Abstract:
Automated plant disease identification is an enduring research subject. Leaves are available for most of the season, and they have a flat (2d) surface that is why practically, it is physible to detect disease symptoms using image analysis. Data collection and pre-processing are the most significant and crucial stages to obtain the data that can be taken as accurate and appropriate for further processing. Machine learning techniques require a large amount of data for training. The present paper focuses on process standardization for the creation of an image dataset of Mung bean plant leaves and pre-processing steps to enhanced captured images. The diseases in leaves result in loss of economic and production status in the agricultural industry worldwide. The identification of disease in leaves using image processing reduces the reliance on the farmers for the safeguard of agricultural crops. In this paper, creation and segmentation process of Mung bean plant leaf is performed. Present dataset will be available to be used by researchers to save their time, efforts, and cost related to dataset creation. Segmentation of images will intensify the accuracy of the identification of various diseases.
