Plant Phenomics / 2021 / Article / Fig 2

Research Article

Automatic Fruit Morphology Phenome and Genetic Analysis: An Application in the Octoploid Strawberry

Figure 2

Data analysis workflow. The input is the segmented internal and external fruit images from workflow in Figure 1. External images are used for linear and multidimensional shape analysis through different standard and machine learning approaches, including deep learning. Internal images are used to estimate the color pattern of the internal fruit section. Additive and dominance genetic components of each of the extracted morphometric and color phenotypes are estimated using Bayesian Linear Modeling using either pedigree or DNA marker information. Code available at https://github.com/lauzingaretti/DeepAFS