Conventional Data Augmentation Techniques for Plant Disease Detection and Classification Systems

Abstract

AbstractA tremendous improvement has been seen in the development of plant leaf disease detection and classification systems using convolutional neural networks from the last few years. The performance of these detection algorithms depends on a large amount and a wide variety of data. But, the collection of such an amount of data relies on so many parameters like weather condition, varying illumination, and non-occurrence of diseases in that specific time. Data augmentation techniques are very much needful to overcome this issue. In this work discussed, various data augmentation techniques that are applied to increase the black gram leaf disease dataset, which is further used to train the customized black gram plant leaf disease detection and classification systems.KeywordsAgricultureData augmentationDatasetPlant diseasesImage annotation