| Abstract Detail
Bioinformatic and Biometric Methods in Plant Morphology Bucksch, Alexander [1], Fleck, Stefan [2], Lynch, Jonathan [3], Weitz, Joshua S. [4]. Automatic plant morphology analysis in the field. Describing and quantifying the morphology of branching structures in plants has many applications: from modeling [1], trait estimation [2] to characterizing the interaction of plants with the environment [3]. Quantifying branching structure from 2D images and data obtained from emerging 3D technologies like terrestrial laser scanning is particularly challenging under field conditions. For example, occlusion, noise and undersampling are obstacles to deciphering the resulting 2D or 3D data and to extract the captured branching structure. Here, we describe innovations in automatic plant morphological analysis under field conditions in two examples: 1.) Tree canopies measured with terrestrial laser scanners and represented as 3D point clouds for which we analyze the point clouds by means of optimal network theory. 2.) Monocot and dicot roots represented in 2D digital photographs. In both examples, our objective is to characterize the branching structure, whether above- or below-ground. As we show, our approaches rely on (distinct) skeletonization algorithms [4] to represent the branching structure as a collection of one-dimensional curves. We emphasize the concepts used to recover and represent the branching structure from the given 2D and 3D data. We compare some results of estimated properties of the tree canopies to manually measured branching hierarchy and branch lengths. In the case of crop roots we adapted and extended the Shovelomics protocol [5] for manual root phenotyping in the field to capture treatment response and phenotypic differences between plant genotypes.
[1] Godin, C. (2000). Representing and encoding plant architecture: a review. Annals of forest science, 57(5):413-438.
[2] Schöler, F., & Steinhage, V. (2012). Towards an automated 3d reconstruction of plant architecture. In Applications of Graph Transformations with Industrial Relevance (pp. 51-64). Springer Berlin Heidelberg.
[3] Furbank, R. T. (2009). Plant phenomics: from gene to form and function. Functional Plant Biology, 36(10):5-6.
[4] Bucksch, A., Lindenberg, R., Menenti, M. (2010). Robust skeleton extraction from imperfect point clouds. The Visual Computer. 26(10):1283-1300.
[5] Trachsel, S., Kaeppler, S., Brown, K., and Lynch, J. (2011). Shovelomics: high throughput phenotyping of maize (Zea mays L.) root architecture in the field. Plant and Soil 341:75-87. Broader Impacts:
Log in to add this item to your schedule
Related Links:
Homepage of Alexander Bucksch
Lab page of Joshua Weitz
A Shovelomics video from Penn State
1 - Georgia Institute of Technology, School of Biology and School of Interactive Computing, 310 Ferst Drive, Cherry Emerson 237, Atlanta, GA, 30332, USA
2 - North-West German Forest Research Institute, Department Environmental Control, Goettingen, Germany
3 - Pennsylvania State University, DEPT OF HORTICULTURE, 102 Tyson Building, UNIVERSITY PARK, PA, 16802-4201, USA
4 - Georgia Institute of Technology, School of Biology and School of Physics, 310 Ferst Drive, Atlanta, GA, 30332, USA
Keywords:
none specified
Presentation Type: Symposium or Colloquium Presentation
Session: C2
Location: Prince of Wales/Riverside Hilton
Date: Monday, July 29th, 2013
Time: 2:00 PM
Number: C2003
Abstract ID:650
Candidate for Awards:None |
