Calibration and Validation of Air-borne Multispectral Imaging ......ISO/TC 211 Geographic information/Geomatics 1 11 December 2019 Calibration and Validation of Air-borne Multispectral

ISO/TC 211 Geographic information/Geomatics 1

11 December 2019

Calibration and Validation of Air-borne

Multispectral Imaging Sensors：State of the art and Standardization Requirements

Presentation for the 49th Plenary of ISO/TC 211, Omiya, Japan

Professor Chun Liu

Tongji University，China

ISO/TC 211


11 December 2019

Background

State of the art

Standardization requirements

Work foundations

Outlook

1、

2、

3、

4、

5、

Contents


11 December 2019

Background1


11 December 2019

Status of geographic information industry and resources monitoring

5

2

6

11

1

2

3

4

5

6

0

2

4

6

8

10

12

1960-1980 1980-2000 2000-至今

typ

es

freq

uen

cy

Development trend of natural

resources monitoring

调查种类调查次数

Geographic information industry is developing rapidly and naturalresource monitoring needs are increasing

0

1000

2000

3000

4000

5000

6000

7000

2016 2017 2018

Ou

tpu

t valu

e/(

bil

lio

n）

Development trend of geographic

information industry

types frequency


11 December 2019

New demands

All-factor, all-coverage and all-weather natural resources monitoring

agricultural forestry rivers townarable land

High precision quantitative

inversion and fine management of

natural resources

Application situation


11 December 2019

Advantages of air-borne multispectral imaging sensors

CBERS

Connect space and ground observation

Cover space and attribute data

High efficiency

High flexible

Rich elements

High precision


11 December 2019

Application of air-borne multispectral imaging sensors

Water quality monitoring

Plant identification of vegetation

Crop stress status analysis

Historic building conservation

Support all-factor, all-coverage and all-weather natural resources monitoring


11 December 2019

Demands of calibration and validation

Geometric distortion

Vignetting effect Central wavelength shift

Calibration Validation

Complex filed radiation environment

Detection system and evaluation indexesare inconformity


11 December 2019

State of the art2


11 December 2019

Status of multispectral imaging sensors

➢ Foreign multispectral imaging sensors：• MicaSense、Tetracam、SlantRange、

MAPIR and PIXELTEQ (America)• Parrot(France)

➢ Domestic multispectral imaging sensors:• Tongji University• Chinese Academy of Sciences

Distribution of airborne multispectral imager research


11 December 2019

Status of calibration

⚫ Gas emission radiant lamp method(Curtiss,2002)；

⚫ Diffuse reflector calibration method(Hedman,2009)；

⚫ Monochromatic light collimation method.

⚫ Based on standard irradiance lamp method (James,1998);

⚫ Integrating sphere calibration method(Zhou,1988);

⚫ Spectral radiometer - parallel light calibration method.

⚫ Brown Model;⚫ BFGS Model;

⚫ Zhang's calibration method.

Indexes are confused and methods are not suitable for low altitude, light and small remote sensing sensors

Construct the calibration and validation scheme of air-borne multispectral imaging sensors

Spectral calibration

Radiometric calibration

Geometric calibration


11 December 2019

Status of calibration and validation filed

Dedicated to calibration and validation of space-borne sensorsand not suitable for low-altitude remote sensing sensors

⚫ White Sand Calibration Field, USA ;⚫ LaCrau tcalibration field, marseille,

France;⚫ Ellesite geometry and radiation

inspection field, Australia.

Establish calibration and validation filed for air-borne multispectral imaging sensors

International calibration

field

China’s calibration

field

⚫ Baotou, Songshan, Heihe, Dunhuang Qinghai Lake and other calibration fields.


11 December 2019

Standardization requirements 3


11 December 2019

Status of standard

There is no multispectral imaging sensor included in the optics section. It only covers frame type and linear array push-broom cameras and lack

of applicability in multiple remote sensing applications.


