Plant Leaf Image Identification with Texture Features using Microstructure Descriptor

Plants are nature's most useful creations. They have become an important part of the survival of mankind. One of the world's biggest problems is that many plants are endangered by the forces of modernization. Plant leaves are used to prepare 50% of ayurvedic medicines, and many of these plant species are endangered. This latest research involves a new tool for identifying the texture of an image. A computational methodology based on GLCM is used to identify specific features within a digital image. A random dataset of leaves is created having different types of leaf image. Computation of the Euclidean distance is performed between the test image and each image in the database. A threshold is applied to the maximum distance obtained, and evaluation metrics such as precision, recall, and f-measure are computed used to check image retrieval performance. The result indicates that this system provides the values of precision, recall, and f-measure for GLCM is 94%, 64%, and 76% respectively.