other information
drafter:Wen Jianguang, Peng Jingjing, You Dongqin, Liu Qiang, Tang Yong, Xiao Qing, Liu Qinhuo, Li Xin, Fan Wenjie, Ge Yong, Wu Hua, Wang Xinhong, Liu Zhaoyan
Drafting unit:Institute of Space Information Technology, Chinese Academy of Sciences, Beijing Normal University, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Peking University, Institute of Geographic Sciences and Natural Resources Research,
Focal point unit:National Remote Sensing Technology Standardization Technical Committee (SAC/TC 327)
Proposing unit:Chinese Academy of Sciences
Publishing department:State Administration for Market Regulation National Standardization Administration
Some standard content:
ICS07.040
National Standard of the People's Republic of China
GB/T40038—2021
Validation of vegetation index remote sensing products
products2021-04-30Release
State Administration for Market Regulation
National Standardization Administration
Implementation on 2021-11-01
Normative reference documents
Terms and definitions
Basic requirements
Test methods
Selection of test methods
Direct test method
Indirect test method
Test report
Cover information
Text information
Test report information summary
Appendix A (Normative Appendix) Ground bidirectional reflectivity factor measurement references
-rrKaeerKa-
GB/T40038—2021
This standard was drafted in accordance with the rules given in GB/T1.1-2009: 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). GB/T40038—2021
The drafting units of this standard are: Institute of Space Information Innovation, Chinese Academy of Sciences, Beijing Normal University, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Peking University, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. The main drafters of this standard are: Wen Jianguang, Peng Jingjing, You Dongqin, Liu Qiang, Tang Yong, Xiao Qing, Liu Qinhuo, Li Xin, Fan Wenjie, Ge Yong, Wu Hua, Wang Xinhong, and Liu Zhaoyan.
-rrKaeerKAca-
GB/T40038—2021
Vegetation index is a series of indices formed by the linear or nonlinear combination of two or more remote sensing sensor band data. It can be used to diagnose the growth status of vegetation, green vegetation vitality, and invert various vegetation parameters. Vegetation index has many specific definitions, such as the ratio vegetation index (RVI) and naturalized vegetation index (NDVI) of the red band and near-infrared band reflectance ratio combination, the soil adjusted vegetation index (SAVI) that reduces the soil background, and the enhanced vegetation index (EVI) that is sensitive to dense vegetation areas.
The reflectance of sensors with red and near-infrared bands is often used to calculate vegetation indices, and multiple global vegetation index products have been released based on this. Due to the limitations of data and algorithms, vegetation index products inevitably have errors, and they need to be verified for authenticity to obtain the accuracy and uncertainty of the product. In the past, the data, processes and evaluation standards of Shanding verification were different, and the accuracy of the products obtained was not comparable, which brought difficulties to users' quantitative application. Therefore, it is necessary to formulate scientific and standardized vegetation index remote sensing product authenticity inspection standards. -iiKaeerkAca
1 Scope
Vegetation index remote sensing product authenticity inspection
GB/T 40038—2021
This standard specifies the basic requirements, inspection methods and inspection reports for the authenticity inspection of vegetation index remote sensing products. This standard applies to the authenticity inspection of vegetation index remote sensing products at the bottom of the atmosphere. 2 Normative references
The following documents are indispensable for the application of this document. For all references with dates, the versions with dates apply to this document. For all references without dates, the latest versions (including all revised versions) apply to this document. GB/T36296—2018 Guidelines for authenticity verification of remote sensing products GB/T394682020 General method for authenticity verification of land quantitative remote sensing products 3 Terms and definitions
GB/T362962018 and (GB/T394682020 defined and the following terms and definitions apply to this document. 3.1
Bidirectional reflectance distribution function bidirectional reflectance distribution function: BRDF bidirectional reflectance distribution function
The ratio of the micro-increment of the reflected radiance in a specific direction on the surface of an object to the micro-increment of the incident irradiance in a specific direction, Note: the unit is per steradian (sr)).
