Precision Viticulture Ampelos 2013

Post on 01-Dec-2014

909 views 2 download

Tags:

description

This slideshow was presented at the Ampelos 2013 International Symposium in Santorini. It's goal is to inform about recent developments in the field of Remote Sensing, that can be used as a supplement to vine grower's/wine maker's experience and knowledge, to aid him/her in achieving better results. The complete title of the study is: Advanced remote sensing techniques & high spatial and spectral resolution data for Precision Viticulture.

Transcript of Precision Viticulture Ampelos 2013

Advanced remote sensing techniques Advanced remote sensing techniques && high spatial and spectral resolutionhigh spatial and spectral resolution data data

for Precision Viticulturefor Precision Viticulture

National Technical University of AthensSchool of Rural and Surveying Engineering

Department of Topography

Authors:Karantzalos, Karakizi, Kandylakis, Oikonomou, Makris, Georgopoulos

Precision ViticulturePrecision Viticulture

2

Precision?

Estimate the within field variability of

various vine/grape.. ..must/wine quality properties

Precision ViticulturePrecision Viticulture

3

Estimate the within field variability of Vines

Grapes

canopy, vigor, foliar pigments (chlorophylls, carotenoids, anthoc.), water stress, health, etc.

Brix, Total acidity, pH, malic acid, polyphenols, color index, ripeness, etc.

Precision ViticulturePrecision Viticulture

4

Estimate the within field variability ofcanopy, vigor, foliar pigments (chlorophylls, carotenoids, anthoc.), water stress, health, etc.

Brix, Total acidity, pH, malic acid, polyphenols, color index, ripeness, etc.

Experience, Expert Organoleptic

Field/Lab Analytical Measurements

Earth Observation/ Remote Sensing

Vines

Grapes

Remote SensingRemote Sensing

5

A supplement to vine grower’s / wine maker’s skills, experience & knowledge

Remote SensingRemote Sensing

6

Spectral Analysis/ Remote Sensors sensitive to:

Optical Near Infrared Thermal Microwave etc

Remote SensingRemote Sensing

7

Spectral Signatures - Vegetation

8

During veraison: Field Work, Satellite Images

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

9

Based on advanced remote sensing techniques & high spatial and spectral resolution satellite data

Detect where the vineyards are Estimate the spectral difference of various vine varieties Calculate within field vine properties

- canopy, Vigor, foliar pigments Estimate grape (must/wine) properties - Brix, Total acidity, pH, malic acid, polyphenols, color index, ripeness

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Where are the vineyards?Where do we cultivate vines?

data pre-processing, image fusion, etc classification (spectral, geometric & texture)

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Where are the vineyards?Where do we cultivate vines?

data pre-processing, image fusion, etc classification (spectral, geometric & texture)

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Where are the vineyards?Where do we cultivate vines?

data pre-processing, image fusion, etc classification (spectral, geometric & texture)

Trapeza Megaplatanos

Quality Indices

Multispectral

DataFused Data

Multispectral

DataFused Data

Completeness 86% 86% 94% 96%

Correctness 89% 92% 81% 92%

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Where are the vineyards?Where do we cultivate vines?

data pre-processing, image fusion, etc classification (spectral, geometric & texture)

What are the spectral differences of each vine variety ?Can we detect or discriminate them remotely ?

detected vineyards supervised classification (spectral)

Authors /Year SensorBand Width &

Number of BandsSpatial

ResolutionVarieties

Lacar et al. (2001)

CASI400-900 nm

12 1 m

1. Cabernet Sauvignon 2. Syrah

Ferreiro-Armán et al. (2006)

CASI400-950 nm

1443 m

1.Cabernet Sauvignon 2. Merlot Noir

Ferreiro-Armán et al. (2007)

CASI-2400-950 nm

144 3 m

1. Cabernet Sauvignon 2. Merlot Noir

CASI-2407,8-942,2 nm

483 m

1. Cabernet Sauvignon 2. Merlot Noir 3. Cabernet Frank , per 2

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Apprx. 20 dif. varieties 300 samples

330 spectral bands 90.000 ground observ.

