Abstract
India loses 35% of the annual crop yield due to plant diseases. Earlydetection of plant diseases remains difficult due to the lack of labinfrastructure and expertise. In this paper, we explore the possibility ofcomputer vision approaches for scalable and early plant disease detection. Thelack of availability of sufficiently large-scale non-lab data set remains amajor challenge for enabling vision based plant disease detection. Against thisbackground, we present PlantDoc: a dataset for visual plant disease detection.Our dataset contains 2,598 data points in total across 13 plant species and upto 17 classes of diseases, involving approximately 300 human hours of effort inannotating internet scraped images. To show the efficacy of our dataset, welearn 3 models for the task of plant disease classification. Our results showthat modelling using our dataset can increase the classification accuracy by upto 31%. We believe that our dataset can help reduce the entry barrier ofcomputer vision techniques in plant disease detection.