GB/T362992018. Definition 3.6
Bidirectional reflectance factor bidirectional reflectance factor
The ratio of the reflected radiance of an object facing a specific direction to the reflected radiance of an imaginary diffuse reflector in that direction under the same irradiance conditions.
［GB/T36299—2018, definition 3.7]
Spectral reflectivity The ratio of the reflected radiation flux of an object to the incident radiation flux of a certain band of electromagnetic waves. 3.4
Spectral response function Spectralresponsefunction A parameter that captures the response rate of the remote sensor to the incident radiation at different wavelengths. 3.5
vegetation index
Vegetation index
A characteristic index that can reflect the growth status and distribution of green plants formed by linear or nonlinear combination of data from different spectral bands of remote sensing images.
［GB/T30115—2013, definition 3.11]1
-rKaeerkca-
GB/T40038—2021
4 Basic requirements
The inspection of vegetation index remote sensing products shall comply with the provisions of Chapter 7 of GB/T36296-2018: At the same time: a) The reference object shall belong to the same vegetation index as the vegetation index remote sensing product to be inspected, correspond to the same area in the same or similar period (vegetation index time change rate), and its data quality is known. b) The reference object shall include two typical areas with different vegetation coverage: 1) low vegetation coverage area, such as sparse vegetation coverage area; 2) high vegetation coverage area, such as dense vegetation coverage area. c) The reference object should be ground measurement data for authenticity verification of vegetation index remote sensing products. It is advisable to select reflectance and precision index collected by ground sensors with the same band and angle as the vegetation index remote sensing product to be tested. When there is no ground measurement data, the tested vegetation index remote sensing product with the same band and angle as the vegetation index remote sensing product to be tested can be used as the reference object. 5 Verification method
5.1 Verification method selection
According to the heterogeneity of lichens, the ground measurement field of view and the pixel scale of the vegetation index remote sensing product to be tested, select the authenticity verification method: a) When the image scale of the vegetation index remote sensing product to be tested does not exceed 1.5 times the ground measurement field of view, the indirect verification method should be used. 6) When the scale of the vegetation index remote sensing product image to be tested is 1.5 times the field of view of the ground measurement, if the area of the pixel to be tested is a homogeneous surface: the direct verification method should be used: if the area of the pixel to be tested is a non-homogeneous surface, the scale difference should be considered, and a multi-scale step-by-step verification method based on ground measurement and high-resolution remote sensing data should be used. C) When there is no ground measurement data in the test area, a cross-verification method based on vegetation index remote sensing products of known accuracy should be used: 5.2 Direct Verification Method
5.2.1 Ground Sampling Method of Reference Objects
The ground sampling method of reference objects is as follows:
a) Evaluate the surface heterogeneity of the sample area in accordance with the provisions of 4.2 of GB/T39468-2020. h)
The sample area selection should be spatially representative, and it is advisable to have at least 3×3 vegetation index remote sensing product images to be tested. e
The time difference of the remote sensing product of vegetation index to be tested is determined according to the time change rate of vegetation index. It should not exceed 5 degrees in general, and should not exceed 2 degrees for the surface with faster changes. d) Sampling should be carried out in accordance with the provisions of 4.3 of (13/T39468-2020): For homogeneous sample areas, random sampling or systematic sampling should be carried out in the area; for non-homogeneous sample areas, stratified sampling should be carried out in the sample area according to the vegetation growth and vegetation type, so that the poles in the layer are homogeneous, and local random sampling should be carried out in the layer to obtain the vegetation index of each layer. 5.2.2 Inspection process
The inspection process should comply with the provisions of 8.1 of GB/T36296-2018. The main operation process is shown in Figure 1. The specific steps are as follows: a)
Sample selection: Design the ground sampling method according to the provisions of 5.2.1 and determine the sample. b) Measurement of ground bidirectional reflectivity: The measurement method is shown in Appendix A.c)
Spatial matching: spatially match the ground sampling points with the pixels of the vegetation index remote sensing product to be tested. d)
Spectral conversion: according to the spectral response function of the corresponding band of the vegetation index remote sensing product to be tested, convert the ground-measured bidirectional reflectance factor into the band bidirectional reflectance factor corresponding to the vegetation index remote sensing product to be tested. Angle consistency judgment: judge whether the angle of the ground-measured bidirectional reflectance factor is consistent with that of the vegetation index remote sensing product to be tested. e
