Creating a panoramic field image using multi-spectral stereovision system

  • Authors:
  • M. Kise;Q. Zhang

  • Affiliations:
  • United States Department of Agriculture, Agricultural Research Service Russell Research Center, 950 College Station Road, Athens, GA 30605, USA;Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign, 1304 W. Pennsylvania Ave., Urbana, IL 61801, USA

  • Venue:
  • Computers and Electronics in Agriculture
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Satellite-based remote sensing presents a broad view of field scenes while ground-based remote sensing offers detailed observation of the crop to support precision farming decision-making. If it is possible to combine the advantages of both satellite-based and ground-based sensing, the resulting imagery could present both the broad view of a field scene and detailed observation of crop growth, which offers a very informative means for delivering crop production information. This paper describes a fundamental investigation on creating multi-spectral 3D panoramic field imagery by seamlessly integrating ground-based multi-spectral images of the field. A vehicle-mounted stereovision sensor composed of two multi-spectral cameras collects six-spectral channel images to form the multi-spectral 3D field imagery. The collected stereo image stream is then synthesized progressively to create a single seamless field image. The resulting panoramic imagery forms a 3D virtual field scene capable of providing more informative interpretation of crop/field status than a 2D plane image. Field test results proved that the assembled multi-spectral imagery was capable of providing a seamless 3D panoramic view of a soybean field with a broad view for assessing crop growth status with sufficient detail of individual crop plants.