On-orbit radiometric characteristics assessment foroptical imaging remote sensor—VIS-SWIR
Some standard content:
ICS07.040
National Standard of the People's Republic of China
GB/T38935—2020
On-orbit radiometric characteristics assessment for optical imaging remote sensor-VIS-SWIR2020-07-21Release
State Administration for Market Regulation
Standardization Administration of China
2021-02-01Implementation
Normative reference documents
Terms and definitions, abbreviations
Terms and definitions
Abbreviations
Evaluation objects and evaluation indicators
Evaluation objects
Evaluation indicators
Evaluation methods
Signal-to-noise ratio
Radiance resolution
Dynamic range| |tt||Nonlinearity
Blind element rate
Appendix A (Informative Appendix)
Appendix B (Informative Appendix)
References
Signal-to-noise ratio specification method
GB/T38935—2020
On-orbit dynamic range and nonlinearity evaluation of high spatial resolution optical remote sensors Ground target layout requirements 12
This standard was drafted in accordance with the rules given in GB/T1.1-2009 GB/T38935—2020
Please note that some of the contents of this document may involve patents. The issuing agency of this document does not assume the responsibility for identifying these patents. This standard was proposed by the Chinese Academy of Sciences.
This standard is under the jurisdiction of the National Remote Sensing Technology Standardization Technical Committee (SAC/TC327). The drafting units of this standard are: Institute of Optoelectronics, Chinese Academy of Sciences, China Resources Satellite Application Center, National Satellite Meteorological Center, Beijing Institute of Space Mechatronics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, and Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences. The main drafters of this standard are: Li Chuanrong, Li Xiaohui, Wang Xinhong, Gao Caixia, Tang Lingli, Ma Lingling, Wang Ning, Fu Qiaoyan, Fang Xiang, Fu Ruimin Ma Yanhua, Wang Gang, Li Wei, Liu Zhaoyan, Zhao Yongguang, Zhu Bo, Zhang Jing, Zhu Jiajia, Liu Yaokai, Qian Yonggang, Qiu Shi, Zhou Yongsheng, Zhu Xiaohua, Ren Lum
1 Scope
Radiation performance of on-orbit imaging of optical remote sensors
Evaluation method Visible light-shortwave infrared
GB/T38935—2020
This standard specifies the evaluation objects, evaluation indicators and evaluation methods for on-orbit imaging radiation performance of spaceborne imaging optical remote sensors with working wavelengths in the visible light to shortwave infrared range (380nm～2500nm). This standard is applicable to the radiation performance evaluation of passive optical remote sensors carried on satellite platforms and using linear array detectors for scanning imaging during on-orbit operation. The evaluation of the on-orbit field radiation performance of other types of spaceborne passive optical remote sensors can refer to the use of normative reference documents
The following documents are indispensable for the application of this document. For any dated referenced document, only the dated version applies to this document. For any undated referenced document, its latest version (including all amendments) applies to this document GB/T33988-2017 Measurement of visible light-shortwave infrared spectral reflectance of urban objects GB/T36297-2018 Field test evaluation index of optical remote sensing payload performance GB/T36540-2018 Measurement of visible light-shortwave infrared spectral reflectance of water bodies 3 Terms and definitions, abbreviations
3.1 Terms and definitions
The terms and definitions defined in GB/T36297-2018 and the following terms and definitions apply to this document 3.1.1
Radiometric characteristics The ability of a remote sensor to obtain and maintain the relative or absolute radiation energy distribution of a target scene during imaging. 3.1.2
Signal-to-noise ratiosignal-to-noiseratio
The ratio of the effective signal power to the noise power output by the remote sensor. 3.1.3
Eradiometricresolution
Radiometric resolution
The minimum radiometric difference that the remote sensor sensing (sensitive) element can distinguish when receiving the spectral radiation signal, or the ability to distinguish the radiation of two different radiation sources.
[GB/T14950—2009, definition 4.103] 3.1.4
Dynamic rangedynamicrange
The interval defined by the minimum incident radiation and the maximum incident radiation that the optical remote sensor output can change with the input. 3.1.5
non-linearity
non-linearity
Within the dynamic range, the maximum deviation of the actual value of the response from the corresponding fitting straight line is 1
GB/T38935—2020
Note 1: Generally expressed in percentage,
Note 2: Rewritten from GB/T17444—2013, definition 2.37. 3.1.6
blindpixelratio
The percentage of blind pixels in the remote sensor to the total number of pixels. 3.2
Abbreviations
The following abbreviations apply to this document.
