Tracking moving objects in image sequences

• Autor(a) principal: João Manuel Ribeiro da Silva Tavares
• Data de Publicação: 2010
• Tipo de documento: Livro
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/23565
• topic: Processamento de imagem, Engenharia mecânica; Image processing, Mechanical engineering

The computational movement analysis of objects in temporal image sequences is very changeling, given that it usually involves tasks for image enhancement, features segmentation, objects matching and registration, features tracking and motion analysis. Notwithstanding the difficulties, this computational analysis has a wide range of prominent applications; for instance, in engineering, medicine, virtual reality, biology and sports. Difficulties that frequently appear while tracking moving objects include the simultaneous tracking of manifold objects, objects temporary occlusion or definitively disappearance, variations of the viewpoints considered in the imaging acquisition or of the illumination conditions, or even nonrigid deformations or topological alterations that objects may undergo. In this presentation, we are going to introduce and discuss methods often considered in computational movement analysis of objects in image sequences; in particularly, for their segmentation, tracking and matching in images, and for estimation of the deformation involved among images.