\mathrm{\Φ}}(i,j),$ Carry out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; wherein, S _Φ (i, j) represents the gray level of the (i, j)th pixel of each calibration image value, the integration time response non-uniformity correction coefficient includes: a'(i, j) and b'(i, j), a'(i, j) represents the correction of the integration time of the (i, j)th pixel Gain, b'(i, j) represents the correction offset of the integration time of the (i, j)th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the calibration image, $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j),$ or,

in, Indicates the mean value of the gray level of the non-uniform correction calibration image with the integration time t _e in response to the integration time,

Represents the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f ,

Indicates the gray value of the (i, j)th pixel of the integrated time response non-uniform correction calibration image whose integration time is t _e , Indicates the gray value of the (i, j)th pixel of the integration time response non-uniform correction calibration image with integration time _tf ,

Wherein, M and N represent the number of pixels of the integral response non-uniform correction calibration image.

In the above scheme, when it is determined that the star map to be corrected is different from the gain and integration time of the two calibration images, the gain condition transformation and gain response non-uniformity correction, integration time condition transformation and During integral time response non-uniformity correction, the described two calibration images are carried out gain condition transformation and gain response non-uniformity correction, as:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ 或者， $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j);$ 其中，表示增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_n的增益响应非均匀校正定标图像的灰度的线性化处理的均值，表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据， ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_n}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_n}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right];$ 其中，M、N表示所述增益响应非均匀校正定标图像的像素数，表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度值；Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g1+d)+d^{\′}(i,j)]}{\exp [c^{\′}(i,j)\·(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\·S_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein, S _Φ (i, j) represents the gray value of the (i, j)th pixel of each calibration image, The gain response non-uniformity correction coefficient includes: c'(i, j) and d'(i, j), c'(i, j) represents the correction gain of the gain of the (i, j)th pixel, d' (i, j) represents the correction offset of the gain of the (i, j)th pixel, g1 represents the gain of the star map to be corrected, g2 represents the gain of the calibration image,

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j),$ or, $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\&Center\; Dot;Y_{g_{\mathrm{no}}}(i,j);$ in, Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with gain g _m ,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with a gain of g _n , Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is _gm ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n , ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_{\mathrm{no}}}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right];$ Wherein, M and N represent the number of pixels of the gain response non-uniform correction calibration image, Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with gain g _m ,

Represents the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n ;

Correspondingly, the transformation of the integration time condition and the non-uniformity correction of the integration time response for the two calibration images are:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

其中，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值，表示积分时间为t_e的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，表示积分时间为t_f的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值， ${\stackrel{\‾}{S}}_{t_f}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j).$ Scanning each image after performing gain condition transformation and gain response non-uniformity correction on the calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′\′}(i,j)=\frac{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\·(a\&Center\; Dot;t2+b)+b^{\′}(i,j)}\·S_{\mathrm{\Φ}}^{\′}(i,j),$ Carrying out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; wherein, S′ _Φ (i, j) means that each of the calibration images is subjected to gain condition transformation and gain response non-uniformity The gray value of the (i, j)th pixel of the image after linearity correction, the integral time response non-uniformity correction coefficient includes: a'(i, j) and b'(i, j), a'(i , j) represents the correction gain of the integration time of the (i, j)th pixel, b'(i, j) represents the correction bias of the integration time of the (i, j)th pixel, t1 represents the integration time of the star map to be corrected , t2 represents the integration time of the calibration image, $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j),$ or,

in,

Indicates the mean value of the gray level of the non-uniform correction calibration image with the integration time t _e in response to the integration time,

Represents the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f in response to the integration time, Indicates the gray value of the (i, j)th pixel of the integrated time response non-uniform correction calibration image whose integration time is t _e , Indicates the gray value of the (i, j)th pixel of the integration time response non-uniform correction calibration image with integration time _tf , ${\stackrel{\‾}{S}}_{t_f}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j).$

In the above scheme, after it is determined that the star map to be corrected is different from the gain and integration time of the two calibration images, the two calibration images are sequentially transformed from the integration time condition and the integration time response non-uniformity correction, gain condition During transformation and gain response non-uniformity correction, the described two calibration images are carried out integration time conditional transformation and integration time response non-uniformity correction, as:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

其中，表示积分时间为t_e的积分时间响应非均匀校正定标图像的灰度的均值，表示积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

${\stackrel{\‾}{S}}_{t_f}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j);$ Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: ${S^p}_{\mathrm{\Φ}}(i,j)=\frac{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\&Center\; Dot;(a\·t2+b)+b^{\′}(i,j)}\&Center\; Dot;S_{\mathrm{\Φ}}(i,j),$ Carry out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; wherein, S _Φ (i, j) represents the gray level of the (i, j)th pixel of each calibration image Value, the integration time response non-uniformity correction coefficient includes: a'(i, j) and b'(i, j), a'(i, j) represents the correction of the integration time of the (i, j)th pixel Gain, b'(i, j) represents the correction offset of the integration time of the (i, j)th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the calibration image, $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\&Center\; Dot;S_{t_e}(i,j),$ or,

in, Indicates the mean value of the gray level of the non-uniform correction calibration image with the integration time t _e in response to the integration time, Represents the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f ,

Indicates the gray value of the (i, j)th pixel of the integrated time response non-uniform correction calibration image whose integration time is t _e ,

Indicates the gray value of the (i, j)th pixel of the integration time response non-uniform correction calibration image with integration time _tf ,

${\stackrel{\‾}{S}}_{t_f}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j);$

Correspondingly, performing gain condition transformation and gain response non-uniformity correction on the two calibration images is:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ 或者， $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j);$ 其中，

表示增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_n的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据， ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_n}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_n}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right].$ Scan each image after performing integration time conditional transformation and integration time response non-uniformity correction on the calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\·g1+d)+d^{\′}(i,j)]}{{\exp [c}^{\′}(i,j)\&Center\; Dot;(c\·g2+d)+d^{\′}(i,j)]}\&Center\; Dot;{S^p}_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein, S ^p _Φ (i, j) represents that each of the calibration images performs integration time transformation and integration time response non-uniformity The gray value of the (i, j)th pixel of the corrected image, the gain response non-uniformity correction coefficients include: c'(i, j) and d'(i, j), c'(i, j ) represents the correction gain of the gain of the (i, j) pixel, d'(i, j) represents the correction bias of the gain of the (i, j) pixel, g1 represents the gain of the star map to be corrected, and g2 represents the The gain of the scaled image,

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j),$ or, $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\·Y_{g_{\mathrm{no}}}(i,j);$ in,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with gain g _m ,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with a gain of g _n ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is _gm ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n , ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_{\mathrm{no}}}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right].$

In the above scheme, the acquired gain response non-uniform correction calibration image is:

Using the integrating sphere as a uniform light source, two images output by the ICCD under the conditions of gain g _m and g _n are obtained as gain response non-uniform correction calibration images

In the above scheme, the acquisition of the integrated time response non-uniform correction calibration image is:

Using the integrating sphere as a uniform light source, the ICCD output images under the conditions of integration time t _e and t _f are acquired as the integration time response non-uniform correction calibration image.

In the above-mentioned scheme, the method further includes:

According to the photoelectric response model of the imaging device, the imaging model of ICCD is established.

