Pattern Recognition in Thermal Images of Plants Pine Using Artificial Neural Networks

• Autor(a) principal: Colturato, Adimara Bentivoglio
• Data de Publicação: 2013
• Outros Autores: Gomes, Andre Benjamin, Pigatto, Daniel Fernando, Colturato, Danielle Bentivoglio, Roschildt Pinto, Alex Sandro [UNESP], Castelo Branco, Luiz Henrique, Furtado, Edson Luiz [UNESP], Jaquie Castelo Branco, Kalinka Regina Lucas, Iliadis, L., Papadopoulos, H., Jayne, C.
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/197436
• topic: Artificial neural networks; thermal images; Pine tree and UAVs

Pine is used primarily as a source of raw materials for the industries of lumber and laminated plates, resin, pulp and paper. Pine may be affected, from the nursery to adults, in plantations by pathogens such as fungi and/or pests. The aim of this work was to recognize patterns in images obtained from a thermal plants camera in pine. An Unmanned Aerial Vehicle with a thermal camera embedded was used to take video images of pine trees. The video was segmented in pictures and all the pictures were standardized to the same size 240 x 350px. The images were segmented and a two-layer neural network feed-forward and the Scaled Conjugate Gradient (SCG) algorithm were used. The results proved to be satisfactory, with most errors near zero.