Um sistema para o reconhecimento da feição edificação em imagem digital com agentes inteligentes.

• Autor(a) principal: Sousa, Pryscila de Jesus de
• Data de Publicação: 2011
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/10845
• topic: Computer Engineering; Image segmentation; Recognized features; Intelligent Agent; Engenharia da Computação; Segmentação de imagem digital; Reconhecimento de feições em imagem digital; Agentes Inteligentes; CNPQ::ENGENHARIAS

The purpose of this dissertation has been to create a new approach in order to recognition features of buildings in a digital image in an automatic way. Such recognition features would be interesting of government agencies that deals with urban planning for irregular human occupation control. The approach created has employed specialist software agents to proceed with the segmentation processing and features recognition in the digital images. The agents have been programmed to manipulate colored images with the Red, Green and Blue pattern (RGB). The creation of this new approach has been motivated by the fact of existing segmentation techniques and classification of images greatly depend on its users. In other words, with the approach discussed it has been intended to allow less technical users to interact with a classifier system, without requiring deep knowledge of digital image processing. A prototype tool has been developed to test this approach, which employs in an unusual way, intelligent agents, with tests done in digital orthophotos of the city of Angra dos Reis (RJ).