Fig. 1
Download original image
Transfer learning flowchart. The DeepLab model is trained on 200,000 images to predict 21 categories. The learned parameters (in orange) are used to initialize the model training, and the last layer (in blue) is adapted to predict 3 categories (oilseed rape, other plants and soil). Our model input consists of 50 manually labelled images.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.