Fig. 4: A CNN-based learning architecture established for lettuce counting. | Horticulture Research

Fig. 4: A CNN-based learning architecture established for lettuce counting.

From: Combining computer vision and deep learning to enable ultra-scale aerial phenotyping and precision agriculture: A case study of lettuce production

Fig. 4

a The architecture of the trained CNN model, which generates a binary output representing the probability of whether a yellow bounding box contains a lettuce signal. b If the probability is close to 100%, it indicates that it is highly likely that the bounding box encloses a whole lettuce. c The training and validation accuracy and loss curves of the model

Back to article page