Performance Analysis of Rice Plant Diseases Identification and Classification Methodology

Technological help can be used for improving the cultivation of critical crops for optimal production and quality. Automatic plant disease detection is an interesting study issue as it may be beneficial for the monitoring of vast agricultural fields and thus the automatic identification of disease by the symptoms in the various sections of plants. This work contributes an automated diagnosis of different rice-related diseases utilizing image processing, deep learning, machine learning, and methods for meta-heuristic optimization. These measures include picture dataset size, class numbers, preprocessing procedures, classification approaches, performance analysis, etc. Researches from the previous decade are extensively reviewed, including studies on numerous rice plant diseases, and an investigation of the key features is provided. The survey provides insights into the various approaches used to identify disease in rice plants. Different attributes evaluated for the study include the kind of segmentation, dividing technology, extracted features, author name, dataset size and year of publication, disease category, techniques, accuracy of detection as well as classification and constraints. Furthermore, a model using a hybrid deep learning technique is proposed to identify diseases in rice plant such as rice blast, brown spots, leaf smut, tungsten and sheath.