The present invention also provides a high dynamic star sensor image non-uniformity correction device, which includes: a gain and integration time correction module and a non-uniformity correction module; wherein,

Gain and integration time correction module, after determining that the star map to be corrected is different from the gain and/or integration time of the two calibration images, according to the imaging model of ICCD, the gain condition conversion and gain response are performed on the two calibration images Non-uniformity correction, and/or integration time condition transformation and integration time response non-uniformity correction, and send the corrected image to the non-uniformity correction module;

The non-uniformity correction module is used to correct the non-uniformity of the star map to be corrected by using the corrected calibration image after receiving the corrected image sent by the gain and integration time correction module;

The imaging model of the ICCD is: S=f(Φ) exp(c g+d) (a t+b); wherein, S represents the ICCD response, and Φ represents the irradiance received by the star sensor , t represents the exposure time of ICCD, g represents the gain of ICCD, and a, b, c, d are constants.

In the above solution, the device further includes: a setting module, which is used for the photoelectric response model of the imaging device, and establishes the imaging model of the ICCD.

In the high dynamic star sensor image non-uniformity correction method and device provided by the present invention, after determining that the star map to be corrected is different from the gain and/or integration time of the two calibration images, according to the imaging model of ICCD, the two Perform gain condition transformation and gain response non-uniformity correction on the calibration image, and/or integration time condition transformation and integration time response non-uniformity correction; use the corrected calibration image to perform non-uniformity correction on the star map to be corrected , so that the non-uniformity of the star map can be effectively suppressed, thereby effectively improving the measurement accuracy of the star sensor.

In addition, the high dynamic star sensor image non-uniformity correction method and device provided by the present invention are suitable for any gain and any integration time condition of the star map to be corrected, therefore, it can also effectively improve the gain in the non-uniformity correction of the star map And integration time condition adaptability.

Description of drawings

Fig. 1 is a schematic flow chart of a method for correcting image non-uniformity of a high dynamic star sensor in the present invention;

Fig. 2 is a schematic flow chart of a method for correcting image non-uniformity of a high dynamic star sensor in Embodiment 1;

Fig. 3 is a schematic flow chart of a method for correcting image non-uniformity of a high dynamic star sensor in Embodiment 2;

FIG. 4 is a schematic flow diagram of a method for correcting gain response non-uniformity of two calibration images in Embodiment 2;

FIG. 5 is a schematic flow diagram of a method for performing integration time response non-uniformity correction on two calibration images in Embodiment 2;

Fig. 6 is a schematic flow chart of the method for performing non-uniformity correction on the star map to be corrected by using the corrected calibration image in the second embodiment;

Fig. 7 is a schematic structural diagram of a device for correcting image non-uniformity of a high dynamic star sensor according to the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

The method for correcting the image non-uniformity of the high dynamic star sensor of the present invention, as shown in Figure 1, comprises the following steps:

Step 101: After determining that the star map to be corrected is different from the gain and/or integration time of the two calibration images, according to the ICCD imaging model, perform gain condition transformation and gain response non-uniformity correction on the two calibration images , and/or integration time condition transformation and integration time response non-uniformity correction;

Specifically, when it is determined that the star map to be corrected is different from the gain of the two calibration images, according to the ICCD imaging model, the gain condition transformation and gain response non-uniformity correction are performed on the two calibration images; After the star map and the integration time of the two calibration images are different, according to the imaging model of ICCD, the integration time condition transformation and the integration time response non-uniformity correction are carried out to the two calibration images; when it is determined that the star map to be corrected is different from the two After the gain and integration time of the two calibration images are different, according to the imaging model of ICCD, the gain condition transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction are performed on the two calibration images. .

Before this step, the method may further include:

According to the photoelectric response model of imaging device, set up the imaging model of ICCD; The imaging model of described ICCD is: S=f(Φ) exp(c g+d) (a t+b); Wherein, S Represents ICCD response, Φ represents the irradiance received by the star sensor, t represents the exposure time of ICCD, g represents the gain of ICCD, and a, b, c, d are constants. In actual application, when establishing the imaging model of ICCD, in addition to relying on the photoelectric response model of the imaging device, it also needs to be combined with actual experiments; among them, the specific process of establishing ICCD is a conventional technical means for those skilled in the art.

After it is determined that the star map to be corrected is different from the gains of the two calibration images, the gain condition transformation and gain response non-uniformity correction are performed on the two calibration images, specifically:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ 或者， $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j);$ 其中，表示增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，表示增益为g_n的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据， ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_n}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_n}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right];$ 其中，M、N表示所述增益响应非均匀校正定标图像的像素数，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度值；举个例子来说，假设所述增益响应非均匀校正定标图像的像素数为1024×1024，则M＝1024，N＝1024。采用公式： $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j)$ 计算出的d′(i，j)与采用公式： $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j)$ 计算出的d′(i，j)相同。Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\·g1+d)+d^{\′}(i,j)]}{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\&Center\; Dot;S_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein, S _Φ (i, j) represents the gray value of the (i, j)th pixel of each calibration image, The gain response non-uniformity correction coefficient includes: c'(i, j) and d'(i, j), c'(i, j) represents the correction gain of the gain of the (i, j)th pixel, d' (i, j) represents the correction offset of the gain of the (i, j)th pixel, g1 represents the gain of the star map to be corrected, g2 represents the gain of the calibration image,

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ or, $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\·Y_{g_{\mathrm{no}}}(i,j);$ in, Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with gain g _m , Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with a gain of g _n ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is _gm ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n , ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_{\mathrm{no}}}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right];$ Wherein, M and N represent the number of pixels of the gain response non-uniform correction calibration image,

Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with gain g _m ,

Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with a gain of g _n ; for example, assume that the number of pixels of the gain response non-uniform correction calibration image is 1024× 1024, then M=1024, N=1024. Using the formula: $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j)$ The calculated d'(i, j) and the adopted formula: $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\·Y_{g_{\mathrm{no}}}(i,j)$ The calculated d'(i, j) is the same.

Here, the calibration image refers to the calibration image used when performing non-uniform corrective correction on the star map to be corrected.

Wherein, the acquisition of the gain response non-uniform correction calibration image is specifically:

Using the integrating sphere as a uniform light source, two images output by the ICCD under the conditions of gain g _m and g _n are obtained as gain response non-uniform correction calibration images; here, m and n can be selected arbitrarily.

