Thesis
Identification of pruning branches in tall spindle apple trees for automated pruning
Washington State University
Master of Science (MS), Washington State University
2012
Handle:
https://hdl.handle.net/2376/102963
Abstract
Apple orchard architectures are evolving towards more simple, narrow, accessible and productive (SNAP) systems that are friendlier to automation and mechanization. These systems have potential to reduce production costs by improving yield and/or by reducing labor cost. Pruning is a labor intensive operation that constitutes a significant component of total production cost of apples. One of the SNAP architectures which lends better to mechanization is Tall Spindle fruiting wall system. This work focused on identification of pruning branches on apple trees in this Tall Spindle architecture. A time-of-flight-of-light-based three dimensional (3D) camera was used to obtain 3D information of apple trees in a commercial Tall Spindle orchard nearby Washington State University Irrigated Agriculture Research and Extension Center (IAREC), Prosser, WA. The camera provided 3D coordinates of the captured scene with reference to the camera image plane. This information was used to construct 3D skeletons of apple trees using a 3D medial axis thinning-based skeletonization algorithm. An algorithm to identify trunk and branches was introduced. Pruning branches were identified in the reconstructed trees using a simplified two step pruning rule; i) maintain specified branch spacing and ii) maintain specified branch length. Tests were conducted to determine the performance of the algorithm to identify all the branches and branches that need to be pruned out. Performance of the algorithm was compared against human pruning. The pruning accuracy of the algorithm compared to human pruner was 61%.The accuracy of the branch identification and pruning branches identification algorithms were 76% and 86% respectively. The pruning branch identification of the algorithm was closest to the pruning branch identification of the most consistent worker with 10 years of experience. The algorithm suggested to remove 17%±8.7% of branches whereas the most consistent worker removed 16%±7.3% of branches in average.
Metrics
56 File views/ downloads
64 Record Views
Details
- Title
- Identification of pruning branches in tall spindle apple trees for automated pruning
- Creators
- Bikram Adhikari
- Contributors
- Manoj Karkee (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Biological Systems Engineering, Department of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; [Pullman, Washington] :
- Identifiers
- 99900525124701842
- Language
- English
- Resource Type
- Thesis