This paper proposes an accurate and dynamic method for diagnosing of crop diseases. This method adopts Bayesian networks to represent the relationships among the symptoms and crop diseases. This method has one main difference from the existing diagnosis methods - it does not use all the symptoms in the diagnosis, but purposively selects a subset of symptoms which are the most relevant to diagnosis; the active symptom selection is based on the concept of a Markov blanket in a Bayesian network. Theoretical analysis demonstrates that the proposed method can significantly enhance the performance of crop disease diagnosis.

Keywords: plant disease recognition, mathematical model, Bayesian network, machine learning, Markov blanket