If they are consistent, they can be directly used for the next test; otherwise, the BRDF characteristics of the ground sample are used for angle normalization, and the ground-measured bidirectional reflectance factor is corrected to the bidirectional reflectance factor consistent with the angle of the vegetation index remote sensing product to be tested. Spatial consistency judgment: According to the pixel area of the vegetation index remote sensing product to be tested, judge whether the spatial degree of the bidirectional reflectivity factor measured on the ground is consistent. If consistent, directly use the next step of inspection: Otherwise, perform spatial scale conversion according to the provisions of 4.4.1 in (G3/T39468-2020) to obtain the bidirectional reflectivity factor consistent with the pixel scale of the vegetation index remote sensing product to be tested. Calculation of relative true value of vegetation index: The relative true value of vegetation index is calculated using the band reflectivity combination method defined by the vegetation index remote sensing product to be tested. If there are empirical coefficients in the vegetation index calculation process, they should be accurately calculated based on the characteristics of the measurement area. Accuracy evaluation: The accuracy of the vegetation index remote sensing product to be tested is quantitatively expressed according to the accuracy evaluation indicators specified in 6.1 in G13/T36296-2018.
Uncertainty analysis: Quantitatively express the uncertainty of the vegetation index remote sensing product to be tested according to the uncertainty evaluation index specified in 6.2 of (G13/T362962018).
Vegetation index remote sensing
Sample selection
Ground bidirectional reflectivity factor
Spatial matching
Spectrum conversion
Angle consistency?
Spatial consistency?
Vegetation index relative true value calculation
Accuracy evaluation
Uncertainty analysis
Ground sampling method design
Angle normalization
Spatial scale conversion
Direct inspection operation flow chart of vegetation index remote sensing products 1
rrKaeerKAca-
GB/T40038—2021
5.3 Indirect inspection method
5.3.1 Multi-scale step-by-step inspection method based on ground measurement and high-resolution remote sensing data The inspection process should comply with the provisions of 8.1 and 8.2.1 in GB3/T36296-2018. The main operation process is shown in Figure 2. The specific steps are as follows: a) Preparation of high-resolution reflectivity image data: The spatial scale of the high-resolution reflectivity image should be the same or similar to the bidirectional reflectivity factor measured on the ground, and the time should be the same or similar to the vegetation index remote sensing product to be inspected. b)
Sample selection: Design the ground sampling method according to the provisions of 5.2.1 to determine the sample, c
Measure the ground bidirectional reflectivity factor: The measurement method is shown in Appendix A. Use the bidirectional reflectance factor measured on the ground to verify the high-resolution image reflectance: According to the provisions of c), d) and c) in 5.2.2, d)
convert the bidirectional reflectance factor measured on the ground into a reference reflectance consistent with the high-resolution image, and evaluate the accuracy and uncertainty of the high-resolution reflectance. If the accuracy and uncertainty of the high-resolution reflectance meet the user's established chest value requirements, it can be used for the next test; otherwise, re-prepare the high-resolution reflectance image data and verify e
high-resolution image reflectance spectrum conversion: according to the spectral response function of the corresponding band of the vegetation index remote sensing product to be tested, convert the high-resolution image reflectance that meets the user's requirements into the band reflectance that matches the vegetation index remote sensing product to be tested. Angle consistency judgment: judge whether the high-resolution image reflectance is consistent with the vegetation index remote sensing product to be tested. If so, it can be directly used for the next test; otherwise, perform angle normalization according to the sample BRDF characteristics, and correct the high-resolution image reflectance to a reflectance consistent with the low-resolution vegetation index remote sensing product to be tested. g) Spatial scale conversion: According to the provisions of 4.4.1 of GB/T39468-2020, the verified high-resolution image reflectance is scaled to the spatial scale of the low-resolution vegetation index remote sensing product to be tested. h) Relative true value calculation of vegetation index: The relative true value of the low-resolution vegetation index is calculated using the band reflectance combination method defined by the vegetation index remote sensing product to be tested. If there are empirical coefficients in the vegetation index calculation process, they should be accurately assigned according to the characteristics of the measurement area.
