Artificial neural networks that recognize and quantify relevant aspects of crop plants show great promise in image-based phenomics, but their training requires many annotated images. The acquisition of these images is comparatively simple, but their manual annotation is time-consuming. Realistic plant models, which can be annotated automatically, thus present an attractive alternative to real plant images for training purposes. Here we show how such models can be constructed and calibrated quickly, using maize and canola as case studies.
Download the paper from the publisher's site or here (PDF, 6.4 MB).
Download Movie 1 -- Calibration of a maize model here (MP4, 11.1 MB)
Download Movie 2 -- Calibration of a canola model here (MP4, 9.9 MB)
Download the maize model here (TGZ archive, 22.2 MB)
Download the canola model here (TGZ archive, 31.4 MB)
You will need
VLAB to run the models.
They were tested using VLAB verison 4.5.1-3582 on macOS 10.13 (High Sierra).
Watch related talks presented at the Seminarios del Insituto de Biologia (UNAM) here and Canola Innovation Day of Canola Week 2021 here.