Precision agriculture in cotton: Definition of the optimal imaging resolution required for purple nutsedge detection Tal Miller, Liraz Cohen, Eldar Peleg,

Slides:



Advertisements
Similar presentations
Remote Sensing GIS/Remote Sensing Workshop June 6, 2013.
Advertisements

Scaling Biomass Measurements for Examining MODIS Derived Vegetation Products Matthew C. Reeves and Maosheng Zhao Numerical Terradynamic Simulation Group.
Predicting and mapping biomass using remote sensing and GIS techniques; a case of sugarcane in Mumias Kenya Odhiambo J.O, Wayumba G, Inima A, Omuto C.T,
SKYE INSTRUMENTS LTD Llandrindod Wells, United Kingdom.
Application Of Remote Sensing & GIS for Effective Agricultural Management By Dr Jibanananda Roy Consultant, SkyMap Global.
Remote Sensing of Aphid-Induced Stress in Wheat BAE/SOIL Precision Agriculture Oklahoma State University Victor W. Slowik April 20, 2001.
Giant Kelp Canopy Cover and Biomass from High Resolution SPOT Imagery for the Santa Barbara Channel Kyle C Cavanaugh, David A Siegel, Brian P Kinlan, Dan.
Use of remote sensing on turfgrass Soil 4213 course presentation Xi Xiong April 18, 2003.
Verify the Effectiveness UAS-mounted Sensors for Crop and Livestock Production Management North Dakota State University Pulsar Operational Boundary, Inc.
Use of Multispectral Imagery for Variable Rate “Application-zone” Identification in Cotton Production Tim Sharp Beltwide Cotton Conference January 6-10,
Relationships Between NDVI and Plant Physical Measurements Beltwide Cotton Conference January 6-10, 2003 Tim Sharp.
1 Sensor technologies for Precision Agriculture Dr. Athanasios Gertsis PERROTIS COLLEGE ( American Farm School (AFS), Thessaloniki,
VEGETATION DATA Viviana Maggioni Dr. Jeffrey Walker.
DROUGHT MONITORING THROUGH THE USE OF MODIS SATELLITE Amy Anderson, Curt Johnson, Dave Prevedel, & Russ Reading.
Agricultural, Water, and Health-Related Satellite Products from NESDIS-STAR Felix Kogan NOAA/NESDIS Center for Satellite Applications and Research October.
Satellite Data Access – Giovanni, LAADS, and NEO Training Workshop in Partnership with BAAQMD Santa Clara, CA September 10 – 12, 2013 Applied Remote SEnsing.
GreenSeekerTM Variable Rate Applicator Equipment and Applications
Site-Specific Management Factors influencing plant growth Water Light Temperature Soil Compaction Drainage.
Dr. M. Ahsan Latif Department of Computer Science
Developing a temperature-light based spatial growth model for purple nutsedge The 2 nd International Conference on: Novel and Sustainable Weed Management.
NUE Workshop: Improving NUE using Crop Sensing, Waseca, MN
Contrasting Precision Ag Technology Between Different Crop Species By Dodi Wear.
A comparison of remotely sensed imagery with site-specific crop management data A comparison of remotely sensed imagery with site-specific crop management.
AVHRR-NDVI satellite data is supplied by the Climate and Water Institute from the Argentinean Agriculture Research Institute (INTA). The NDVI is a normalized.
West Hills College Farm of the Future. West Hills College Farm of the Future Precision Agriculture – Lesson 5 What is Precision Agriculture?? Managing.
Using Sonar and Digital Imagery To Estimate Crop Biomass Introduction Sonar: may be used to detect proximity and distance in machine vision (Senix 2003)
ARI Agricultural Research Institute Kromeriz Ltd. Fluorescence imaging - a new tool for weed sensing? Karel KLEM, Ladislav NEDBAL.
Operational Agriculture Monitoring System Using Remote Sensing Pei Zhiyuan Center for Remote Sensing Applacation, Ministry of Agriculture, China.
Lesson 7 Understanding Remote Sensing Technology.
Use of aerial imagery to detect N response in corn following alfalfa FR 5262 Matt Yost Stephen Palka.
Karnieli: Introduction to Remote Sensing
