Artificial neural network to estimate the basic density of cerrado wood

• Autor(a) principal: Silva, Jeferson Pereira Martins
• Data de Publicação: 2018
• Outros Autores: Cabacinha, Christian Dias, Assis, Adriana Leandra, Monteiro, Thiago Campos, Araújo Júnior, Carlos Alberto, Maia, Renato Dourado
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Pesquisa Florestal Brasileira (Online)
• Texto Completo: https://pfb.cnpf.embrapa.br/pfb/index.php/pfb/article/view/1656
• topic: Artificial intelligence; Wood density; Pilodyn; Inteligência artificial; Densidade da madeira; Pilodyn

The basic density of wood is an important property because it is related to the final product in the various uses that wood has. However, its determination demands time and costs, which justifies the use of more refined techniques for its estimation, such as artificial neural networks (ANN). The objective was to evaluate the use of artificial neural networks to estimate the basic density of species of cerrado stricto sensu with the use of Pilodyn and dendrometric variables. To compare the results obtained by ANN, regression models were adjusted. The best performing neural network was the one that used as input variables the depth of penetration (Pilodyn), species and DAP, presenting R² values of 0.72 and with root mean square error in percentage (RMSE%) of 5.69. The regression model presented R² value of 0.72 and RMSE% of 9.19. The artificial neural networks can estimate the basic wood density of species of cerrado stricto sensu studied in this study with satisfactory results.