According to the imaging model of ICCD, the gray level of each pixel of each gain response non-uniform correction calibration image is linearized, specifically:

scanning each of the gain response non-uniform correction calibration images line by line from top to bottom;

Taking the logarithm of the gray scale of each pixel of each of the gain response non-uniform correction calibration images;

Wherein, after taking the logarithm of the gray scale of each pixel of each of the gain response non-uniform correction calibration images, it indicates that the linearization process for each of the gain response non-uniform correction calibration images is completed; After counting, for the gain response non-uniform correction calibration image with gain g _m , then we have For the gain response non-uniform correction calibration image with gain g _n , then we have $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right].$

When it is determined that the star map to be corrected is different from the integration time of the two calibration images, the integration time condition transformation and integration time response non-uniformity correction are performed on the two calibration images, specifically:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

其中，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

其中，M、N表示所述积分响应非均匀校正定标图像的像素数；这里，采用公式：

计算出的b′(i，j)与采用公式： $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_f}-a^{\′}(i,j)\·S_{t_f}(i,j)$ 计算出的b′(i，j)相同。Scanning each image obtained by correcting the gain response non-uniformity of the calibration image line by line from top to bottom, and then according to the formula: ${S^p}_{\mathrm{\Φ}}(i,j)=\frac{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t2+b)+b^{\′}(i,j)}\&Center\; Dot;S_{\mathrm{\Φ}}(i,j),$ Carry out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; wherein, S _Φ (i, j) represents the gray level of the (i, j)th pixel of each calibration image Value, the integration time response non-uniformity correction coefficient includes: a'(i, j) and b'(i, j), a'(i, j) represents the correction of the integration time of the (i, j)th pixel Gain, b'(i, j) represents the correction offset of the integration time of the (i, j)th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the calibration image, $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j),$ or,

in,

Indicates the mean value of the gray level of the non-uniform correction calibration image with the integration time t _e in response to the integration time,

Represents the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f ,

Indicates the gray value of the (i, j)th pixel of the integrated time response non-uniform correction calibration image whose integration time is t _e ,

Indicates the gray value of the (i, j)th pixel of the integration time response non-uniform correction calibration image with integration time _tf ,

Wherein, M and N represent the number of pixels of the integral response non-uniform correction calibration image; here, the formula is adopted:

The calculated b'(i, j) and the adopted formula: $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_f}-a^{\′}(i,j)\&Center\; Dot;S_{t_f}(i,j)$ The calculated b'(i, j) is the same.

The acquisition of the integrated time response non-uniform correction calibration image is specifically:

Using the integrating sphere as a uniform light source, the ICCD output image under the condition of integration time t _e and t _f is obtained as the integration time response non-uniform correction calibration image; here, e and f can be selected arbitrarily.

When it is determined that the star map to be corrected is different from the gain and integration time of the two calibration images, the gain time transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction are performed on the two calibration images in sequence. During uniformity correction, the gain condition transformation and gain response non-uniformity correction are performed on the two calibration images, specifically:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j),$ 或者， $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j);$ 其中，

表示增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，

表示增益为g_n的增益响应非均匀校正定标图像的灰度的线性化处理的均值，表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度的线性化处理后的数据， ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_n}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_n}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right];$ 其中，M、N表示所述增益响应非均匀校正定标图像的像素数，

表示增益为g_m的增益响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示增益为g_n的增益响应非均匀校正定标图像的第(i，j)像素的灰度值；举个例子来说，假设所述增益响应非均匀校正定标图像的像素数为1024×1024，则M＝1024，N＝1024。采用公式： $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\·Y_{g_m}(i,j)$ 计算出的d′(i，j)与采用公式： $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_n}-c^{\′}(i,j)\·Y_{g_n}(i,j)$ 计算出的d′(i，j)相同。Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g1+d)+d^{\′}(i,j)]}{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\·S_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein, S _Φ (i, j) represents the gray value of the (i, j)th pixel of each calibration image, The gain response non-uniformity correction coefficient includes: c'(i, j) and d'(i, j), c'(i, j) represents the correction gain of the gain of the (i, j)th pixel, d' (i, j) represents the correction offset of the gain of the (i, j)th pixel, g1 represents the gain of the star map to be corrected, g2 represents the gain of the calibration image,

$d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j),$ or, $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\&Center\; Dot;Y_{g_{\mathrm{no}}}(i,j);$ in,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with gain g _m ,

Represents the mean value of the linearization process of the gray scale of the gain response non-uniform correction calibration image with a gain of g _n , Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is _gm ,

Represents the data after the linearization processing of the gray level of the (i, j)th pixel of the gain response non-uniform correction calibration image whose gain is g _n , ${\stackrel{\‾}{Y}}_{g_m}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_m}(i,j),$ ${\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{g_{\mathrm{no}}}(i,j),$ $Y_{g_m}(i,j)=\ln \left[S_{g_m}(i,j)\right],$ $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right];$ Wherein, M and N represent the number of pixels of the gain response non-uniform correction calibration image,

Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with gain g _m ,

Indicates the gray value of the (i, j)th pixel of the gain response non-uniform correction calibration image with a gain of g _n ; for example, assume that the number of pixels of the gain response non-uniform correction calibration image is 1024× 1024, then M=1024, N=1024. Using the formula: $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j)$ The calculated d'(i, j) and the adopted formula: $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}-c^{\′}(i,j)\·Y_{g_{\mathrm{no}}}(i,j)$ The calculated d'(i, j) is the same.

Here, the calibration image refers to the calibration image used when performing non-uniform corrective correction on the star map to be corrected.

Correspondingly, performing integration time condition transformation and integration time response non-uniformity correction on the two calibration images is specifically:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

其中

表示积分时间为t_e的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值，

表示积分时间为t_e的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

表示积分时间为t_f的积分时间响应非均匀校正定标图像的第(i，j)像素的灰度值，

${\stackrel{\‾}{S}}_{t_f}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j);$ 这里，采用公式： $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j)$ 计算出的b′(i，j)与采用公式：

计算出的b′(i，j)相同。Scanning each image after performing gain condition transformation and gain response non-uniformity correction on the calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′\′}(i,j)=\frac{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t2+b)+b^{\′}(i,j)}\&Center\; Dot;S_{\mathrm{\Φ}}^{\′}(i,j),$ Carrying out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; wherein, S′ _Φ (i, j) means that each of the calibration images is subjected to gain condition transformation and gain response non-uniformity The gray value of the (i, j)th pixel of the image after linearity correction, the integral time response non-uniformity correction coefficient includes: a'(i, j) and b'(i, j), a'(i , j) represents the correction gain of the integration time of the (i, j)th pixel, b'(i, j) represents the correction bias of the integration time of the (i, j)th pixel, t1 represents the integration time of the star map to be corrected , t2 represents the integration time of the calibration image, $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j),$ or,

in

Indicates the mean value of the gray level of the non-uniform correction calibration image with the integration time t _e in response to the integration time,

Represents the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f ,

Indicates the gray value of the (i, j)th pixel of the integrated time response non-uniform correction calibration image whose integration time is t _e ,

Indicates the gray value of the (i, j)th pixel of the integration time response non-uniform correction calibration image with integration time _tf ,

${\stackrel{\‾}{S}}_{t_f}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{S}_{t_f}(i,j);$ Here, the formula is used: $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\·S_{t_e}(i,j)$ The calculated b'(i, j) and the adopted formula:

The calculated b'(i, j) is the same.

Wherein, the acquisition of the gain response non-uniform correction calibration image is specifically:

Using the integrating sphere as a uniform light source, two images output by the ICCD under the conditions of gain g _m and g _n are obtained as gain response non-uniform correction calibration images; here, m and n can be selected arbitrarily.