Giant Kelp Canopy Cover and Biomass from High Resolution Multispectral Imagery for the Santa Barbara Channel Kyle C Cavanaugh, David A Siegel, Brian P.
NDVI: What It Is and What It Measures Danielle Williams.
AJ Foster Soils 4213 Spring AJ Foster, Department of Plant and Soil Sciences, Oklahoma State University.
The Use of Red and Green Reflectance in the Calculation of NDVI for Wheat, Bermudagrass, and Corn Robert W. Mullen SOIL 4213 Robert W. Mullen SOIL 4213.
SATELLITE AND AERIAL IMAGE DATA, MOBILE COMPUTING, GIS, AND GPS FOR INTEGRATED CROP MANAGEMENT (ICM) Chuck O’Hara, Dan Reynolds, Roger King John Cartwright,
Variable-Rate N Fertilization of Wheat and Corn in the Mid-Atlantic Variable-Rate N Fertilization of Wheat and Corn in the Mid-Atlantic Wade Thomason,
14 ARM Science Team Meeting, Albuquerque, NM, March 21-26, 2004 Canada Centre for Remote Sensing - Centre canadien de télédétection Geomatics Canada Natural.
Development of a Small Remotely Piloted Vehicle for the Collection of Normalized Difference Vegetative Index Readings Dr. Randy R. Price, Goutam Nistala.
Development of Vegetation Indices as Economic Thresholds for Control of Defoliating Insects of Soybean James BoardVijay MakaRandy PriceDina KnightMatthew.
Generalized Algorithm for Variable Rate Nitrogen Application on Cereal Grains John B. Solie, Regents Professor Biosystems and Agri. Engineering Dept. William.
Precision Agriculture an Overview. Precision Agriculture? Human need Environment –Hypoxia –$750,000,000 (excess N flowing down the Mississippi river/yr)
Factors affecting soil sub-surface phase of purple nutsedge (Cyperus rotundus) development Tal Naamat 1,2, Hanan Eizenberg 1 and Baruch Rubin 2 1 Newe.
Agronomic Spatial Variability and Resolution Resolution for Sensing/Soil Sampling And Yield Measurements.
A Remote Sensing Approach for Estimating Regional Scale Surface Moisture Luke J. Marzen Associate Professor of Geography Auburn University Co-Director.
Updated Cover Type Map of Cloquet Forestry Center For Continuous Forest Inventory.
Estimating Cotton Defoliation with Remote Sensing Glen Ritchie 1 and Craig Bednarz 2 1 UGA Coastal Plain Experiment Station, Tifton, GA 2 Texas Tech, Lubbock,
REMOTE SENSING INDICATORS FOR CROP GROWTH MONITORING AT DIFFERENT SCALES Zongnan Li 1, 2 and Zhongxin Chen 1, 2* 1 Key Laboratory of Resources Remote Sensing.
Generalized Algorithm for Variable Rate Nitrogen Application on Cereal Grains John B. Solie, Regents Professor Biosystems and Agri. Engineering Dept. William.
Measuring Vegetation Health NDVI Analysis of East Sacramento 1.
Weekly NDVI Relationships to Height, Nodes and Productivity Index for Low, Medium, and High Cotton Productivity Zones T. Sharp, G. Evans and A. Salvador.
GIS: The Systematic Approach to Precise Farm Management Robert Biffle Precision Agriculture April,
Assessment on Phytoplankton Quantity in Coastal Area by Using Remote Sensing Data RI Songgun Marine Environment Monitoring and Forecasting Division State.
Farms, sensors and satellites. Using fertilisers Farming practice are changing Growing quality crops in good yields depends on many factors, including.
Mapping Variations in Crop Growth Using Satellite Data
Remote sensing data for detection of Rhizoctonia solani in sugar beets
Drone applications in Forestry APEC/APFL forum, February 2017
Mapping wheat growth in dryland fields in SE Wyoming using Landsat images Matthew Thoman.
Hui Wang1,2*, Anders K. Mortensen1 and René Gislum1
Precision Agriculture an Overview
Precision Agriculture
Sensing Resolution in Corn
Precision Agriculture an Overview
E.V. Lukina, K.W. Freeman,K.J. Wynn, W.E. Thomason, G.V. Johnson,
By Blake Balzan1, with Ramesh Sivanpillai PhD2
Remote Sensing Section 3.
Rice monitoring in Taiwan
Detection of emerging plants using computer vision
Precision Irrigation in Oklahoma
An introduction to Machine Learning (ML)
Presentation transcript:

