Deficiency Identification of Greenhouse Lettuce using Explainable AI


A.R.S.P. Rodrigo
D.D. Karunaratne
S. Fernando


green house lettuce
convolutional neural networks
explainable artificial intelligence
nutrient deficiency identification

Issue Date:

18 th February 2022



The Convolutional Neural Network (CNN) based solutions are used to identify the nutrient deficiencies of the crops based on the color variances of the leaves. However, one of the major problems in CNN based solutions is the lack of ability to explain the results obtained. This research is focused on overcoming this challenge by combining the results obtained from CNN with the TensorFlow Inference Engine to provide humanely understandable results for deficiency identification of crops. Therefore, a custom testbed is created to gather data/images and using those data/images YOLOv3 object detection model was trained to detect calcium, nitrogen, and magnesium nutrient deficiencies of greenhouse lettuce. The results demonstrate a mean average precision of 94.38% on training data and 75.53% on custom data. The trained weights were combined with the TensorFlow inference engine to provide explainable results using a local knowledge base of deficiencies.