Automatic plant identification via computer vision techniques has been greatly important for a number of professionals, such as environmental protectors, land managers, and foresters. In this chapter, we propose two learning-based leaf image recognition frameworks for automatic plant identification and conduct a comparative study between them with existing approaches. First, we propose to learn sparse representation for leaf image recognition. In order to model leaf images, we learn an over-complete dictionary for sparsely representing the training images of each leaf species. Each dictionary is learned using a set of descriptors extracted from the training images in such a way that each descriptor is represented by linear combination of a small number of dictionary atoms. Second, we also propose a general bag-of-words (BoW) model-based recognition system for leaf images, mainly used for comparison. We experimentally compare the two learning-based approaches and show unique characteristics of our sparse representation- based framework. As a result, efficient leaf recognition can be achieved on public leaf image dataset based on the two proposed methods. We also show that the proposed sparse representation-based framework can outperform our BoWbased one and state-of-the-art approaches, conducted on the same dataset.
CITATION STYLE
Hsiao, J. K., Kang, L. W., Chang, C. L., & Lin, C. Y. (2015). Learning-based leaf image recognition frameworks. Studies in Computational Intelligence, 591, 77–91. https://doi.org/10.1007/978-3-319-14654-6_5
Mendeley helps you to discover research relevant for your work.