Precision agriculture in cotton: Definition of the optimal imaging resolution required for purple nutsedge detection Tal Miller, Liraz Cohen, Eldar Peleg, Matan Gilad and Anat Stein Western Galilee Regional Highschool Hanan Eizenberg Department of Weed Research, Newe Ya’ar Research Center, Agricultural Research Organization (ARO) The 2 nd International Conference of Novel and Sustainable Weed Management in Arid and Semi-Arid Agro Ecosystems September 6 th -10 th, Santorini, Greece

Introduction Purple nutsedge is a troublesome weed, causing severe damage in cotton Weeds may compete on resources such as water, light, space, nutrients etc.

The effect of purple nutsedge infestation on cotton biomass Purple nutsedge biomass (g m -2 ) Cotton biomass (g m -2 )

Precision agriculture approach Precision agriculture, specifically, site specific weed management is a modern approach for reducing herbicide rates This could be achieved by spraying herbicides only on weed patches based on the detection of the spatial distribution of weeds (and not on the entire field)

50% savings

66% savings

Weeds are easily detected visually because they are green plants on brown soil Several indexes were developed for this purpose How can we detect weeds grown in the field?

Normalized Difference Vegetation index (NDVI) NIR = Reflection in Near Infra Red (770  m) Red = Reflection in Red (660  m) It was reported in the literature that NDVI is highly correlated to vegetative growth, nitrogen and chlorophyll levels

NDVI image by NASA

NDVI of wheat field pre planting savings 50%

Wheat field pre planting

When NIR channel is not available, NGRDI index may be used:

The main goal of this study was to determine the optimal resolution required for the detection of purple nutsedge in cotton Research objectives

Specific objectives were: Detecting purple nutsedge on bare soil (inter rows) using NGRDI index Defining the threshold resolution for purple nutsedge detection Research objectives (cont.) High resolution RGB image (0.05 x 0.05 m per pixel)

Hypothesis We hypothesize that NGRDI values, greater than bare soil NGRDI (~0.01) represent vegetative growth, in our case purple nutsedge infestation

Experiments were performed in a commercial cotton field in the Jesreel Valley, Northern Israel Aerial images were captured at the same day of data collection Image resolution was 0.05 x 0.05 m per pixel Twenty-five plots were randomly selected for data collection Materials & Methods

Materials & Methods (cont.) Weed coverage (%) was visually estimated Purple nutsedge shoots were counted Plot locations were marked using a differential GPS (dGPS – sub-meter) Data were imported into a Geographical Information Software (GIS) software for advance analysis

Results 10m

Data processing Originally pixel size was 0.05 x 0.05 m Computing values of RGB channels Reducing the resolution by increasing the size of the pixels Re-computing values for the merged pixels by using the average value 0.45 m 1.70 m

Imaging GIS Observations Fixing imageMarking location by dGPS Increasing pixel size Validation Determining threshold value NoYes Creating multi-layer map Choosing plots Color channels analysis Computing NGRDI index Is index value higher than ground value?

Relations between pixel size and NGRDI index NGRDI Pixel size (cm 2 )

Conclusions The threshold resolution for purple nutsedge detection from an aerial RGB image is 0.5 X 0.5 m per pixel (using NGRDI index) Although NIR imaging is separating better weed from soil, using RGB channels (NGRDI index) is much cheaper and available for weed detection Weed coverage that causes damage to cotton could be detected with a resolution of 0.5 x 0.5 m per pixel

Acknowledgment EWRS for supporting my trip Anat Stein for her assistance and motivation Research team in the Newe Ya’ar Research Center, Department of Weed Research Dr. Yafit Cohen, Sensing, Information and Mechanization Engineering, ARO Jimmie Ipen, field crops action, Alonim Shay Mey-tal, Agam LTD The school, for support and resources

Thank you for your attention!