TY - GEN
T1 - Automated Plant and Leaf Separation
T2 - 2016 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2016
AU - Frolov, Kirill
AU - Fripp, Jurgen
AU - Nguyen, Chuong V.
AU - Furbank, Robert
AU - Bull, Geoff
AU - Kuffner, Peter
AU - Daily, Helen
AU - Sirault, Xavier
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/12/22
Y1 - 2016/12/22
N2 - In recent years, the use of imaging based, non-invasive, and non-destructive plant phenotyping platforms have become popular. The analysis of the imaging data acquired from these platforms is still challenging. Current, 2D methods are limited in the information available, while 3D methods are more challenging to analyze. Plants like wheat are particularly challenging due to their thin leaves which are often in contact or occluded. In this paper, we introduce a method to more reliably reconstruct the complete 3D mesh from plants with thin leaves. This includes a spline based method to automatically separate merged leaves and plants. A conveyor belt multiview imaging system was used to image 200 trays that contained 3-10 young wheat plants that were grown in a glasshouse. All meshes were reliably reconstructed, however, 42 trays exhibited leaf merging. The proposed spline method allowed the each plant in 35 of the 42 trays to be fully separated. Thus the proposed process provides a promising first step towards automating a higher throughput non-invasive quantitative monitoring of wheat phenotypic features.
AB - In recent years, the use of imaging based, non-invasive, and non-destructive plant phenotyping platforms have become popular. The analysis of the imaging data acquired from these platforms is still challenging. Current, 2D methods are limited in the information available, while 3D methods are more challenging to analyze. Plants like wheat are particularly challenging due to their thin leaves which are often in contact or occluded. In this paper, we introduce a method to more reliably reconstruct the complete 3D mesh from plants with thin leaves. This includes a spline based method to automatically separate merged leaves and plants. A conveyor belt multiview imaging system was used to image 200 trays that contained 3-10 young wheat plants that were grown in a glasshouse. All meshes were reliably reconstructed, however, 42 trays exhibited leaf merging. The proposed spline method allowed the each plant in 35 of the 42 trays to be fully separated. Thus the proposed process provides a promising first step towards automating a higher throughput non-invasive quantitative monitoring of wheat phenotypic features.
UR - http://www.scopus.com/inward/record.url?scp=85011106034&partnerID=8YFLogxK
U2 - 10.1109/DICTA.2016.7797011
DO - 10.1109/DICTA.2016.7797011
M3 - Conference contribution
T3 - 2016 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2016
BT - 2016 International Conference on Digital Image Computing
A2 - Liew, Alan Wee-Chung
A2 - Zhou, Jun
A2 - Gao, Yongsheng
A2 - Wang, Zhiyong
A2 - Fookes, Clinton
A2 - Lovell, Brian
A2 - Blumenstein, Michael
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 30 November 2016 through 2 December 2016
ER -