Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6866377 | Neurocomputing | 2014 | 12 Pages |
Abstract
Uniform multi-angle images were used to study the spatial distribution of plant roots. First, a method of two-step ellipse detection, based on the Hough transformation, was designed to detect all ellipses from multi-angle skeleton images. Then the original locations of root points were deduced using the ellipse equation obtained, and the root length distribution ratios of different horizontal regions, and different vertical layers in the growth medium were calculated. Finally, the effectiveness of the proposed method was assessed according to an orthography projection method. Experimental results showed that the proposed method can detect a large number of overlapping and occluding ellipses whose points scattered in different images, and has the advantages of requiring no three-dimensional reconstruction and economical.
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Xiaohong Xu, Zhihui Wu, Yu Chen, Guanlin Huang, Tonglin Zhu,