CAN_EYE 1/11 The CAN_EYE software for processing Digital Hemispherical Photos F. Baret, M. Weiss.

CAN_EYE 1/11

The CAN_EYE software for processing Digital Hemispherical Photos

F. Baret, M. Weiss

CAN_EYE 2/11

How to derive LAI from gap fraction

• fCover= 1-Po(0)

• fAPAR(s)fIPAR(s)=1-Po(s)

•

)cos(/).()( LAIGePo

CAN_EYE 3/11

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

LAI

Po(

57;5

°)

)cos(/).()( LAIGePo

The LAI Po relationship

CAN_EYE 4/11

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5-2.5

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

LAI

LAI(

Po)

-LA

I(P

o+er

ror)

Uncertainties in LAI estimates

-0.08-0.04

-0.02

-0.01

0.08

0.04

0.02

0.01

Error on Po

CAN_EYE 5/11

Start Setup Imageselection

Preprocessing

Classification EndProcessing & reporting

CAN_EYE: suite of steps

CAN_EYE 6/11

Some results over Barrax

0

1

2

3

4

5

6

7

0 2 4 6 8

INRA-CSE

Un

iver

sity

of

Val

enci

a

Comparison of 2 independent processing

CAN_EYE 7/11

Comparison with other methods

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

PAI destructif

PAI v

rai C

an_e

ye

Maïs Tournesol Soja Blé

r = 0.917

Comparison with LAI2000 over BarraxBy B. Martinez

Comparison with destructive measurements by V. Desmarez (CESBIO)

CAN_EYE 8/11

The P57 photosUsing only the 57° direction allows to get better resolution

But no more access to fAPAR (except at tetas=57°!) and fCover

CAN_EYE 9/11

Evidence of finer resolution

Hemispherical photosP57 photos

CAN_EYE 10/11

Some results

CAN_EYE 11/11

Conclusion• Software available (for free) (www.avignon.inra.fr/can_eye)• Advantages:

– Quick and easy to use– Accounting for clumping in LAI– Provides estimates of fAPAR and fCover– Traceability of the processing– Some encouraging validation results

• Potential problems:– Importance of illumination conditions: diffuse for looking down-ward– Potential problems when small gaps (resolution)– Not yet full documentation and references!!

• Future enhancements– Inclusion of Chen’s/Leblanc/Fernandes clumping computation– Processing P57 photos– Slope correction– More validation

• Development of autonomous light sensors (dynamics)