Optimization Techniques for Detection and Recognition of Plant Leaf Diseases Using IOT and Image Segmentation

Abstract

On the earlier days, considerable effort has been put into building a database of plant diseases that is sufficient to develop effective methods for the detection and recognition of plant diseases. Since then, nutrition become an important factor throughout Asia, where the poorest people live. At the moment, the main purpose of agriculture is to produce and nourish the nation. Diagnosis and identification on a very large scale area through automated procedures is very helpful as it reduces labor, time and cost of diagnosing and diagnosing disease symptoms. The project reports on the novel's method of diagnosing diseases in rice leaf by employing an advanced algorithm such as the modified SqueezeNet to enhance accuracy to a greater stage. Incorporating the loss reduction process as one of the categorical cross entropy to enhance accurate level. An improved application is being developed where we can capture the image of the affected crop and propose the type of chemical fertilizer that will be used by the farmer to reduce crop losses due to improper fertilizer use.