VIS-SWIR
Ground sampling distance (groundsamplingdistance) Visible to short-wave infrared (visible to short-wave infrared) 4 Evaluation objects and evaluation indicators
4.1 Evaluation objects
This standard is aimed at passive optical remote sensors (such as visible light panchromatic cameras, multispectral cameras, hyperspectral imagers, etc.) carried on satellite platforms, with working wavelengths in the visible to short-wave infrared range (380nm~2500nm) and using linear array detectors for scanning imaging. The image data obtained during its on-orbit operation is used to analyze and monitor the radiation performance of the remote sensor. 4.2
Evaluation indicators
The evaluation indicators of the on-orbit imaging radiation performance of space-borne optical remote sensors include: signal-to-noise ratio, radiation resolution, dynamic range, nonlinearity, and blind pixel rate.
5 Evaluation methods
5.1 Signal-to-noise ratio
Evaluation is based on the remote sensor observation image data that has only been processed with relative radiation correction. The specific steps are as follows: a) Select several images of large-area uniform scenes with different grayscale levels (grayscale level ≥ 5) acquired by the remote sensor in a similar time period, which are less affected by the atmosphere, as sample images, such as water bodies such as deep sea and lakes, large-scale ice and snow covered areas in Antarctica or Greenland, deserts, Gobi or uniform dense vegetation covered areas, etc. The uniform area in the image should be as large as possible (at least larger than 50 pixels × 50 pixels), and the best case is to cover all pixels in the array along the direction of the remote sensor array and be greater than or equal to 100 pixels along the scanning direction of the remote sensor array (vertical to the remote sensor array direction). b) For a certain band l of the sample image (l=1, 2, Nburd, Nhnd is the number of remote sensor bands), select a uniform area in the image that meets the signal-to-noise ratio evaluation requirements (the grayscale mean of all pixels in the area is D., corresponding to a certain grayscale level k, k=1, 2, K, K is the number of grayscale levels of the uniform scene participating in the evaluation) as the uniform area sub-image I. If the remote sensor adopts array scanning imaging, the uniform area sub-image I is transposed using formula (1), where:
pij..=pi.il..
........(
the gray value of the uniform area sub-image I.. in the ith row and jth column at the gray level k of the lth band. pii
pi.i——the gray value of the uniform area in the image I. in the jth row and ith column at the gray level k of the /th band. c) The difference image dij.i.. of the uniform area sub-image I. is calculated using formula (2). =pi+)jfpi.j..
Where:
GB/T38935—2020
The gray value of the ith row and column of the difference image of the uniform area sub-image I. at the gray level k of the first band; The gray value of the ith plus 1st row and ith column of the uniform area sub-image I. at the gray level k of the first band; d) Calculate the column noise of each column of the uniform area sub-image I at the gray level k of the first band using formula (3). (dij.)
Where:
のj——the noise of the ith column of the uniform area sub-image I. at the gray level k of the first band; Mi——the number of rows of the uniform area sub-image I. at the gray level k of the first band. e) Calculate the uniform area sub-image I. at the gray level k of the first band using formula (4)..The signal-to-noise ratio of each column. RsNRj
Where:
The column signal-to-noise ratio of the jth column of the uniform area sub-image I. at the gray level k of the first band. (3)
f) Use formula (5) to calculate the average value of the column signal-to-noise ratio as the signal-to-noise ratio of the uniform area sub-image I. at the gray level k of the first band. RR
RsNRI=
Where:
The signal-to-noise ratio of the uniform area sub-image I at the gray level k of the first band; N.
--The number of columns of the uniform area in the image I at the gray level k of the first band (5)
g) According to steps b) to f), calculate the signal-to-noise ratio of the uniform area sub-image I. at all gray levels of the first band in the sample image. h) Using the signal-to-noise ratio RsNR obtained in step g) (k=1, 2, KK is the number of uniform scene gray levels involved in the evaluation), the signal-to-noise ratio normalization method (see Appendix A) is used to calculate the signal-to-noise ratio RsNR.
obtained in step h) at a certain reference entrance pupil radiance level Lo. The signal-to-noise ratio RsNR.
is converted into the power representation of the signal-to-noise ratio using formula (6). i)
RDBSNR.=20·Ig(RsNRa.）
Where:
RDBSNRO.t
The power representation of the signal-to-noise ratio of the first band at a certain reference human pupil radiance level L. ., in decibels (dB). According to step b) i), the signal-to-noise ratio RpB-SNRa.r of all bands of the remote sensor is obtained (l=1, 2, Nband, Nband is the number of bands of the remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, the vertical axis is the signal-to-noise ratio) to display the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiometric resolution
Use noise equivalent radiance or noise equivalent reflectance to evaluate based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select sample images according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2,, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation! , the same as step 3 in 5.1 (step b) to h).
c) Use equations (7) to (9) to calculate the radiation resolution of the first band. Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.(7)
..(8)
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",JThe column signal-to-noise ratio of the jth column. (3)
f) Use formula (5) to calculate the average value of the column signal-to-noise ratio as the signal-to-noise ratio of the uniform area sub-image I. at the gray level k of the first band. RR
RsNRI=
Wherein:
The signal-to-noise ratio of the uniform area sub-image I at the gray level k of the first band; N.
