Rice Leaves Disease Detection by Using TensorFlow

Abstract

This research work focused on the automatic detection method for image analysis on rice leaves under a wide range of natural conditions for further analysis. In image processing, the syndrome is an essential part of feature eradication and regulation. However, some of the threats are still flawed to predict the inflammation. To meet those threats, the expected algorithm focuses on an exact problem to predict the inflammation from early warning. Bacterial Leaf Blight and Brown Spot are a major bacterial and fungal inflammation respectively in rice (Oryza sativa) crops, it causes harvest loss and reduces the quality of the grain. Various hybrid techniques for image analysis and regulation algorithms were analyzed and an automatic detection method has been proposed for identifying the exact inflammation in rice leaves under different natural conditions. This system can classify the percentage of infected and healthy rice leaves by using image data. This system can make our work easy with big agro farms to identify infected plants. I hope that it will make our work more classified in less time.