LEAF DISEASE DETECTION AND IDENTIFICATION USING HYBRID MULTICLASS SVM (HM-SVM)

Abstract

The agricultural industry is critical to long-term economic growth & food security. Crop diseases, on the other hand, can pose a significant threat to achieving this expansion. Early diagnosis and categorization of plant diseases are essential for a good outcome. This opened up a slew of new options for study in this field. A lot of effort is being done now to use neural networks to better identify and categorise plant diseases. A Hybrid Multiclass SVM (HM-SVM) model strategy towards leaf disease detection is presented in this research. To distinguish healthy and sick leaves, the researchers developed an HM-SVM for automated feature extraction and classification. Experiment results indicate that the proposed technique is capable of reaching high accuracy. Disease detection as well as identification in large fields using automated techniques is beneficial since it minimises people or farmers' labour, as well as time and money spent on observation and study of illness signs. This study explains how to use Hybrid multiclass SVM to identify and detect leaf diseases. Hybrid Multiclass SVM classifier is used to classify illnesses, and thus the detection accuracy is increased by maximising the information exploitation. We are applying image processing algorithms to classify diseases in this suggested system, and diagnosis may be done fast according to the disease. Crop productivity will be increased as a result of this strategy. Acquisition, image pre-processing, segmentation, and feature extraction are some of the procedures involved