The number of columns of the uniform area in the image I at the gray level k of the first band (5)
g) According to steps b) to f), calculate the signal-to-noise ratio of the uniform area sub-image I. at all gray levels of the first band in the sample image. h) Using the signal-to-noise ratio RsNR obtained in step g) (k=1, 2, KK is the number of uniform scene gray levels involved in the evaluation), the signal-to-noise ratio normalization method (see Appendix A) is used to calculate the signal-to-noise ratio RsNR.
obtained in step h) at a certain reference entrance pupil radiance level Lo. The signal-to-noise ratio RsNR.
is converted into the power representation of the signal-to-noise ratio using formula (6). i)
RDBSNR.=20·Ig(RsNRa.）
Where:
RDBSNRO.t
The power representation of the signal-to-noise ratio of the first band at a certain reference human pupil radiance level L. ., in decibels (dB). According to step b) i), the signal-to-noise ratio RpB-SNRa.r of all bands of the remote sensor is obtained (l=1, 2, Nband, Nband is the number of bands of the remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, the vertical axis is the signal-to-noise ratio) to display the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiometric resolution
Use noise equivalent radiance or noise equivalent reflectance to evaluate based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select sample images according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2,, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation! , the same as step 3 in 5.1 (step b) to h).
c) Use equations (7) to (9) to calculate the radiation resolution of the first band. Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.(7)
..(8)
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",JThe column signal-to-noise ratio of the jth column. (3)
f) Use formula (5) to calculate the average value of the column signal-to-noise ratio as the signal-to-noise ratio of the uniform area sub-image I. at the gray level k of the first band. RR
RsNRI=
Wherein:
The signal-to-noise ratio of the uniform area sub-image I at the gray level k of the first band; N.
The number of columns of the uniform area in the image I at the gray level k of the first band (5)
g) According to steps b) to f), calculate the signal-to-noise ratio of the uniform area sub-image I. at all gray levels of the first band in the sample image. h) Using the signal-to-noise ratio RsNR obtained in step g) (k=1, 2, KK is the number of uniform scene gray levels involved in the evaluation), the signal-to-noise ratio normalization method (see Appendix A) is used to calculate the signal-to-noise ratio RsNR.
obtained in step h) at a certain reference entrance pupil radiance level Lo. The signal-to-noise ratio RsNR.
is converted into the power representation of the signal-to-noise ratio using formula (6). i)
RDBSNR.=20·Ig(RsNRa.）
Where:
RDBSNRO.t
The power representation of the signal-to-noise ratio of the first band at a certain reference human pupil radiance level L. ., in decibels (dB). According to step b) i), the signal-to-noise ratio RpB-SNRa.r of all bands of the remote sensor is obtained (l=1, 2, Nband, Nband is the number of bands of the remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, the vertical axis is the signal-to-noise ratio) to display the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiometric resolution
Use noise equivalent radiance or noise equivalent reflectance to evaluate based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select sample images according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2,, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation! , the same as step 3 in 5.1 (step b) to h).
c) Calculate the radiation resolution of the first band using equations (7) to (9). Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.(7)
..(8)
...(9)
The noise equivalent radiance of the remote sensor in the first band, in watts per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",J
——Number of columns of uniform area in image I at gray level k of band 1 (5)
g) According to steps b) to f), calculate the signal-to-noise ratio of uniform area sub-image I, at all gray levels of band 1 in the sample image. h) Using the signal-to-noise ratio RsNR obtained in step g) (k=1, 2, KK is the number of gray levels of uniform scenes involved in the evaluation), the signal-to-noise ratio RsNR. of band 1 at a certain reference entrance pupil radiance level Lo. is calculated by using the signal-to-noise ratio normalization method (see Appendix A).
