Application of neural networks to image recognition of plant diseases

Digital image recognition of plant diseases could reduce the dependence of agricultural production on the professional and technical personnel in plant protection field and is conducive to the development of plant protection informatization. In order to find out a method to realize image recognition of plant diseases, four kinds of neural networks including backpropagation (BP) networks, radial basis function (RBF) neural networks, generalized regression networks (GRNNs) and probabilistic neural networks (PNNs) were used to distinguish wheat stripe rust from wheat leaf rust and to distinguish grape downy mildew from grape powdery mildew based on color features, shape features and texture features extracted from the disease images. The results showed that identification and diagnosis of the plant diseases could be effectively achieved using BP networks, RBF neural networks, GRNNs and PNNs based on image processing. For the two kinds of wheat diseases, the best prediction accuracy was 100% with the fitting accuracy equal to 100% while BP networks, GRNNs or PNNs were used, and the best prediction accuracy was 97.50% with the fitting accuracy equal to 100% while RBF neural networks were used. For the two kinds of grape diseases, the best prediction accuracy was 100% with the fitting accuracy equal to 100% while BP networks, GRNNs or PNNs were used, and the best prediction accuracy was 94.29% with the fitting accuracy equal to 100% while RBF neural networks were used.

I. Introduction

Wheat stripe rust caused by Puccinia striiformis f. sp. tritici and wheat leaf rust caused by P. recondita f. sp. tritici, are both important airborne wheat diseases in China. The occurrence of these two kinds of diseases in wheat seedling stage is of great significance for forecast and integrated management of the diseases. However, their symptoms are extremely similar in wheat seedling stage, and it is difficult to distinguish these two kinds of diseases in the practical production [1]. So this poses some difficulties for disease investigation and disease forecast. Grape downy mildew caused by Plasmopara uiticola and grape powdery mildew caused by Uncinula necator, are two kinds of devastating grape diseases in greenhouse and field [2]. It is also difficult to distinguish them from each other. Traditionally, identification and diagnosis of these plant diseases mainly depends on the real-time visual identification by the professional and technical personnel in field or pathogen identification in laboratory. These approaches require high professional knowledge and many professional and technical personnel. Some mistakes are often made if relevant personnel are lack of experience. Therefore, in order to obtain fast and accurate disease diagnosis methods and provide strong support for disease forecast and disease control, it is necessary to conduct the studies on identification and diagnosis of these diseases.