11 December 2019

Requirements

Promotes the construction of international standard for air-borne multispectral imaging sensors and fill in the gaps in this field

Calibration methodsFacilitiesValidation methods


11 December 2019

Requirements

Air-borneMultispectralSensor

CA_Air-borneMultispectralSensor

CA_CalibrationValidation

+ calibrationType: CA_CalibrationType

+ observtionType:CA_ObservationType

+ multispectralType: CA_MultispectralType

+ centralWavelength:Real

+ numberofWavelengths:Real+ bandWidth:Real

CA_MultispectralCalibrationGeometry

CA_MultispectralCalibrationRadiometry

CA_MultispectralLaboratoryFacility

CA_MultispectralValidation

CA_MultispectralCalibrationSpbectral

CA_MultispectralFieledFacility

<<CodeLIst>>

CA_CalibrationType

<<CodeLIst>>

CA_MultispectralType

<<CodeLIst>>

CA_ObservationType

Calibration methods

Calibration facilities

Validation methods


11 December 2019

Work foundations4


11 December 2019

Novel air-borne multispectral imaging sensor

Multispectral imaging sensor Air-borne multispectral remote

sensing integrated system

3

5

6

2

4

1 1.Position and pose module

2.Light sensitive

elements

3.Synchronization module

4.Shock absorption frame

5.Aerial camera lens

6.Narrow band filters


11 December 2019

Comprehensive calibration and validation field

High precision control network

TargetsEquipment storage area

Field environment

Support the calibration of multi - spectral, laser radar, video, synthetic aperture radar and other novel sensors.


11 December 2019

Spectral calibration

Spectral bandwidth is 10 nm; Center wavelength shift is less

than 1nm; Transmittance is better than 98%.

Discrete spectral curves

Gauss fitting

Spectral response curve

➢ Monochromatic light collimationmethod


11 December 2019

Radiometric calibration

Standard light source radiation response curve

Absolute calibration

Absolute calibration test image

Standard light source

Uncertainty is less than 4%; Dynamic response range of standard light source is 18bit; Correctable band is 350nm~2500nm.

➢ Integral sphere-target sensor model

Test image Dark current coefficients Uniformity coefficient


11 December 2019

Geometric calibration

Reprojection error is less than 0.10 pixels; Pixel-by-pixel nonlinear radiometric correction.

➢ Chessboard calibration method

Original image Corrected image

Feature point extraction Reprojection error


11 December 2019

Multi-band spectral registration

675 nm waveband 705 nm waveband

850 nm waveband Three-waveband false color image

➢ Multi-channel geometric registration methodbased on feature points

Registration success rate reaches 100%； Registration accuracy is less than 0.3 pixel.


11 December 2019

Relative radiometric calibration

Principle of relative radiometric correction

Standard white board

Ground objects image acquisition

Grayscale image of surface reflectance

Result

Effectively eliminate the impact of radiation in the field; Realize rapid and wide range field relative radiometric correction.


11 December 2019

Validation of geometric calibration results

Geometric checkpoints distribution Geometric targets Reprojection error

Reprojection error is less than 1 pixels; The coordinates of ground points can be accurately obtained.


11 December 2019

Validation of radiometric calibration results

Based on typical ground objects

CementGround Wheat

Broad bean

Rape flower

Rape flowerwavelength（nm）

radiance （W/(sr m2）

Groundwavelength（nm）


Broad beanwavelength（nm）


Wheat


wavelength（nm）

Cementwavelength（nm）


Standard reflecting plate

Standard reflecting platewavelength（nm）

radiance（W/(sr m2）Based on standard

reflectors

Reflectivity deviation is less than 1.5%


11 December 2019

Outlook5


11 December 2019

Research the generic calibration technologies of "optics, video, synthetic aperture radar and lidar".

Promotes the construction of flied and laboratory standardized calibration environment for novel remote sensing sensors.

SARLiDAR

Video cameraMultispectral camera

Corner reflector

Feature targets

Promotes the construction of novel calibration and validation field


11 December 2019

Outlook

◆ Research on the novel air-borne multispectral imaging

sensors calibration and validation methods.

◆ Establish the core performance evaluation system of

novel remote sensing observation.

◆ Realize the standard transformation of standardized

verification method and process.

◆ Fill the technical and standard gaps in the field.


11 December 2019

Thanks！

Documents

Calibration and Validation of Air-borne Multispectral Imaging ......ISO/TC 211 Geographic information/Geomatics 1 11 December 2019 Calibration and Validation of Air-borne Multispectral