Use formula (6) to convert the signal-to-noise ratio RsNR. obtained in step h) into a power representation of the signal-to-noise ratio. i)
RDBSNR.=20·Ig(RsNRa.）
Where:
RDBSNR.t
The first band at a certain reference human pupil radiance level L. . The power representation of the signal-to-noise ratio under the condition of . is expressed in decibels (dB). According to step b) i), the signal-to-noise ratio RpB-SNRa.r of all bands of the remote sensor is obtained (l=1, 2,, Nband, Nband is the number of bands of remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at a frequency of at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, and the vertical axis is the signal-to-noise ratio) to display the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiation resolution
is expressed by noise equivalent radiance or noise equivalent reflectance, and is evaluated based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select a sample image according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation!, the same as step 3 b) to h in 5.1).
c) Calculate the radiation resolution of the first band using equations (7) to (9). Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.（7）
..（8）
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:bzxz.net
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",J
——Number of columns of uniform area in image I at gray level k of band 1 (5)
g) According to steps b) to f), calculate the signal-to-noise ratio of uniform area sub-image I, at all gray levels of band 1 in the sample image. h) Using the signal-to-noise ratio RsNR obtained in step g) (k=1, 2, KK is the number of gray levels of uniform scenes involved in the evaluation), the signal-to-noise ratio RsNR. of band 1 at a certain reference entrance pupil radiance level Lo. is calculated by using the signal-to-noise ratio normalization method (see Appendix A).
Use formula (6) to convert the signal-to-noise ratio RsNR. obtained in step h) into a power representation of the signal-to-noise ratio. i)
RDBSNR.=20·Ig(RsNRa.）
Where:
RDBSNR.t
The first band at a certain reference human pupil radiance level L. . The power representation of the signal-to-noise ratio under the condition of . is expressed in decibels (dB). According to step b) i), the signal-to-noise ratio RpB-SNRa.r of all bands of the remote sensor is obtained (l=1, 2,, Nband, Nband is the number of bands of remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at a frequency of at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, and the vertical axis is the signal-to-noise ratio) to display the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiation resolution
is expressed by noise equivalent radiance or noise equivalent reflectance, and is evaluated based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select a sample image according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation!, the same as step 3 b) to h in 5.1).
c) Calculate the radiation resolution of the first band using equations (7) to (9). Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.（7）
..（8）
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",Jt
The power expression of the signal-to-noise ratio of the band under a certain reference human pupil radiance level L. ., in decibels (dB). According to step b) i), the signal-to-noise ratio of all bands of the remote sensor RpB-SNRa.r (l=1, 2,, Nband, Nband is the number of bands of the remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, the vertical axis is the signal-to-noise ratio) to show the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiometric resolution
Use noise equivalent radiance or noise equivalent reflectance to express it, and evaluate it based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select a sample image according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation!, the same as step 3 b) to h in 5.1).
c) Calculate the radiation resolution of the first band using equations (7) to (9). Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.（7）
..（8）
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5 pixels.
b) Select clear and cloudless weather with clean atmosphere (visibility should be greater than 23km under normal circumstances) to carry out satellite-ground synchronization test while the remote sensor is passing, and synchronously measure the ground parameter data and atmospheric parameter data of the test area or ground target: 1) In accordance with the provisions of GB/T33988-2017 and GB/T36540-2018, use a spectrometer to measure the surface reflectance spectral data of the test area or target ti, t2,, tk (K≥4); use sun photometer, ground-based lidar and other atmospheric measurement equipment to synchronously obtain atmospheric parameter data such as atmospheric water vapor content and aerosol optical characteristics parameters at the time of remote sensor imaging. c) According to the imaging time of the remote sensor and the satellite orbit related parameters, obtain the auxiliary parameters such as the solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc. when the remote sensor images the test area or target. d) Based on the surface reflectance spectrum data of the test area or target measured in the above synchronous ground test, calculate the equivalent reflectance of each band of the remote sensor according to formula (10):
Where:
In, In12
p(a)s,(a)d)
S, a)da
GB/T38935—2020
..（10）
The equivalent reflectivity of the test area or target t (k=1, 2,, K, K≥4) in the Ith (l=1, 2, \, Nhand, Nhand is the number of remote sensor bands) band of the remote sensor; the surface reflectivity spectral data of the test area or target measured by the spectrometer; the normalized spectral response function of the first band of the remote sensor; the wavelength range of the first band of the remote sensor, in micrometers (μm). e) Based on the surface reflectance spectral data, atmospheric parameter data and auxiliary parameters (solar zenith angle, solar azimuth angle, observation zenith angle, observation azimuth angle, satellite platform height, etc.) obtained and calculated by the above synchronous test, the atmospheric radiation transfer model is used to simulate and calculate the simulated radiance L1, L2, Lk (l1, 2, Nband) of each test area or target t1t2, tk at the entrance pupil of the remote sensor in each band, the unit is watt per square meter steradian micrometer [W/(m2·sr·μm)]. f) For a certain band I (I=1, 2, Nband, Nhand is the number of remote sensor bands) of the observation image data of each test area or target obtained by the remote sensor in the synchronous test, the average grayscale value D. of each test area or target ti, t2, tk in the image corresponding to the uniform area in the image I.. is calculated using formula (11) (l=1, 2, Nhand; k=1, 2, K, K≥4); Mi.t N!
