Automated Cover Estimation Using Ground Digital Images

Qiming Zhou*, Marc Robson

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingChapter

Abstract

This paper reports the development of a method to automate classification of digital images acquired from digital still-frame cameras. As the images to be processed record extensive details of the ground, it is difficult and inaccurate to classify such images without considering pattern structures of features such as bare soil, shrubs, grass and gravels. To overcome the problem, a statistical label relaxation algorithm was developed and implemented. The results from this method were encouraging showing high accuracy in the resulting classification which in turn provides essential statistics on cover types of testing images, with minimal human-machine interaction. Together with the development of field instruments, this classification method has made automated and accurate field sampling possible for rangeland vegetation investigation.
Original languageEnglish
Title of host publicationSpatial information science, technology and its applications
Subtitle of host publicationRS, GPS, GIS their integration and applications
EditorsDeren Li, Jianya Gong, Xiaoling Chen
Place of PublicationWuhan
PublisherWuhan Technical University of Surveying and Mapping Press
Pages745-751
Number of pages7
Edition1st
ISBN (Print)9787810306676, 7810306677
Publication statusPublished - Dec 1998

User-Defined Keywords

  • Label Relaxation
  • Classification
  • Image Processing
  • Remote Sensing

Fingerprint

Dive into the research topics of 'Automated Cover Estimation Using Ground Digital Images'. Together they form a unique fingerprint.

Cite this