Advanced

Geometric quality assessment of multi-rotor unmanned aerial vehicle borne remote sensing products for precision agriculture

Wijesingha, Jayan LU (2016) In Student thesis series INES NGEM01 20161
Dept of Physical Geography and Ecosystem Science
Abstract
Sub-meter level spatial resolution remote sensing products are essential for Precision Agriculture (PA) applications. Recent development of Unmanned Aerial Vehicle (UAV) with imaging sensors provides opportunity to generate sub-meter level, timely, and cloud free remote sensing products. This study is a preliminarily assessment of a multi-rotor UAV with consumer grade optical camera and five spectral band multispectral camera for PA applications especially in geometric aspect. The UAV was flown over the agriculture area and images from both cameras were acquired and Digital Surface Models (DSM) and Ortho-Mosaics were derived from the collected image data. Geometric and visual quality of the derived products were assessed and limitations... (More)
Sub-meter level spatial resolution remote sensing products are essential for Precision Agriculture (PA) applications. Recent development of Unmanned Aerial Vehicle (UAV) with imaging sensors provides opportunity to generate sub-meter level, timely, and cloud free remote sensing products. This study is a preliminarily assessment of a multi-rotor UAV with consumer grade optical camera and five spectral band multispectral camera for PA applications especially in geometric aspect. The UAV was flown over the agriculture area and images from both cameras were acquired and Digital Surface Models (DSM) and Ortho-Mosaics were derived from the collected image data. Geometric and visual quality of the derived products were assessed and limitations were identified regarding
to PA applications. The optical camera images derived 2.1 cm spatial resolution orthomosaic while multispectral ortho-mosaic from the UAV multipsectral images gave 5.6 cm spatial resolution. The horizontal geometric accuracies of the optical camera product and multispectral camera product were 2 pixels and less than one pixel respectively. Relative average elevation difference of agriculture crop area and non-crop area were 0.27 m and 0.14 m in derived DSM from optical images and multispectral images respectively. Blurriness of the UAV-borne images was identified as a limitation of the UAV remote sensing exercise and UAV motion blur, cloud shadow, and wind were noted as possible causes for the blurriness in this study. (Less)
Popular Abstract
Miniaturization of the electronic devices opens doors for remote sensing using Unmanned Aerial Vehicle (UAV) with small imaging sensors. Cloud free, very high spatial resolution, real time remote sensing products can be obtained from the UAV remote sensing. Vegetation monitoring is the most applied remote sensing application using UAV. Precision Agriculture (PA) is a new level of farming strategy which used remote sensing as a main tool. UAV remote sensing provides the opportunity to collect accurate, timely data over agriculture field s to conduct PA applications. This study focus on geometric accuracy of the UAV remote sensing products from two different cameras that attached to the UAV for the PA. Moreover, remote sensing product... (More)
Miniaturization of the electronic devices opens doors for remote sensing using Unmanned Aerial Vehicle (UAV) with small imaging sensors. Cloud free, very high spatial resolution, real time remote sensing products can be obtained from the UAV remote sensing. Vegetation monitoring is the most applied remote sensing application using UAV. Precision Agriculture (PA) is a new level of farming strategy which used remote sensing as a main tool. UAV remote sensing provides the opportunity to collect accurate, timely data over agriculture field s to conduct PA applications. This study focus on geometric accuracy of the UAV remote sensing products from two different cameras that attached to the UAV for the PA. Moreover, remote sensing product generation form UAV mages and difficulties of the process were identified from the study. A multi-rotor UAV with a consumer grade optical camera and a five band multispectral camera were used in this study. UAV was flew over test agriculture fields and image were acquired from both cameras. Acquired images were tested for light condition and blurriness effects. Based on the quality level UAV images were processed to obtain Digital Surface Models (DSM) and Ortho-rectified Image Mosaics from each image set. Geometric accuracy assessment using Ground Control Points (GCPs) were performed with different number of GCP and different pattern GCP distribution. Vertical accuracy of the products were not evaluated but average relative height difference of the two DSMs were evaluated. The DSM with 4.2 cm resolution and the ortho-mosaic with 2.1 cm resolution were derived from optical camera images. Similarly, the DSM with 22.6 cm and 5-band ortho-mosaic with 5.6 cm resolutions were obtained from multispectral images. Relative average elevation difference of agriculture crop area and non-crop area were 0.27 m and 0.14 m in derived DSM. The horizontal geometric accuracies of the optical camera product and multispectral camera product were 2 pixels and less than one pixel respectively. According to the area of 0.07 square kilometres, 06 GCPs were enough to obtain required geometric accuracy for PA. Blurriness of the UAV-borne images was identified as a limitation of the UAV remote sensing exercise and UAV motion blur, cloud shadow, and wind were noted as possible causes for the blurriness in this study. (Less)
Please use this url to cite or link to this publication:
author
Wijesingha, Jayan LU
supervisor
organization
course
NGEM01 20161
year
type
H2 - Master's Degree (Two Years)
subject
keywords
OrthoMosaic, blurriness, geometric accuracy, UAV, precision agriculture (PA), Digital Surface Model (DSM), geomatics, Physical Geography and Ecosystems Analysis
publication/series
Student thesis series INES
report number
401
funder
Svenska Institutet
language
English
id
8893471
date added to LUP
2016-10-14 12:40:24
date last changed
2016-10-14 12:40:24
@misc{8893471,
  abstract     = {Sub-meter level spatial resolution remote sensing products are essential for Precision Agriculture (PA) applications. Recent development of Unmanned Aerial Vehicle (UAV) with imaging sensors provides opportunity to generate sub-meter level, timely, and cloud free remote sensing products. This study is a preliminarily assessment of a multi-rotor UAV with consumer grade optical camera and five spectral band multispectral camera for PA applications especially in geometric aspect. The UAV was flown over the agriculture area and images from both cameras were acquired and Digital Surface Models (DSM) and Ortho-Mosaics were derived from the collected image data. Geometric and visual quality of the derived products were assessed and limitations were identified regarding
to PA applications. The optical camera images derived 2.1 cm spatial resolution orthomosaic while multispectral ortho-mosaic from the UAV multipsectral images gave 5.6 cm spatial resolution. The horizontal geometric accuracies of the optical camera product and multispectral camera product were 2 pixels and less than one pixel respectively. Relative average elevation difference of agriculture crop area and non-crop area were 0.27 m and 0.14 m in derived DSM from optical images and multispectral images respectively. Blurriness of the UAV-borne images was identified as a limitation of the UAV remote sensing exercise and UAV motion blur, cloud shadow, and wind were noted as possible causes for the blurriness in this study.},
  author       = {Wijesingha, Jayan},
  keyword      = {OrthoMosaic,blurriness,geometric accuracy,UAV,precision agriculture (PA),Digital Surface Model (DSM),geomatics,Physical Geography and Ecosystems Analysis},
  language     = {eng},
  note         = {Student Paper},
  series       = {Student thesis series INES},
  title        = {Geometric quality assessment of multi-rotor unmanned aerial vehicle borne remote sensing products for precision agriculture},
  year         = {2016},
}