+ 23 million remote observ.

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

What are the spectral differences of each vine variety ?Can we detect or discriminate them remotely ?

detected vineyards supervised classification (spectral)

COMPLETENESS Ground Truth

Classification Results

Syrah Ι MerlotSauvignon

Blanc IISauvignon

Blanc I

Syrah Ι 82,49% 2,11% 14,88% 12,66%

Merlot 0,49% 96,97% 0,44% 0,00%

Sauvignon Blanc II

5,50% 0,92% 83,66% 0,70%

Sauvignon Blanc I

11,52% 0,00% 1,03% 86,64%

Overall Accuracy 85,21%

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

What are the spectral differences of each vine variety ?Can we detect or discriminate them remotely ?

Karakizi et al., 2013. Vineyard detection and vine variety discrimination from high resolution satellite data, European Conference on Precision Agriculture

COMPLETENESS Ground Truth

Classification ResultsCabernet

SauvignonSyrah Robola Merlot

Sauvignon Blanc

Cabernet Sauvignon 68,35% 20,30% 17,28% 0,76% 27,92%

Syrah 4,60% 45,79% 7,09% 3,55% 7,01%

Robola 16,59% 19,40% 67,27% 2,24% 15,97%

Merlot 0,05% 1,62% 1,12% 91,37% 2,78%

Sauvignon Blanc 10,39% 12,89% 7,22% 2,08% 46,27%

Overall Accuracy 63,59%

Karakizi et al., 2013. Vineyard detection and vine variety discrimination from high resolution satellite data, European Conference on Precision Agriculture

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

What are the spectral differences of each vine variety ?Can we detect or discriminate them remotely ?

Estimate Vine properties - Within Field Variability

Canopy, Vigor, Foliar pigments (chlorophyll, carotenoids, anthoc..), Water stress, Health, etc

Johnson et al., 2003. Mapping vineyard leaf area with multispectral satellite imagery. Computers and Electronics in Agriculture.

Haboudane et al. 2004. Hyperspectral Vegetation indices.. ..for predicting green LAI, Remote Sensing of Environment.

Zarco-Tejada et al., 2005. Assessing vineyard condition with hyperspectral indices, Remote Sensing of Environment.

Meggio et al. 2010. Grape quality assessment in vineyards.. ..using narrow-band physiological remote sensing indices. Remote Sensing of Environment.

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Estimate Vine properties - Within Field Variability

Canopy, Vigor, Foliar pigments (chlorophyll, carotenoids, anthoc..), Water stress, Health, etc

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Estimate Vine properties - Within Field Variability

Canopy, Vigor, Foliar pigments (chlorophyll, carotenoids, anthoc..), Water stress, Health, etc

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Estimate Vine properties - Within Field Variability

Canopy, Vigor, Foliar pigments (chlorophyll, carotenoids, anthoc..), Water stress, Health, etc

Estimate Grape (Must/Wine) properties

Brix, pH, Total Acidity, Malic Acid Polyphenols, Color Index Ripeness, etc.

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Estimate Grape (Must/Wine) properties

Brix, pH, Total Acidity, Malic Acid Polyphenols, Color Index Ripeness, etc.

Kandylakis et al., 2013. Evaluating spectral indices from WorldView-2 satellite data for selective harvesting in vineyards, European Conference on Precision Agriculture

GIS

GeographicInformation

System

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Remote SensingRemote Sensing for for Precision ViticulturePrecision Viticulture

Thank you !!!Thank you !!!

Authors:Karantzalos, Karakizi, Kandylakis, Oikonomou, Makris, Georgopoulos

National Technical University of AthensSchool of Rural and Surveying Engineering

Department of Topography