Identificação de indíviduos da espécie Cordia trichotoma (Vell.) Arráb. Ex Steud em dois remanescentes florestais a partir de dados geoespaciais
Forests provide us with goods and services continuously, contributing to a constant improvement in the quality of air, water, and, consequently, life, by supplying products of the highest quality, whether woody or not. In this sense, the species Cordia trichotoma (Vell.) Arráb. Ex Steud, popularly k...
| Autor principal: | Lindao, Andres Angel Leonardo |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2022
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/29299 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Forests provide us with goods and services continuously, contributing to a constant improvement in the quality of air, water, and, consequently, life, by supplying products of the highest quality, whether woody or not. In this sense, the species Cordia trichotoma (Vell.) Arráb. Ex Steud, popularly known as louro-pardo, is a characteristic element of these forests, particularly inserted within the Atlantic Forest, with importance from several points of view: ecological, economic and cultural. This species has a medium mechanical strength and a pleasant aspect, which makes it attractive for the construction of luxury furniture and civil construction, besides contributing to the recovery of degraded areas and the formation of new forests. Thus, there is a need to know the distribution and location of the species, and thus act in decision making that allows better planning in time and space. In this sense, the work aimed at the spatial identification of individuals of the louro-pardo, starting from high-resolution digital images generated by remotely piloted aircraft –RPA in two forest remnants. The identification of this species began with the manual classification through visual variables such as crown shape, color and texture, followed by the vectorization of the crowns of individuals. The visual identification was validated by field visits and comparison of the morphological characteristics of the species itself. In order to optimize this process, the images were automatically classified using the Maximum Likelihood Method and the results were compared to the manual classification. Finally georeferenced thematic maps were obtained with the spatial distribution of the individuals located in the two remnants.It is concluded that the methodology proved to be efficient, fast and safe, being very appropriate to work with high resolution images to realize the location of individuals of interest in the native forest, highlighting that the identification of the species is only possible when the totality of the selectedpattern is verified, this means that only when the 3 parameters of visual variables are established (canopy shape, color and texture) are the appropriate classification is assertive. |
|---|