How Germination Changes During Individual Seed RGB-Space Differentiation: The Case of Pinus sylvestris L. сv. Negorelskaya

Abstract

Abstract To watch the growth of 1200 P. sylvestris cv. Negorelskaya trees from seeds to young or even old stage is a big grant project. We want to make a «seed–culture» passport. Each individual seed (N = 1200) was weighed, and image acquisition via a flatbed scanner in the VIS wavelength region and seeded into an individual 120 cm 3 cell of a 40-cell container. On day 30, container-grown germination was evaluated according to the following dichotomous criterion: 1 – germinated (n 1 = 942), 0 – did not germinate (n 0 = 258), and 0-group and 1-group datasets were formed. The RGB space color of the individual seed epidermis between the 0- and the 1-group were compared via the Kolmogorov‒Smirnov criterion D. The lower individual weight of the seed in the 0-group compared with the 1-group was not accidental (p = 0.0045). Additionally, in the 0 group, the median values of R, G, and B brightness of pixels from individual seeds are not accidental (p = 0.0000381) compared with those of the 1 group. Therefore, in this experiment, seeds that reflected most of the light from the epidermis showed a lower germination when placed in the container.

Plant phenotyping relevance

個体種子を対象にスキャナ画像からRGB形質を抽出し、発芽との関連を評価する画像ベースの表現型取得が研究の中心である。

Code and data availability

The paper's data availability statement openly deposits all three paper-specific phenotyping assets in Mendeley Data: Dataset 1 (individual seed morphometric/weight data, N=1200), Dataset 2 (original VIS flatbed-scanner seed images, N=1200), and Dataset 3 (individual container germination data, N=1200). These directly供