accuracy assessment and reference data collection kamini yadav dr. russ congalton
TRANSCRIPT
Accuracy Assessment and Reference data
CollectionKamini Yadav
Dr. Russ Congalton
Review of the Reference data Excel spreadsheet
• Africa-Juno Received excel spread sheet o Had first conference call on April 16, 2015
• Australia-Pardhao Received excel spread sheet, GCE v.2 along with 1/3rd Ground data for Validationo Working on Scheduling conference call
• North America-Richard/Tekio Received excel spread sheeto Working on Scheduling conference call
• Europe-Aparna/Mutluo Received excel spread sheeto Working on Scheduling conference call
• South America-Chandra• Not received
Ground Data Sources Ground data (collected by our team including Murali)
Received shape files for Ethiopia, Tanzania, Malawi, Rawanda, Burundi
India (South India, Rajasthan) Ground data sourced from other projects (e.g., CORINE)
Curt Reynolds's field data from USDA/FAS 2015 corn map for South Africa and 2014 cotton / rice
map for Australia GDA Corp Ground data from literature
Authors will be contacted to access the reference data they used or the map they produced if possible
LUCAS Data (Received photos) 2012: 250,000 locations, 85,500 for validation 2009: 200,000 locations, 66,000 for validation 2006: 150,000 locations, 49,500 for validation Mixed pixels, Positional error and Independent data for validation
Reference data from other standard work (e.g., USDA CDL, Canada Agri)
Reference data from Literature Paper Title Journal Contact Data
1
Crop area mapping in West Africa using landscape stratification of MODIS time series and comparison with existing global land products
International Journal of Applied Earth Observation and Geoinformation, Volume 14, Issue 1, February 2012, Pages 83–93
[email protected], [email protected]
A ground data set collected during the 2009 and 2010 cropping seasons (744 GPS waypoints at the validation sites)
2
Generating plausible crop distribution maps for Sub-Saharan Africa using a spatially disaggregated data fusion and optimization approach
Agricultural Systems, Volume 99, Issues 2–3, February 2009, Pages 126–140
Crop distribution map of sub Saharan Africa
3Generating global crop distribution maps: From census to grid
Agricultural Systems, Volume 127, May 2014, Pages 53–60
Global Rainfed/Irrigated crop map
4
Disaggregating and mapping crop statistics using hyper temporal remote sensing
International Journal of Applied Earth Observation and Geoinformation, Volume 12, Issue 1, February 2010, Pages 36–46
[email protected], sunflower, Barley crop maps of southern Spain
5
Global rain-fed, irrigated, and paddy croplands (GRIPC) J.Meghan Salmon, Mark A.
Friedl , Steve Frolking, Dominik Wisser,Ellen M. Douglas
https://dl.dropboxusercontent.com/u/12683052/GRIPCmap.zip.
Irrigated/Rainfed Map
6
Finer resolution observation and monitoring of global land cover: first mapping results with Landsat TM and ETM+ data
International Journal of Remote SensingVolume 34, Issue 7, 2013
Landsat/MODIS Mapping 91,000 Training smaples; 38,000 Test samples
Reference data from Literature Paper Title Journal Contact Data
7 Data Mining, A Promising Tool for Large-Area Cropland Mapping
IEEE Journal of selected topics in applied earth observations and remote sensing, vol. 6, no. 5, october 2013
The field surveys were conducted in Mali during the 2009 and 2010 crop seasons (980 Way points)
8 GlobeLand30 (http://www.globallandcover.com/GLC30Download/index.aspx)
ISPRS Journal of Photogrammetry and Remote Sensing 103 (2015) 7–27
Jun Chen 154,587 pixel samples 2010 year
9 Mapping and discrimination of soyabean and corn crops using spectrotemporal profiles of vegetation indices
International Journal of Remote Sensing, 2015, Vol. 36, No. 7, 1809–1824,
Field data from 19 different croplands (state of Paraná, located in the South ofBrazil, between)
10 Improving Crop Area Estimation in West Africa Using Multiresolution Satellite Data
Proceedings of Global Geospatial Conference 2013
field survey conducted between May and July 2012.
11 Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using Support Vector Machines
ISPRS Journal of Photogrammetry and Remote Sensing 85 (2013) 102–119
Extensive field survey conducted in Four test sites in Middle Asia.
12 MODIS Collection 5 global land cover: Algorithm refinements and characterization of new datasets
Remote Sensing of Environment 114 (2010) 168–182
[email protected] 1860 Training sites globally
13 Cropland for sub-Saharan Africa: A synergistic approach using five land cover data sets
Calibrated synergy map for Africa (http://onlinelibrary.wiley.com/doi/10.1029/2010GL046213/abstract)
2553 samples distributed over Africa
Way Forward• Approach the respective authors or producers who have worked
on standard products mapping on small area to get either the ground data or final product along with their accuracy
• Generate independent reference data from existing cropland layers for 30x30m and 250x250m validation
• Work on VHRI to build independent reference data (trying to make use of temporal information to label the image segments)
• Compile the independent ground data coming from different sources (e.g. ICRISAT/GDA or others)
Obstacles
Challenge for 30 m data as demonstrated aboveWhat about MODIS pixels?
Field Form
8
REFERENCE DATA COLLECTION FORM
Basic Required Information:
Date of Collection (month, day, year): __/__/____
Local Weather Conditions _____________________________________________________
Observer Name: _____________________________________ Expert: Yes or No
Village, Sub-Country, Country__________________________________________________
GPS Coordinates of Center of Sampling Unit (WGS84?, DD) _________________________
Datum_______________
Latitude (X): _________ [Dec. Degrees] Longitude (Y): _________ [Dec. Degrees]
Estimated distance to nearest road: ______________ [m]
Accessibility Class/Collection Method: Offset Vantage Center
Approximate Size of Sampling Unit
Landsat TM (min. = 90mx90m): __________
MODIS (min. = 250mx250m): __________
250m
250m
Drawings of cropping pattern in the Training sample area (Surrounding area including location of roads; must annotate with north arrow direction)
(i) Sample area for Landsat TM (ii) Sample area for MODIS
i) Crop Type (Wheat, Corn, Rice, Barley, Soybeans, Pulses, Cotton, Potatoes, or Other):
____________
ii) Dominant Crop____________ Secondary Crop ____________
Were pictures acquired? Y or N
Associated Picture IDs
Photo ID
Direction Photo ID
Direction
North East
South Up
West Down
Any issues or anomalies at site:_________________________________________________
___________________________________________________________________________
___________________________________________________________________________
Additional Desired Information:
Irrigation: Irrigated or Rain fed
Cropping intensity: single, double, triple, continuous
90 m
90 m
250m
250m
Thanks