A Survey on Plant Classification Based on Multi Organ Features Using HGO-CNN
- February 8, 2019
- Posted by: RSIS
- Category: Computer Science and Engineering
International Journal of Research and Scientific Innovation (IJRSI) | Volume VI, Issue I, January 2019 | ISSN 2321–2705
G.Kousalya1, Dr. P.D. Sheba Kezia Malarchelvi2
M.E. Student1, Professor & HOD 2,
Department of Computer Science and Engineering, JJ College of Engineering and Technology, Trichy, Tamil Nadu, India
Abstract: – A good understanding of plants is essential to help in identifying new or rare plant species in order to improve the drug industry, balance the ecosystem as well as the agricultural productivity. Categorizations of plants are still remains a tedious task due to limited knowledge and information of world’s plant families. Due to the intra or interspecies diversity of plants in nature, some species are difficult or impossible to differentiate from one another using only the leaf organ. Median Filtering is a noise removal algorithm that simultaneously reduces noise and preserves edges of input image. Here first propose a HGO-CNN (Hybrid Generic Organ – Convolutional Neural Network) model to automatically learn the generic and organ features representation for plant categories, replacing the need of designing hand-crafted features as to previous approaches. After having both organ and generic features migrate its convolutional layers to learn the fusion features. Second, we propose a new framework of plant structural learning based on recurrent neural networks (RNN), namely the Plant-StructNet. After classification of plant, the specific features and uses of the plant will be analyzing through this project.
Keywords:-Multi-organ, Plant classification, deep learning, CNN-HGO.
Plants play the vital role that provides the food and oxygen to all species in the world. Understanding the plants is essential to help for identifying new or rare plant species in order to develop the drug industry, to balance the ecosystem and also the agricultural productivity, sustainability. All botanists use variations on leaf characteristics as a comparative tool for their study on plant i.e., leaf characteristic are available to be experiential, annual plants or year-round in evergreen perennials and examined throughout the year in deciduous.
To identify the plant for the botanist computer makes it possible and easier task. The majority of computer vision selects leaf to identify, as leaf characters have been predominantly used to clarify plants. Characters such as size, texture, shape and venation are the common feature that is generally used to define the leaves in different species. Other than that, due to the interspecies or intra diversity in nature of plants, it makes us difficult to differentiate the plant species from one another by using the leaf organ.