According to the imaging model of ICCD, the gray level of each pixel of each gain response non-uniform correction calibration image is linearized, specifically:

scanning each of the gain response non-uniform correction calibration images line by line from top to bottom;

Taking the logarithm of the gray scale of each pixel of each of the gain response non-uniform correction calibration images;

对于增益为g_n的增益响应非均匀校正定标图像，则有 $Y_{g_n}(i,j)=\ln \left[S_{g_n}(i,j)\right].$ Wherein, after taking the logarithm of the gray scale of each pixel of each of the gain response non-uniform correction calibration images, it indicates that the linearization process for each of the gain response non-uniform correction calibration images is completed; After counting, for the gain response non-uniform correction calibration image with gain g _m , then we have

For the gain response non-uniform correction calibration image with gain g _n , then we have $Y_{g_{\mathrm{no}}}(i,j)=\ln \left[S_{g_{\mathrm{no}}}(i,j)\right].$

The acquisition of the integrated time response non-uniform correction calibration image is specifically:

Using the integrating sphere as a uniform light source, the ICCD output image under the condition of integration time t _e and t _f is obtained as the integration time response non-uniform correction calibration image; here, e and f can be selected arbitrarily.

When it is determined that the star map to be corrected and the gain and integration time of the two calibration images are different, and the two calibration images are sequentially transformed from the integration time condition and the non-uniformity correction of the integration time response, the gain condition transformation and the gain response non-uniformity During linearity correction, the described two calibration images are carried out integration time condition transformation and integration time response non-uniformity correction, specifically:

Obtain an integrated time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, according to the imaging model of ICCD, calculating the mean value of the gray level of each said integral time response non-uniform correction calibration image;

Scanning the integral time response non-uniform correction calibration image line by line from top to bottom, using the mean value of the gray levels of all the integral time response non-uniform correction calibration images to calculate the integral time response non-uniformity correction coefficient;

Each of the calibration images is scanned line by line from top to bottom, and then according to the formula: ${S^p}_{\mathrm{\Φ}}(i,j)=\frac{a^{\′}(i,j)\·(a\·t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\&Center\; Dot;(a\&Center\; Dot;t2+b)+b^{\′}(i,j)}\&Center\; Dot;S_{\mathrm{\Φ}}(i,j),$ Carry out integration time condition transformation and integration time response non-uniformity correction to the two calibration images; here, the meaning of each parameter in the right side of the formula is the same as performing gain condition transformation and gain response non-uniformity on the two calibration images in turn In the case of linearity correction and integration time response non-uniformity correction, the meanings of the parameters in the formulas used for the integration time conditional transformation and integration time response non-uniformity correction of the two calibration images are the same, and will not be repeated here.

Correspondingly, performing gain condition transformation and gain response non-uniformity correction on the two calibration images is specifically:

Obtain a calibration image for gain response non-uniformity correction;

According to the imaging model of ICCD, linearize the gray level of each pixel of each of the gain response non-uniform correction calibration images, and use the linearized data to calculate each of the gain response non-uniform correction The mean value of the linearization process of the calibration image;

Scan the linearized image from top to bottom line by line, and calculate the gain response non-uniformity correction coefficient by using the mean value of the linearized processing of all the gain response non-uniform correction calibration images;

Scan each image after performing integration time conditional transformation and integration time response non-uniformity correction on the calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g1+d)+d^{\′}(i,j)]}{{\exp [c}^{\′}(i,j)\·(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\&Center\; Dot;{S^p}_{\mathrm{\Φ}}(i,j),$ Perform gain conditional transformation and gain response non-uniformity correction on the two calibration images; wherein, S ^p _Φ (i, j) represents that each of the calibration images performs integration time conditional transformation and integration time response non-uniformity The gray value of the (i, j)th pixel of the image after linear correction, here, the meaning of each parameter except S ^p _Φ (i, j) on the right side of the formula is the same as the gain condition transformation of the two calibration images in turn and gain response uniformity correction, integration time condition transformation and integration time response non-uniformity correction, the meanings of each parameter in the formula used for the integration time condition transformation and integral gain non-uniformity correction of the two calibration images are the same, I won't go into details here.

Here, it needs to be explained that in the actual application process, when performing gain condition transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction on two calibration images, the gain condition Transform and gain response non-uniformity correction, and then perform integration time condition transformation and integration time response non-uniformity correction on the image after gain condition transformation and gain response non-uniformity correction, or, you can first perform integration time condition transformation and integration Time response non-uniformity correction, and then perform gain response non-uniformity correction on the image after integration time condition transformation and integration time response non-uniformity correction.

Step 102: Using the corrected calibration image to perform non-uniformity correction on the star map to be corrected;

Specifically, each corrected calibration image is scanned line by line from top to bottom;

performing linearization processing on the grayscale of each pixel of each corrected calibration image, and using the data after linearization processing to calculate the mean value of the linearization processing of each calibration image after correction;

Scan the linearized image from top to bottom line by line, and calculate the star map non-uniformity correction coefficient by using the mean value of the linearized processing of all corrected calibration images;

Scanning the star map to be corrected line by line from top to bottom, and performing linearization processing on the gray scale of each pixel of the star map to be corrected;

$s(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-r(i,j)\·G_{{\mathrm{\Φ}}_u}(i,j),$ 或者， $s(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}-r(i,j)\·G_{{\mathrm{\Φ}}_v}(i,j),$ S_线性(i，j)＝ln[255/S_星图(i，j)-1]；其中，表示校正后的两个定标图像中的一个定标图像的线性化处理的均值，

表示校正后的两个定标图像中的另一个定标图像的线性化处理的均值，

表示校正后的两个定标图像中的一个定标图像的第(i，j)像素的灰度的线性化处理后的数据，

表示校正后的两个定标图像中的另一个定标图像的第(i，j)像素的灰度的线性化处理后的数据，S_星图(i，j)表示所述待校正星图的第(i，j)像素的灰度值， ${\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}=\frac{1}{M\×N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{G}_{{\mathrm{\Φ}}_v}(i,j),$ 其中， $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_v}^{\′}(i,j)-1],$ 或者， $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/{S^p}_{{\mathrm{\Φ}}_u}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/{S^p}_{{\mathrm{\Φ}}_v}(i,j)-1],$ 或者， $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′\′}(i,j)-1],$

其中，

表示进行增益条件变换及增益响应非均匀性校正后的两个定标图像中的一个定标图像的第(i，j)像素的灰度值，

表示进行增益条件变换及增益响应非均匀性校正后的两个定标图像中的另一个定标图像的第(i，j)像素的灰度值，

表示进行积分时间条件变换及积分时间响应非均匀性校正后的两个定标图像中的一个定标图像的第(i，j)像素的灰度值，表示进行积分时间条件变换及积分时间响应非均匀性校正后的两个定标图像中的一个定标图像的第(i，j)像素的灰度值，表示进行增益条件变换及增益响应非均匀性校正、积分时间条件变换及积分时间响应非均匀性校正后的两个定标图像中的一个定标图像的第(i，j)像素的灰度值，

表示进行增益条件变换及增益响应非均匀性校正、积分时间响应非均匀性校正后的两个定标图像中的另一个定标图像的第(i，j)像素的灰度值；采用公式： $s(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-r(i,j)\·G_{{\mathrm{\Φ}}_u}(i,j)$ 计算出的s(i，j)与采用公式： $s(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}-r(i,j)\·G_{{\mathrm{\Φ}}_v}(i,j)$ 计算出的s(i，j)相同。这里，需要说明的是：步骤101涉及的所有定标图像的像素与所述待校正星图的像素完全相同，换句话说，M、N表示所有定标图像及待校正图像的像素数。Then according to the formula: S _correction (i, j) = 255/{exp[r (i, j) S _linear (i, j) + s (i, j)] + 1}, the star map to be corrected Perform non-uniformity correction for each pixel; wherein, the star map non-uniformity correction coefficient includes: r(i, j) and s(i, j), r(i, j) represents the star to be corrected The correction gain of the (i, j)th pixel in the map, s(i, j) represents the correction bias of the (i, j)th pixel of the star map to be corrected, S _linear (i, j) represents the Correct the linearized data of the gray level of the (i, j)th pixel of the star map,

