A New Imaging Method for Automatic Phenotyping of the Primary Root in the Context of Tap-Rooted Legumes

Plant Phenotyping consists in characterising the plant morphometric parameters which allows correlating them to genome expression, modulated by plant environment. The observation of legumes roots and the manual evaluation of their characteristics are quite difficult and tedious. Imaging techniques are thus more and more used, but currently there is still a serious lack of fully automated, fast and complete methods. This paper proposes a fast and automated image processing technique based both on skeletonization technique and specific local image processing to determine very accurately plant roots traits, such as the identification of the primary root and its length. The results obtained will be used for the determination of the traits of other roots such as second order and lateral roots which will allow the dynamic characterization of plant root architecture.