Plant Species Identification using Leaf Image Retrieval: A Study

摘要

Human beings along with other living beings and their ecological system are completely inter-dependent. In the past few decades, the technological development has affected the environment more radically than ever before. It has posed grave threats to the natural resources including habitat loss and degradation, over-exploitation of resources and change in climatic condition. Most of plant species are on the verge of extinction. In the present circumstances, it is essential to conserve ecological system. Plant identification is a crucial step towards ecosystem diversity conservation. It is time consuming and requires lots of efforts, specialized knowledge and in-depth training. Recent technological advancement in the field of imaging, data analysis, and plant morphology has improved the decision such as yield prediction, crop management, veterinary diet, improving climate and many more. This has also made it possible to develop plant species identification system. In this paper, we present a comprehensive study about plant species identification methods based on various feature extraction methods, classification and other challenges. Furthermore, the performance of two widely used classifiers viz. Support Vector Machine (SVM) and Probabilistic Neural Network (PNN) on Flavia dataset is examined in terms of Precision, Recall and F-Score. It is observed that SVM performs better than PNN for plant species identification..