Feature extraction and XML representation of plant leaf for image retrieval

Abstract

Leaf recognition and retrieval plays an important role in plant recognition and retrieval and its key issue lies in whether selected features are stable and have good ability to discriminate different kinds of leaves. From the view of plant leaf morphology, domain-related visual features and semantic features of plant leaf are analyzed and extracted first. Then these features are translated into a hierarchy that is easily represented by XML. On such a basis, the leaf image retrieval system proposed in this paper provides two types of retrieval methods, which could give better precision and flexibility. Experiment results prove the effectiveness of selected features and performance superiority of the leaf image retrieval system. © Springer-Verlag Berlin Heidelberg 2006.