Accuracy evaluation: The accuracy of the vegetation index remote sensing product to be tested is quantitatively expressed according to the accuracy evaluation index specified in 6.1 of GB/T36296-2018.
Uncertainty analysis: The uncertainty of the vegetation index remote sensing product to be tested is quantitatively expressed according to the uncertainty evaluation index specified in 6.2 of GB/T36296-2018.
-riKacerKAca-
Ground sampling method design
Direct verification method
Vegetation index remote sensing product
High resolution reflectance image number
|Data preparation
Sample selection
Ground bidirectional reflectivity factor
High-resolution reflectivity verification
Accuracy and uncertainty
Does it meet the requirements?
Spectral conversion
Angular consistency?
Spatial scale conversion
Vegetation index relative true value calculation
Accuracy evaluation
Uncertainty analysis
Angular normalization
Figure 2 Multi-scale step-by-step verification operation flow of vegetation index remote sensing products 5.3.2 Cross-verification method based on vegetation index remote sensing products with known accuracy GB/T 40038—2021
The inspection process should comply with the provisions of 8.2.1 of GB/T36296-2018. The main operation process is shown in Figure 3. The specific steps are as follows: a) Selection of reference vegetation index remote sensing products: It is advisable to select the inspected vegetation index remote sensing products with the same or similar time as the vegetation index remote sensing products to be inspected as reference objects.
Spatial matching: geometrically match the reference vegetation index remote sensing products with the vegetation index remote sensing products to be inspected. c
Spectral matching: geometrically match the reference vegetation index remote sensing products with the vegetation index remote sensing products to be inspected. 1) Consistency judgment: According to the reflectivity spectrum characteristics of the corresponding band of the vegetation index remote sensing product to be tested, judge whether the reflectivity spectrum of the corresponding band of the reference vegetation index remote sensing product is consistent. If consistent, the next step of judgment can be performed: if inconsistent, the spectrum should be converted according to the spectral response function of the vegetation index remote sensing product to be tested and the reference vegetation index remote sensing product. 2) Angle consistency judgment: According to the angle characteristics of the vegetation index remote sensing product to be tested, judge whether the reflectivity angle of the corresponding band of the reference vegetation index remote sensing product is consistent. If consistent, the next step of judgment can be performed; if inconsistent, the angle should be normalized according to the BRDF characteristics of the reflectivity of the corresponding band of the reference vegetation index remote sensing product, and the reference vegetation index remote sensing product should be corrected to a product with an angle consistent with the vegetation index remote sensing product to be tested. Spatial consistency demarcation: According to the pixel size of the vegetation index remote sensing product to be tested, determine whether the spatial scale of the vegetation index remote sensing product to be tested is consistent. If it is consistent, the reference vegetation index remote sensing product can be directly used as the relative truth value for verification; otherwise, the spatial scale conversion is performed according to the provisions of 4.4.1 of GB/T39468-2020 to obtain the reference vegetation index remote sensing product that is consistent with the pixel scale of the vegetation index remote sensing product to be tested as the relative truth value. Accuracy evaluation: The accuracy of the vegetation index remote sensing product to be tested is quantitatively expressed according to the accuracy evaluation index specified in 6.1 of GB/T36296-2018. Uncertainty analysis: The uncertainty of the vegetation index remote sensing product to be tested is quantitatively expressed according to the uncertainty evaluation index specified in 6.2 of GB/T36296-2018.
Vegetation index
Sensing product
Reference vegetation index sensing product
Spatial matching
Spectral consistency?
Angle consistency?
Spatial consistency?