Di.=M..XN..
Wherein:
.(11)
The average grayscale value of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data; pi.je
The grayscale value of the i-th row and column of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data;
The number of rows of the uniform area sub-image I. corresponding to the k-th test area or target t in the first band image data; -The number of columns of the uniform area sub-image I corresponding to the k-th test area or target t in the first band image data. For band 1, select the test area or target t that is within the dynamic range of the load and has not reached saturation (k=1,,",Jt
The power expression of the signal-to-noise ratio of the band under a certain reference human pupil radiance level L. ., in decibels (dB). According to step b) i), the signal-to-noise ratio of all bands of the remote sensor RpB-SNRa.r (l=1, 2,, Nband, Nband is the number of bands of the remote sensor j
).
k) Calculate the signal-to-noise ratio of each band of the remote sensor according to steps a) to j) at least once a year. Use a two-dimensional line graph (the horizontal axis is the evaluation period, the vertical axis is the signal-to-noise ratio) to show the characteristics and trends of the signal-to-noise ratio of each band of the remote sensor over time (for multispectral remote sensors, the signal-to-noise ratio change curves of different bands can be displayed in the same line graph). 5.2 Radiometric resolution
Use noise equivalent radiance or noise equivalent reflectance to express it, and evaluate it based on the remote sensor observation image data that has only been processed by relative radiation correction. The specific steps are as follows:
a) Select a sample image according to the same method and requirements as the signal-to-noise ratio evaluation, the same as step a in 5.1). b) For a certain band l of the sample image (l=1, 2, Nband, Nhand is the number of remote sensor bands), calculate the signal-to-noise ratio RsNRo. of the first band of the remote sensor at a certain reference human pupil radiance level Lo. according to the same method as the signal-to-noise ratio evaluation!, the same as step 3 b) to h in 5.1).
c) Calculate the radiation resolution of the first band using equations (7) to (9). Lo.
RNEALn=
RsNRo-t
元Lrd2
po..=E..coso
Where:
.（7）
..（8）
...(9)
The noise equivalent radiance of the remote sensor in the first band, the unit is watt per square meter steradian micrometer [W/(m·sr ·μm)]; RNEAPn-
noise equivalent reflectivity of the remote sensor band; Lo.t
RsNRo,2
reference pupil radiance of the remote sensor band, in watts per square meter steradian micrometer [W/(m2·sr·μm)];
apparent reflectance value of the remote sensor band 1 corresponding to Lo,; signal-to-noise ratio of the remote sensor band 1 at the reference entrance pupil radiance L. , level; sun-earth distance factor (astronomical unit);
top of atmosphere solar irradiance of the remote sensor band 1, in watts per square meter micrometer [W/(m·μm)] solar zenith angle, in degrees (°). d)
repeat steps b) to c) to complete the radiometric resolution evaluation of all bands of the remote sensor. Calculate the radiometric resolution of each band of the remote sensor according to steps a) to d) at least once a year. Use a two-dimensional line graph e)
(the horizontal axis is the evaluation period, the vertical axis is the radiometric resolution) to display the characteristics and trends of the radiometric resolution of each band of the remote sensor over time (for multispectral remote sensors, the radiometric resolution change curves of different bands can be displayed in the same line graph).
5.3 Dynamic range
Evaluate based on the remote sensor observation image data that has only been processed with relative radiometric correction. The specific steps are as follows: a) According to the GSD of the remote sensor, select a test area (usually a specific radiation calibration field, verification field) or a ground target (suitable for remote sensors with high spatial resolution, the layout requirements of the ground target refer to Appendix B) t,, t2,...., tk (K≥4) with flat terrain, good surface uniformity and different reflectivities, and meet the following requirements: 1) There is at least one high-reflectivity test area or target where the remote sensor response output reaches saturation; 2) There is at least one high-reflectivity test area or target where the remote sensor response output is within the dynamic range and has not reached saturation, and at least one low-reflectivity test area or target; 3) The size of the test area or target should be as large as possible, and the corresponding uniform area in the image should be at least larger than 5 pixels × 5
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.