Convolutional Neural Networks for Detection of Crop Diseases and Weed
Bhavansh Gupta, Saurabh Bomble, Onkar Gaikar, Shubham Chalekar, Sushma Vispute, K. Rajeswari
Abstract
Agriculture is around $10 trillion industry worldwide and has growth of more than 6% year on year. To enhance crop output and modernize the system, an automated and speedy disease detection mechanism is being developed. Another area where these technologies play an important role is the detection of Weed from farm using images of farmlands. The images will be used to locate weed among the crops and can be used along with aerial drones or robots for automated weed spraying operations. This paper compares proposes a system for detection of crop/plant diseases using images of plant and detection of weed using aerial images of farms. We compared MobileNetV2, InceptionV1, NASNetMobile for crop disease detection on dataset of 3 different crops having 12 disease classes. Of these NASNetMobile performed the best. Similarly we used ResNet50V2, MobileNetV2, InceptionV3, NASNetMobile for weed detection on soyabean plant image dataset and InceptionV3 had the best performance with 81.23% Validation Accuracy for Bounding Boxes. Using the result of this study as the basis we were able to develop an application that can be used for weed detection and crop disease detection using images.