Applying data mining techniques to the classification of successions forest stages mixed shade tolerant forest environments Técnicas de mineração de dados aplicadas à classificação do estádio sucessional da vegetação em áreas de floresta ombrófila mista

• Main Author: Sothe C.*
• Publication Date: 2016
• Other Authors: De Almeida C.M., Schimalski, Marcos Benedito, De Souza C.F.*, Liesenberg, Veraldo, De Souza J.B.*
• Language: eng
• Source: Repositório Institucional da Udesc
• dARK ID: ark:/33523/0013000003wjs
• Download full: https://repositorio.udesc.br/handle/UDESC/10777

This study aimed to evaluate open source softwares in order to classify secondary successional forest stages in Shade Tolerant Mixed Forest (FOM) environments in Southern Brazil. Two test sites were selected in the mountainous region of Santa Catarina State. We used scenes from the airborne system for acquisition and post-processing of images (SAAPI) with a spatial resolution of 0.39m. The dataset consists of orthorectified images containing three spectral bands in the visible range (i.e. 0.38-0.70pm), three spectral bands in the near infrared (i.e. 0.76- 0.78pm) and a digital surface model. The methodologies were developed using feature selection and decision tree algorithms in the following open source softwares: InterlMAGE, WEKA and QGIS. The results were satisfactory to classify successional stages of FOM as well as other classes of land use and land corer. The obtained Kappa indices ranged from 0.53 to 0.59 and the conditional Kappa varied from 0.29 to 0.83 for the successional forest stages. These results demonstrate the potential of these approaches for the extraction of information in high spatial resolution imagery as well as the possibility of providing subsidies for the implementation of public policies and monitoring of forest resources.