$\mathrm{the\; s}(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-r(i,j)\·G_{{\mathrm{\Φ}}_u}(i,j),$ or, $\mathrm{the\; s}(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}-r(i,j)\·G_{{\mathrm{\Φ}}_v}(i,j),$ S _linear (i, j) = ln[255/S _{star map} (i, j)-1]; where, Represents the mean value of the linearization process of one of the two calibration images after correction,

Represents the linearized mean of the other of the two corrected calibration images,

Represents the linearized data of the gray level of the (i, j)th pixel of one of the two calibration images after correction,

Represents the linearized data of the gray level of the (i, j)th pixel of the other calibration image of the two calibration images after correction, and S _{star map} (i, j) represents the star map to be corrected The gray value of the (i, j)th pixel, ${\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{G}_{{\mathrm{\Φ}}_v}(i,j),$ in, $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_v}^{\′}(i,j)-1],$ or, $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/{S^p}_{{\mathrm{\Φ}}_u}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/{S^p}_{{\mathrm{\Φ}}_v}(i,j)-1],$ or, $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′\′}(i,j)-1],$

in,

Indicates the gray value of the (i, j)th pixel of one of the two calibration images after gain condition transformation and gain response non-uniformity correction,

Indicates the gray value of the (i, j)th pixel of the other calibration image of the two calibration images after gain condition transformation and gain response non-uniformity correction,

Indicates the gray value of the (i, j)th pixel of one of the two calibration images after integration time condition transformation and integration time response non-uniformity correction, Indicates the gray value of the (i, j)th pixel of one of the two calibration images after integration time condition transformation and integration time response non-uniformity correction, Indicates the gray value of the (i, j)th pixel of one of the two calibration images after gain condition transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction ,

Represents the gray value of the (i, j)th pixel of the other calibration image of the two calibration images after gain condition transformation, gain response non-uniformity correction, and integral time response non-uniformity correction; using the formula: $\mathrm{the\; s}(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-r(i,j)\&Center\; Dot;G_{{\mathrm{\Φ}}_u}(i,j)$ The calculated s(i, j) and the adopted formula: $\mathrm{the\; s}(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}-r(i,j)\·G_{{\mathrm{\Φ}}_v}(i,j)$ The calculated s(i, j) is the same. Here, it should be noted that the pixels of all the calibration images involved in step 101 are exactly the same as the pixels of the star map to be corrected. In other words, M and N represent the pixel numbers of all calibration images and the image to be corrected.

当确定待校正星图与两个定标图像的增益及积分时间不同后， $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′\′}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_v}^{\′\′}(i,j)-1].$ Here, it needs to be explained that: when it is determined that the star map to be corrected is different from the gains of the two calibration images, $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_v}^{\′}(i,j)-1];$ When it is determined that the integration time of the star map to be corrected is different from that of the two calibration images,

When it is determined that the star map to be corrected and the gain and integration time of the two calibration images are different, $G_{{\mathrm{\Φ}}_u}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_u}^{\′\′}(i,j)-1],$ $G_{{\mathrm{\Φ}}_v}(i,j)=\ln [255/S_{{\mathrm{\Φ}}_v}^{\′\′}(i,j)-1].$

The more specific process of performing non-uniformity correction on the star map to be corrected by using the corrected calibration image can adopt the existing technology.

The corrected calibration image refers to: a calibration image after gain response non-uniformity correction and/or integral time response non-uniformity correction.

The present invention will be further described in detail below in conjunction with the examples.

Embodiment one

The non-uniform correction method for the high dynamic star sensor image of this embodiment, as shown in Figure 2, includes the following steps:

Step 201: Input the star map to be corrected and two calibration images, and then perform step 202;

Here, it is assumed that the gain of the star map to be corrected is g1, the gain of the two calibration images is g2, the integration time of the star map to be corrected is t1, and the integration time of the two calibration images is t2.

Step 202: Determine whether g1 is equal to g2, if yes, execute step 204, otherwise, execute step 203.

Step 203: Perform gain condition transformation and gain response non-uniformity correction on the two calibration images, and then perform step 204.

Step 204: Judging whether t1 is equal to t2, if yes, execute step 206, otherwise, execute step 205.

Step 205: Perform integration time condition transformation and integration time response non-uniformity correction on the two calibration images, and then perform step 206.

Step 206: Perform non-uniformity correction on the star map to be corrected, and then end the current processing flow.

Embodiment two

In this embodiment, it is determined that the star map to be corrected is different from the gain and integration time of the two calibration images, and the gain condition transformation and gain response non-uniformity correction and integration time are sequentially performed on the two calibration images. The processing method of condition transformation and integration time response non-uniformity correction, set the gain of the star map to be corrected as g1, the gain of the two calibration images as g2, the integration time of the star map to be corrected as t1, and the gain of the two calibration images The integration time is t2.

The non-uniform correction method for the high dynamic star sensor image of this embodiment, as shown in Figure 3, includes the following steps:

Step 301: Input the star map to be corrected and two calibration images, and then perform step 302.

Step 302: After determining that the star map to be corrected is different from the gain and integration time of the two calibration images, according to the ICCD imaging model, perform gain condition transformation, gain response non-uniformity correction and integration time on the two calibration images Response non-uniformity correction;

The specific process of performing gain condition transformation and gain response non-uniformity correction on the two calibration images, as shown in Figure 4, includes the following steps:

Step 302a: Acquire a calibration image for gain response non-uniformity correction;

Specifically, using the integrating sphere as a uniform light source, two images output by the ICCD under the conditions of gain g _m and g _n are obtained as gain response non-uniform correction calibration images; here, m and n can be selected arbitrarily.

Step 302b: Scan each of the gain response non-uniform correction calibration images row by row from top to bottom, and then execute step 302c.

Step 302c: Take the logarithm of the gray level of each pixel of each of the gain response non-uniform correction calibration images;

具体地，对于增益为g_m的增益响应非均匀校正定标图像，则有

对于增益为g_n的增益响应非均匀校正定标图像，则有

Here, after taking the logarithm, we have

Specifically, for the gain-response non-uniform correction calibration image with gain g _m , then there is

For the gain response non-uniform correction calibration image with gain g _n , then we have

Step 302d: Determine whether all the pixels of the two gain response non-uniformity correction calibration images have been processed, if yes, perform step 302e, otherwise, perform step 302b.

Step 302e: Using the data after linearization processing, calculate the mean value of the linearization processing of each of the gain response non-uniform correction calibration images;

计算增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值，采用公式：

计算增益为g_m的增益响应非均匀校正定标图像的灰度的线性化处理的均值。Specifically, using the formula:

Calculation gain is the average value of the linearization processing of the gray level of the gain response non-uniform correction calibration image of g _m , adopts the formula:

Calculate the mean value of the linearization process of the gray level of the gain response non-uniform correction calibration image with the gain g _m .

