Author(s): Backes, André Richard ; Sá, Jarbas Joaci de Mesquita ; Kolb, Rosana Marta [UNESP]
Date: 2018
Persistent ID: http://hdl.handle.net/11449/177550
Origin: Oasisbr
Subject(s): Adaxial epidermis; Color; Gravitational system; Texture analysis; Adaxial epidermis; Adaxial epidermis; Color; Color; Gravitational system; Gravitational system; Texture analysis; Texture analysis
Made available in DSpace on 2018-12-11T17:25:57Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-01-01
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
The leaves are very informative plant organs. They are extensively used in plant anatomical studies focusing taxonomy. Their both inner and outer structures provide very discriminant features from vegetal species. In this study, we propose using images from adaxial epidermis for plant classification. The adaxial epidermis is a very variable region in a plant leaf cross-section. It differs in color, number of layers and presence/absence of hypodermis. To accomplish this task, we propose combining complexity analysis methods with a gravitational collapsing system to extract texture features from adaxial epidermis samples. Experimental results show that this combination of techniques surpasses traditional and state-of-the-art methods in both grayscale and color images of adaxial epidermis.
Universidade Federal de Uberlândia, Av. João Naves de Ávila 2121
Departamento de Engenharia de Computação, Universidade Federal do Ceará, Campus de Sobral, Rua Estanislau Frota, S/N, Centro
Departamento de Ciências Biológicas, Universidade Estadual Paulista UNESP, Av. Dom Antônio, 2100
Departamento de Ciências Biológicas, Universidade Estadual Paulista UNESP, Av. Dom Antônio, 2100
CNPq: #301558/2012-4
CNPq: #453298/2015-0
