Track 1-03

Description

Three dimensional (3D) plant reconstructions, extended to four dimensions with the use of time series and accompanied by visual modelling, is being used for a number of purposes including the estimation of biovolume and as the basis for functional structural plant modelling (FSPM). This has been successfully applied to crop species such as cotton (Paproki et al. 2012). Measuring the growth pattern and arrangement of a pasture sward is a difficult task but can be used as an indirect measure of other variables of interest, such as growth rate, light interception, nutritional quality, herbivore intake, etc. (Laca and Lemaire 2000). Digital representation of individual plants in three dimensions is one way to determine sward structure. The High Resolution Plant Phenomics Centre (HRPPC) has developed PlantScan™ which combines robotics, image analysis and computing advances, to accelerate and automate the measurement of plant growth characteristics and allow discrimination of differences between individual plants within species. Image silhouettes and LiDAR (Light Detection And Ranging) are used and combined to digitise plant architecture in three dimensions with a high level of detail. Colour information, extracted from multispectral sensors, and thermal imaging from infra-red (IR) cameras are then overlaid on these 3D plant representations, thus providing a tool to link plant structure to plant function. Successful reconstructions using data collected by PlantScan™ in controlled conditions, have been conducted for a range of grasses such as wheat (Triticum aestivum), rice (Oryza sativa), corn (Zea mays) and broadleaf species such as canola (Brassica napus), cotton (Gossypium hirsutum) and tobacco (Nicotiana tabacum). This suggests that modelling the sward structure of grass and legume pasture species should be equally achievable. This study explores the use of PlantScanTM to reconstruct 3D images of the important and common pasture legume, subterranean clover (Trifolium subterraneum) with a view to analysing their 3D structure in-silico.

Share

COinS
 

Three Dimensional (3D) Reconstruction of Subterranean Clover

Three dimensional (3D) plant reconstructions, extended to four dimensions with the use of time series and accompanied by visual modelling, is being used for a number of purposes including the estimation of biovolume and as the basis for functional structural plant modelling (FSPM). This has been successfully applied to crop species such as cotton (Paproki et al. 2012). Measuring the growth pattern and arrangement of a pasture sward is a difficult task but can be used as an indirect measure of other variables of interest, such as growth rate, light interception, nutritional quality, herbivore intake, etc. (Laca and Lemaire 2000). Digital representation of individual plants in three dimensions is one way to determine sward structure. The High Resolution Plant Phenomics Centre (HRPPC) has developed PlantScan™ which combines robotics, image analysis and computing advances, to accelerate and automate the measurement of plant growth characteristics and allow discrimination of differences between individual plants within species. Image silhouettes and LiDAR (Light Detection And Ranging) are used and combined to digitise plant architecture in three dimensions with a high level of detail. Colour information, extracted from multispectral sensors, and thermal imaging from infra-red (IR) cameras are then overlaid on these 3D plant representations, thus providing a tool to link plant structure to plant function. Successful reconstructions using data collected by PlantScan™ in controlled conditions, have been conducted for a range of grasses such as wheat (Triticum aestivum), rice (Oryza sativa), corn (Zea mays) and broadleaf species such as canola (Brassica napus), cotton (Gossypium hirsutum) and tobacco (Nicotiana tabacum). This suggests that modelling the sward structure of grass and legume pasture species should be equally achievable. This study explores the use of PlantScanTM to reconstruct 3D images of the important and common pasture legume, subterranean clover (Trifolium subterraneum) with a view to analysing their 3D structure in-silico.