Classificação e reconhecimento de frutos por processamento digital de imagem para previsão da produção

• Autor(a) principal: Pereira, Tiago Filipe da Mota
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.6/7849
• topic: Agricultura de Precisão; Deteção de Arestas; Espaços de Cor; Máquina de Suporte Vetorial; Pessegueiro; Previsão de Produção; Segmentação; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

The concept of Precision Agriculture is usually associated with the usage of high-end technology equipments (being hardware or software), to first evaluate or control the conditions of a determined portion of land, applying afterwards various factors of production accordingly, like seeds, fertilizers, phytopharmaceuticals, growing regulators, water, etc. The proposed theme of this dissertation aims to create an algorithm capable of analyse and process images to characterize trees, particularly peach trees, distinguish fruits and calculate it dimensions, like volume and weight. The recognition of peaches on their natural conditions, on trees, is complex and requires segmentation algorithms, also complex. The proposed algorithm applies image segmentation for extraction of characteristics such as colour and shape. These characteristics are then used to train a classification method through a support vector machine to improve the recognition rate of fruits, accomplishing results around 70%. The production prediction is obtained with the volume values already calculated for the recognised peaches, providing a prediction of 29.3 tons per hectare.