ISBN-13: 9786204726069 / Angielski / Miękka / 76 str.
Agricultural domain plays the vital role in our daily life. Due to that, it is important to clear that measures are taken to detect and mitigate any diseases on agricultural plants leaf. Plant leaf disease are major problem or factor to losses on crop in agricultural framing. In this proposed system we have approach multiple classification step to eliminate possibilities of disease. We have approach YOLOv3 with pressing and data augmentation for extracting leaf as an input images and extracted the leaf through resnet50 based model and also flask server to define which category of leaf disease is it. We trained plant village dataset. Mainly used of dataset name are Tomato, Pepper bell, and Potato images from plant village dataset. In our proposed system, the classification is performed in multiple stages to eliminate possibilities at every stage, hence providing better accuracy during predictions. And also gives better accuracy than the existing algorithms.