Learning sparse representation for leaf image recognition

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 paper, a novel leaf image recognition technique via sparse representation is proposed for automatic plant identification. In order to model leaf images, we learn an overcomplete 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. For each test leaf image, we calculate the correlation between the image and each learned dictionary of leaf species to achieve the identification of the leaf image. As a result, efficient leaf recognition can be achieved on public leaf dataset based on the proposed framework leading to a more compact and richer representation of leaf images compared to traditional clustering approaches. Moreover, our method is also adapted to newly added leaf species without retraining classifiers and suitable to be highly parallelized as well as integrated with any leaf image descriptors/features.