Step 302f: Determine whether the two gain response non-uniformity correction calibration images have been processed, if yes, perform step 302g, otherwise, perform step 302e.

Step 302g: Scan the linearized image progressively from top to bottom, and then perform step 301h.

Step 302h: adopt the formula: $c^{\′}(i,j)=\frac{{\stackrel{\‾}{Y}}_{g_m}-{\stackrel{\‾}{Y}}_{g_{\mathrm{no}}}}{Y_{g_m}(i,j)-Y_{g_{\mathrm{no}}}(i,j)},$ $d^{\′}(i,j)={\stackrel{\‾}{Y}}_{g_m}-c^{\′}(i,j)\&Center\; Dot;Y_{g_m}(i,j),$ Calculate the gain response non-uniformity correction coefficient, and then execute step 302i.

Step 302i: Determine whether all pixels have been processed, if yes, execute step 301j, otherwise execute step 302g.

Step 302j: Scan each calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′}(i,j)=\frac{\exp [c^{\′}(i,j)\&Center\; Dot;(c\·g1+d)+d^{\′}(i,j)]}{\exp [c^{\′}(i,j)\&Center\; Dot;(c\&Center\; Dot;g2+d)+d^{\′}(i,j)]}\·S_{\mathrm{\Φ}}(i,j),$ Perform gain condition transformation and gain response non-uniformity correction on each pixel of the two calibration images, and then execute step 302k.

Step 302k: Judging whether all the pixels of the two calibration images have been processed, if yes, execute step 302l, otherwise, execute step 302j.

Step 302l: Correct the non-uniformity of the integration time for the calibration image after gain condition transformation and gain response non-uniformity correction.

The specific process of carrying out integration time condition transformation and integration time response non-uniformity correction to the calibration image, as shown in Figure 5, includes the following steps:

Step 302A: Acquiring the integrated time response non-uniform correction calibration image;

Specifically, using the integrating sphere as a uniform light source, the ICCD output image under the conditions of integration time t _e and t _f is obtained as the integration time response non-uniform correction calibration image; here, e and f can be selected arbitrarily.

Step 302B: Integrate the time-response non-uniform correction calibration image from top to bottom progressively, and then perform step 302C.

Step 302C: Determine whether all the pixels of the two integral time response non-uniform correction calibration images have been processed, if yes, perform step 302D, otherwise, perform step 302B.

Step 302D: Calculate the mean value of the gray scale of each of the integrated time response non-uniform correction calibration images;

计算积分时间为t_f的积分时间响应非均匀校正定标图像的灰度的均值。Specifically, using the formula: Calculate the mean value of the gray level of the integral time response non-uniform correction calibration image whose integral time is t _e , adopt the formula:

Calculate the mean value of the gray level of the non-uniform correction calibration image with the integration time t _f in response to the integration time.

Step 302E: Determine whether the two integral time response non-uniform correction calibration images have been processed, if yes, perform step 302F, otherwise, perform step 302D.

Step 302F: Integrate the time-response non-uniform correction calibration image from top to bottom progressively, and then execute step 301G.

Step 302G: Adopt the formula: $a^{\′}(i,j)=\frac{{\stackrel{\‾}{S}}_{t_e}-{\stackrel{\‾}{S}}_{t_f}}{S_{t_e}(i,j)-S_{t_f}(i,j)},$ $b^{\′}(i,j)={\stackrel{\‾}{S}}_{t_e}-a^{\′}(i,j)\&Center\; Dot;S_{t_e}(i,j),$ Calculate the integration time response non-uniformity correction coefficient, and then execute step 302H.

Step 302H: Determine whether all pixels have been processed, if yes, execute step 302I, otherwise execute step 302F.

Step 302I: scan each image obtained by performing gain condition transformation and gain response non-uniformity correction on the calibration image line by line from top to bottom, and then according to the formula: $S_{\mathrm{\Φ}}^{\′\′}(i,j)=\frac{a^{\′}(i,j)\·(a\·t1+b)+b^{\′}(i,j)}{a^{\′}(i,j)\·(a\&Center\; Dot;t2+b)+b^{\′}(i,j)}\&Center\; Dot;S_{\mathrm{\Φ}}^{\′}(i,j),$ Perform integration time condition transformation and integration time response non-uniformity correction on the two calibration images, and then perform step 302J.

Step 302J: judge whether all the pixels of the two calibration images have been processed, if yes, execute step 303, otherwise, execute step 302I;

Here, the calibration image in this step refers to the calibration image after gain condition transformation and gain response non-uniformity correction.

Step 303: Use the corrected calibration image to correct the non-uniformity of the star map to be corrected, and then end the current processing flow;

The specific implementation of this step, as shown in Figure 6, includes the following steps:

Step 303a: Scan each corrected calibration image line by line from top to bottom, and then perform step 303b.

Step 303b: Linearize the grayscale of each pixel of each corrected calibration image;

对校正后的两个定标图像中的另一个定标图像的每个像素的灰度进行线性化处理。Specifically, using the formula: Linearize the gray level of each pixel of one of the two corrected calibration images; using the formula:

A linearization process is performed on the gray level of each pixel of the other of the two corrected calibration images.

Step 303c: Determine whether all the pixels of the two corrected calibration images have been processed, if yes, execute step 303d, otherwise, execute step 303a.

Step 303d: Using the linearized data, calculate the linearized mean value of each corrected calibration image;

计算校正后的两个定标图像中的一个定标图像的线性化处理的均值，采用公式：

计算校正后的两个定标图像中的另一个定标图像的线性化处理的均值。Specifically, using the formula:

Computes the linearized mean of one of the two corrected calibration images using the formula:

Computes the linearized mean of the other of the two corrected calibration images.

Step 303e: Determine whether the two corrected calibration images have been processed, if yes, execute step 303f, otherwise, execute step 303d.

Step 303f: Scan the linearized image progressively from top to bottom, and then perform step 303g.

Step 303g: adopt the formula: $r(i,j)=\frac{{\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-{\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_v}}{G_{{\mathrm{\Φ}}_u}(i,j)-G_{{\mathrm{\Φ}}_v}(i,j)},$ $\mathrm{the\; s}(i,j)={\stackrel{\‾}{G}}_{{\mathrm{\Φ}}_u}-r(i,j)\&Center\; Dot;G_{{\mathrm{\Φ}}_u}(i,j),$ Calculate the star map non-uniformity correction coefficient, and then execute step 303h.

303h: Determine whether all pixels have been processed, if yes, execute step 303i, otherwise execute step 303f.

Step 303i: Scan the star map to be corrected line by line from top to bottom, and then perform step 303j.

Step 303k: Linearize the grayscale of each pixel of the star map to be corrected;

Specifically, the formula: S _linear (i, j)=ln[255/S _{star map} (i, j)-1] is used to linearize the grayscale of each pixel of the star map to be corrected.

Step 303k: According to the formula: S _correction (i, j) = 255/{exp[r (i, j) · S _linear (i, j) + s (i, j)] + 1}, the to-be-corrected Perform non-uniformity correction on each pixel of the star map, and then perform step 3031 .

