Orthogonal Locally Discriminant Projection for Classification of Plant Leaf Diseases

Looking for fast, automatic, less expensive and accurate method to detect plant diseases is of great realistic significance. By using the symptoms of the plant disease leaves, a supervised orthogonal nonlinear dimensionality reduction algorithm, named orthogonal locally discriminant projection (OLDP), is presented for plant disease recognition in this paper. The proposed algorithm aims to find a projecting matrix by pulling the data points in the same class as close as possible, while pushing the data points in different classes as far as possible. The highlights of OLDP include (1) it takes both of the local information and the class information of the data into account, (2) it considers the effect of the noisy points and outliers, (3) it is supervised and orthogonal. The experimental results on real maize disease leaf images demonstrate that the proposed method is effective and feasible for the detection of plant leaf diseases.