Intelligent deep transfer learning Based rice plant disease diagnosis and Classification framework for Precision agriculture

Abstract

As Indian population is increasing at a faster rate, agricultural productivity newlinealso needs to be rapidly increased. Rice is considered the essential food crop newlinein India. But the rice crop tends to be easily affected by disease causing newlineagents and resulted in decreased yield. Though various challenging issues newlinedegrade crop productivity like pests, climate changes, diseases, crop diseases newlineremain the main problem in rice cultivation. Most diseases are introduced newlineby/associated with bacterial or fungi and can affect the crop in almost all newlinestages from nursery to harvesting. Conventionally, human vision based newlineapproaches have been employed to detect leaf diseases. They require expert s newlineknowledge, laborious, and expensive process. In addition, the accurateness of newlinethe human vision based process is mainly based on the vision of the farmer or newlineexperts. For resolving the limitations of classical approaches, it is needed to newlinedesign automated Machine Learning (ML) based classifier models. Earlier newlineidentification of Rice Plant Diseases (RPD) enables to take preventive actions newlineand reduce the loss of productivity. For accomplishing better crop quality and newlinequantity, it is needed to handle the spreading of diseases. The recently newlinedeveloped Computer Vision (CV) and Artificial Intelligence (AI) techniques newlinecan be used for the detection of plant diseases at an earlier stage, thereby newlinereducing productivity loss and improving crop quality. The plant disease newlinedetection process involves different stages of operations such as image newlineacquisition, pre-processing, segmentation, feature extraction, and newlineclassification. newline