Step 3031: Determine whether all the pixels of the star map to be corrected have been processed, if yes, end the current processing flow, otherwise, execute step 303i.

Embodiment three

In this embodiment, when the high dynamic star sensor of the ICCD image sensor collects the star map, the gain used when collecting the star map is 105, and the integration time is 4ms. The gain used when shooting the calibration image is 75, and the integration time is 6ms. That is: the gain of the star map to be corrected is 105, the integration time is 4ms, the gain of the calibration image is 75, and the integration time is 6ms.

Utilize the improved two-point temperature calibration correction method based on the detector nonlinear response model and the high dynamic star sensor image non-uniformity correction method provided by the present invention to perform star map non-uniformity correction. Table 1 provides two methods the calibration result.

Table 1

As can be seen from Table 1, compared with the improved two-point temperature calibration correction method based on the nonlinear response model of the detector, the star map correction is performed by using the high dynamic star sensor image non-uniformity correction method provided by the present invention Finally, the non-uniformity of the star map is effectively suppressed, and the accuracy of the star sensor is effectively improved.

In order to realize the above method, the present invention also provides a high dynamic star sensor image non-uniformity correction device, as shown in Figure 7, the device includes: a gain and integration time correction module 71 and a non-uniformity correction module 72; wherein,

Gain and integration time correction module 71, after determining that the star map to be corrected and the gain and/or integration time of the two calibration images are different, according to the imaging model of ICCD, perform gain condition conversion and gain adjustment on the two calibration images Response non-uniformity correction, and/or integration time condition transformation and integration time response non-uniformity correction, and send the corrected image to the non-uniformity correction module 72;

The non-uniformity correction module 72 is configured to perform non-uniformity correction on the star map to be corrected by using the corrected calibration image after receiving the corrected image sent by the gain and integration time correction module 71;

The imaging model of the ICCD is: S=f(Φ) exp(c g+d) (a t+b); wherein, S represents the ICCD response, and Φ represents the irradiance received by the star sensor , t represents the exposure time of ICCD, g represents the gain of ICCD, and a, b, c, d are constants. .

Wherein, the device may further include: a setting module, which is used for the photoelectric response model of the imaging device, and establishes the imaging model of the ICCD.

Here, the specific processing procedures of the gain and integration time correction module and the non-uniformity correction module of the device of the present invention have been described in detail above, and will not be repeated here.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention.

Claims (6)

1. A method for correcting the image non-uniformity of a high-dynamic star sensor is characterized by comprising the following steps:

after determining that the star map to be corrected and the two calibration images have different gains and/or different integration times, performing gain condition transformation and gain response non-uniformity correction and/or integration time transformation and integration time response non-uniformity correction on the two calibration images according to an imaging model of an image enhancement type charge coupled device (ICCD);

carrying out non-uniformity correction on the star map to be corrected by adopting the corrected calibration image;

the imaging model of the ICCD is as follows: s ═ f (Φ) · exp (c · g + d) · (a · t + b); wherein S represents ICCD response, phi represents irradiance received by the star sensor, t represents exposure time of the ICCD, g represents gain of the ICCD, and a, b, c and d are constants; wherein,

when the gains of the star map to be corrected and the two calibration images are different, performing gain condition transformation and gain response non-uniformity correction on the two calibration images, wherein the gain condition transformation and the gain response non-uniformity correction are as follows:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:

carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the gain-responsive non-uniformity correction coefficients comprising: c '(i, j) and d' (i, j), c '(i, j) representing the correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected, g2 representing the gain of the scaled image,

or,

wherein,

represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

wherein M, N represents the number of pixels of the gain response non-uniformity correction scaled image,

represents a gain of g_mGain ofIn response to the gray value of the (i, j) th pixel of the non-uniformity correction scaled image,

represents a gain of g_nIn response to the gray value of the (i, j) th pixel of the non-uniform correction scaled image; or,

when the integration time of the star map to be corrected is determined to be different from the integration time of the two calibration images, the integration time condition transformation and the integration time response non-uniformity correction are carried out on the two calibration images, and the method comprises the following steps:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each of the gain response non-uniformity corrected images of the scaled image line by line from top to bottom, and then according to the formula:

carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the integrated time response non-uniformity correction factor comprising: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,wherein,

representing integration time t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,

representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

wherein MN represents the pixel number of the integral response non-uniformity correction calibration image; or,

when the gain and the integration time of the star map to be corrected and the two calibration images are determined to be different, and the two calibration images are subjected to gain condition transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction in sequence, the two calibration images are subjected to gain condition transformation and gain response non-uniformity correction, and the method comprises the following steps:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the gain-responsive non-uniformity correction coefficients comprising: c '(i, j) and d' (i, j), c '(i, j) representing the correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected, g2 representing the gain of the scaled image,

or,wherein,

represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

wherein M, N represents the number of pixels of the gain response non-uniformity correction scaled image,

represents a gain of g_mIn response to the gray value of the (i, j) th pixel of the non-uniformly corrected scaled image,

represents a gain of g_nIn response to the gray value of the (i, j) th pixel of the non-uniform correction scaled image;

correspondingly, the performing integral time condition transformation and integral time response non-uniformity correction on the two calibration images comprises:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each image obtained by performing gain condition transformation and gain response non-uniformity correction on the calibration image line by line from top to bottom, and then according to a formula:

carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein, S'_Φ(i, j) represents a gray value of an (i, j) th pixel of each image after the gain condition transformation and the gain response non-uniformity correction are performed on the calibration image, and the integration time response non-uniformity correction coefficient includes: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,

wherein,

representing integration time t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

or,

when the gain and the integration time of the star map to be corrected and the two calibration images are determined to be different, and the two calibration images are subjected to integral time condition transformation and integral time response non-uniformity correction, gain condition transformation and gain response non-uniformity correction in sequence, the integral time condition transformation and the integral time response non-uniformity correction are performed on the two calibration images, and the steps are as follows:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:

carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the integrated time response non-uniformity correction factor comprising: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,

wherein,

representing integration time t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,

representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

correspondingly, the performing gain condition transformation and gain response non-uniformity correction on the two scaled images includes:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning the images obtained after the integral time condition transformation and integral time response non-uniformity correction are carried out on the calibration images from top to bottom line by line, and then according to a formula:

carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is^p _Φ(i, j) represents a gray value of an (i, j) th pixel of each image after integration time conversion and integration time response non-uniformity correction of the scaled image, and the gain response non-uniformity correction coefficient includes: c '(i, j) and d' (i, j), c '(i, j) representing the gain correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected, g2 representing the gain of the star map to be corrected, c '(i, j) and d' (i, j) representing the gain of the (i, j) th pixel, g 3578 representing the gain of the star map toThe gain of the image is scaled in such a way that,

or,

wherein,

represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

2. the method of claim 1, wherein said obtaining a gain response non-uniformity correction scaled image is:

using an integrating sphere as a uniform light source to obtain a gain g_mAnd g_nTwo images of the ICCD output under the conditions were scaled images for gain response non-uniformity correction.

3. The method of claim 1, wherein said obtaining an integrated time response non-uniformity correction scaled image is:

using an integrating sphere as a uniform light source to obtain the integration time t_eAnd t_fUnder the condition, the ICCD outputs an image as an integral time response non-uniform correction calibration image.

