article id 1386,
category
Research article
Highlights:
Different strategies for how to utilise machine vision to streamline the mounding head movements were developed and evaluated; The theoretical minimum rate of encountered obstacles while utilising machine vision in continuous mounding is presented, provided that an optimal continuous mounding has been performed; The needed minimum resolution of a machine vision system at work on a clearcut area was found.
Abstract |
Full text in HTML
|
Full text in PDF |
Author Info
To promote the growth and survival of regenerated forests, site preparation prior to tree planting on clearcuts is necessary. This is often performed with scarifiers, either through trenching or mounding. Mounding is generally considered better in a plant survival perspective but is inefficient on obstacle-rich clearcuts. By utilising machine vision through e.g. remote sensing methods, new strategies can enable efficient mound positioning. In this paper, three realistic strategies utilizing ideal clearcut object identification through machine vision have been developed that can be used for more efficient mounding. The results show that mounding efficiency can be significantly improved with a new mound positioning strategy that employs ideal object identification, especially on obstacle-rich clearcuts.
-
Lideskog,
Luleå University of Technology, Department of Engineering Sciences and Mathematics, Division of Product and Production Development, SE-971 87 Luleå, Sweden
E-mail:
hakan.lideskog@ltu.se
-
Karlberg,
Luleå University of Technology, Department of Engineering Sciences and Mathematics, Division of Product and Production Development, SE-971 87 Luleå, Sweden
E-mail:
magnus.karlberg@ltu.se