Skip to main content
Fig. 7 | Plant Methods

Fig. 7

From: High-throughput and automatic structural and developmental root phenotyping on Arabidopsis seedlings

Fig. 7

Performances in the estimation of static traits by comparison between predicted and expertized data. Traits investigated are: A primary root length, B total second-order root length, and C number of second-order roots. The three red points in A–C are outliers that were not considered for linear regression (in blue). The four images, D–G present the source image and its associated reconstructed RSA. Each corresponds to a point in graphs A–C. In D, the reconstruction process captured the architecture efficiently despite many ambiguities: numerous root crossings and a second-order root following the trajectory of the primary root. The three outliers E–G occurred in unexpected situations leading to significant reconstruction errors. In E an active bacterial strain is present on the path of the primary root, yielding a false positive of the primary root detection at an early stage and the interruption of the primary axis path. In F the root reaches the bottom of the petri dish. While this common situation generally does not impact the reconstruction, in this case, a dark structure at the bottom of the dish is interpreted as the subsequent path of the primary root. In G the primary root is infected with a bacterial strain. As the bacterial strain develops, the root exhibits a rapid increase in apparent thickness, interpreted as the emergence of many lateral roots

Back to article page