4. A method according to any one of claims 1 to 3, characterized in that the method further comprises:

and establishing an ICCD imaging model according to the photoelectric response model of the imaging device.

5. A high-dynamic star sensor image non-uniformity correction device is characterized by comprising: a gain and integration time correction module and a non-uniformity correction module; wherein,

the gain and integration time correction module is used for carrying out gain condition transformation and gain response non-uniformity correction and/or integration time condition transformation and integration time response non-uniformity correction on the two calibration images according to an imaging model of the ICCD after determining that the star map to be corrected and the two calibration images have different gains and/or different integration times, and sending the corrected images to the non-uniformity correction module;

the non-uniformity correction module is used for carrying out non-uniformity correction on the star map to be corrected by adopting the corrected calibration image after receiving the corrected image sent by the gain and integration time correction module;

the imaging model of the ICCD is as follows: s ═ f (Φ) · exp (c · g + d) · (a · t + b); wherein S represents ICCD response, phi represents irradiance received by the star sensor, t represents exposure time of the ICCD, g represents gain of the ICCD, and a, b, c and d are constants; wherein,

when the gains of the star map to be corrected and the two calibration images are different, performing gain condition transformation and gain response non-uniformity correction on the two calibration images, wherein the gain condition transformation and the gain response non-uniformity correction are as follows:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:

carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the gain-responsive non-uniformity correction coefficients comprising: c '(i, j) and d' (i, j), c '(i, j) representing the correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected, g2 representing the gain of the scaled image,

or,

wherein,represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

wherein M, N represents the number of pixels of the gain response non-uniformity correction scaled image,

representing the gainIs g_mIn response to the gray value of the (i, j) th pixel of the non-uniformly corrected scaled image,represents a gain of g_nIn response to the gray value of the (i, j) th pixel of the non-uniform correction scaled image; or,

when the integration time of the star map to be corrected is determined to be different from the integration time of the two calibration images, the integration time condition transformation and the integration time response non-uniformity correction are carried out on the two calibration images, and the method comprises the following steps:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each of the gain response non-uniformity corrected images of the scaled image line by line from top to bottom, and then according to the formula:

carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the integrated time response non-uniformity correction factor comprising: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,wherein,representing integration time t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean of the gray levels of the scaled image non-uniformly,

representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,

representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

wherein M, N represents the number of pixels of the integrated response non-uniformity correction scaled image; or,

when the gain and the integration time of the star map to be corrected and the two calibration images are determined to be different, and the two calibration images are subjected to gain condition transformation and gain response non-uniformity correction, integration time condition transformation and integration time response non-uniformity correction in sequence, the two calibration images are subjected to gain condition transformation and gain response non-uniformity correction, and the method comprises the following steps:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the gain-responsive non-uniformity correction coefficients comprising: c '(i, j) and d' (i, j), c '(i, j) representing the correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected, g2 representing the gain of the scaled image,

or,

wherein,

represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

wherein M, N represents the number of pixels of the gain response non-uniformity correction scaled image,represents a gain of g_mIn response to the gray value of the (i, j) th pixel of the non-uniformly corrected scaled image,

represents a gain of g_nIn response to the gray value of the (i, j) th pixel of the non-uniform correction scaled image;

correspondingly, the performing integral time condition transformation and integral time response non-uniformity correction on the two calibration images comprises:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each image obtained by performing gain condition transformation and gain response non-uniformity correction on the calibration image line by line from top to bottom, and then according to a formula:

carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein, S'_Φ(i, j) represents a gray value of an (i, j) th pixel of each image after the gain condition transformation and the gain response non-uniformity correction are performed on the calibration image, and the integration time response non-uniformity correction coefficient includes: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,

wherein,

means product ofTime division of t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,

representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

or,

when the gain and the integration time of the star map to be corrected and the two calibration images are determined to be different, and the two calibration images are subjected to integral time condition transformation and integral time response non-uniformity correction, gain condition transformation and gain response non-uniformity correction in sequence, the integral time condition transformation and the integral time response non-uniformity correction are performed on the two calibration images, and the steps are as follows:

acquiring an integral time response non-uniform correction calibration image;

scanning the integral time response non-uniform correction calibration image line by line from top to bottom, and calculating the mean value of the gray scale of each integral time response non-uniform correction calibration image according to an ICCD imaging model;

scanning an integral time response non-uniform correction calibration image line by line from top to bottom, and calculating an integral time response non-uniform correction coefficient by adopting the mean value of the gray scales of all the integral time response non-uniform correction calibration images;

scanning each of said scaled images line by line from top to bottom, and then according to the formula:carrying out integral time condition transformation and integral time response non-uniformity correction on the two calibration images; wherein S is_Φ(i, j) represents a gray value of an (i, j) th pixel of each of the scaled images, the integrated time response non-uniformity correction factor comprising: a '(i, j) and b' (i, j), a '(i, j) represents the correction gain of the integration time of the (i, j) th pixel, b' (i, j) represents the correction offset of the integration time of the (i, j) th pixel, t1 represents the integration time of the star map to be corrected, t2 represents the integration time of the scaled image,

or,

wherein,

representing integration time t_eThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_fThe integration time response of (a) corrects the mean value of the gray scale of the scaled image non-uniformly,

representing integration time t_eThe (i, j) th pixel of the non-uniform correction scaled image,representing integration time t_fThe (i, j) th pixel of the non-uniform correction scaled image,

correspondingly, the performing gain condition transformation and gain response non-uniformity correction on the two scaled images includes:

acquiring a gain response non-uniform correction calibration image;

according to an ICCD imaging model, carrying out linearization processing on the gray scale of each pixel of each gain response non-uniform correction calibration image, and calculating the mean value of the linearization processing of each gain response non-uniform correction calibration image by adopting data after linearization processing;

scanning the linearized images line by line from top to bottom, and calculating gain response non-uniformity correction coefficients by adopting the mean value of the linearization treatment of all the gain response non-uniformity correction calibration images;

scanning the images obtained after the integral time condition transformation and integral time response non-uniformity correction are carried out on the calibration images from top to bottom line by line, and then according to a formula:

carrying out gain condition transformation and gain response non-uniformity correction on the two calibration images; wherein S is^p _Φ(i, j) represents a gray value of an (i, j) th pixel of each image after integration time conversion and integration time response non-uniformity correction of the scaled image, and the gain response non-uniformity correction coefficient includes: c '(i, j) and d' (i, j), c '(i, j) representing the correction gain of the (i, j) th pixel, d' (i, j) representing the correction offset of the gain of the (i, j) th pixel, g1 representing the gain of the star map to be corrected,g2 represents the gain of the scaled image,

or,

wherein,

represents a gain of g_mIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,represents a gain of g_nIn response to the average of the linearization process of non-uniformly correcting the gray scale of the scaled image,

represents a gain of g_mIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

represents a gain of g_nIn response to the linearized data of the gray scale of the (i, j) th pixel of the non-uniform correction scaled image,

6. the apparatus of claim 5, further comprising: and the setting module is used for establishing an ICCD imaging model by using a photoelectric response model of the imaging device.

Images