Accuracy evaluation
Uncertainty analysis
6Inspection report
6.1 Cover information
Spectral conversion
Angle normalization
Spatial scale conversion
Cross-check operation process of vegetation index remote sensing products The cover of the inspection report should include the following information:6
KaeerkAca-
Inspection report number;
Inspection report name:
Inspection person in charge:
Inspection checker:
Inspection issuer
Inspection unit;
Inspection unit;
Inspection time.
Main text information
Overview of vegetation index remote sensing products to be inspected
Describe the vegetation index remote sensing products to be inspected, including the data source, spatiotemporal coverage, spatiotemporal resolution, projection method and applicability of the products; a)
b) Main algorithms of the products and the characteristics of the algorithms, etc. 6.2.2
Description of reference object
Describe the validation dataset used as the reference object, which should include: GB/T40038-2021
General information of the validation dataset, including name, spatiotemporal resolution, measurement method, measurement instrument, data format, etc.; description of the quality evaluation of the validation dataset;
Description of the applicability of the validation dataset
Verification method and process
Describe the adopted verification method and verification process, which should include: a)
Overview of the verification method;
Verification process;www.bzxz.net
Evaluation index of the verification result;
Record requirements of the verification process;
Archive of the verification result.
Authenticity verification conclusion
Describe the authenticity verification result, which should include: Overall evaluation of the authenticity verification result: Capture the overall accuracy of the verified vegetation index remote sensing product. a)
Evaluation of sub-item indicators: Describe the accuracy of the tested vegetation index remote sensing products in different regions: Describe the accuracy of the tested vegetation index remote sensing products in different vegetation types.
Analyze the uncertainty in the test process.
6.2.5 Additional information
Describe and describe the unusual issues in the authenticity test of vegetation index remote sensing products. 3. Brief form of inspection report information
For the preparation of the brief form of inspection report information for vegetation index remote sensing products, please refer to Appendix D of GB/T36296-2018. [
-rrKaeerKAca-
GB/T40038—2021
A,1 Measurement equipment
Appendix A
(Normative Appendix)
Measurement of ground bidirectional reflectivity factor
A.1.1 Imaging or non-imaging equipment with bands related to the calculation of vegetation index can be used to measure the bidirectional reflectivity factor of ground samples. The spectral range should be 380nm~1050nm. According to the definition of vegetation index, the vegetation index can be calculated. A.1.2 Multi-angle observation frame (optional): used for multi-angle measurement of ground sample bidirectional reflectivity factor Measurement time and environmental conditions
A.2.1 The measurement period should be 9:30~15:30 local time to ensure that the solar radiation is stable during the measurement time; if it is required to be close to the satellite data time, the interval between the measurement time and the satellite transit time should be less than 0.5h. A.2.2 The sample area should have good visibility conditions, with no near obstructions above an altitude of 10°, and no moving objects nearby during measurement. A.2.3 The sample should be representative.
Meteorological conditions at the time of measurement: ground visibility should be better than 5km; the sky should be clear or with few clouds; the wind force should be less than level 3. A.2.4
A.3 Other requirements
A.3.1 In order to reduce the impact of natural reflected light from the measurement personnel on the measured samples, the measurement personnel should wear dark clothing. A,3.2 During the measurement process, the measurement personnel and the measurement equipment should be vertical to the main plane of the sun, and the recorder and other members should stand behind the measurement personnel to avoid walking in the sample area
A.3.3 The shadow of the measurement equipment should have no impact on the sample. A.3.4 When measuring at a single angle, the normal line of the instrument field of view should be kept vertically downward. A.3.5 When measuring at multiple angles, at least the main solar plane and the vertical solar plane should be included. For crops with row structures, the vertical and vertical planes should also be included. For multi-angle measurements of vegetation, low vegetation should be selected. A.3.6 The measurement height should be determined based on the sensor field of view and the ground sample, and should be 1m higher than the upper surface of the ground sample. For crops with ridge structures, the instrument field of view should be at least 3 times the period of change of the vegetation row structure. 8
rKaeerkca-
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.