![]() For instance, the relative pose of its petiole and blade can change in different images. Although such characters are applicable in computer systems, accurate recognition of leaves still remains challenging: (1) a leaf can be imaged under arbitrary poses and its parts can articulate. Nowadays, with the rich development of information technology, it is desired, to have an automated plant identification system, so that a user can take picture of a plant in the field by a build-in camera of a mobile device and feed it to a pre-installed recognition program to find out the information of the possible species.Īmong the organs of a plant, such as flower, fruit, stem and bark, leaf is often used for plant identification since its features are more universal and persistent. However, plant identification is not an easy job, but requires specialized knowledge and in-depth training in botany and plant systematics. It is required by all walks of life, from professionals such as landscape architects, arborists, herbal doctors, conservationists, and biologists, to the general public like ecotourists, hikers, park visitors and nature lovers. Plant identification is not solely the job of botanists and plant ecologists. Experiments conducted on five leaf image datasets demonstrate that our algorithm significantly outperforms the state-of-the-art methods in terms of recognition accuracy, efficiency and storage requirement. The proposed counting-based shape descriptor is not only discriminative for classification but also computationally fast and storage cheap. Furthermore, we advocate that when comparing two leaf individuals it is better to “count” the number of certain shape patterns rather than to match the extracted shape features in a point-wise manner. In specifics, we propose a novel feature that captures global and local shape information independently so that they can be examined individually during classification. Different from existing studies which target at simple leaves, the proposed method can accurately recognize both simple and compound leaves. In this paper, we propose a new method for plant identification using shapes of their leaves. Nevertheless, it is not an easy job but requires specialized knowledge. Plant identification is required by all walks of life, from professionals to the general public.
0 Comments
Leave a Reply. |