Biblioteca Digital da Memória Científica do INPE (National Institute for Space Research) · 2003 · 42 citations · 1 references
ComputingVisualizationComputer VisionMachine VisionSemanticsClassificationImage ClassificationImage SegmentationObject RecognitionSemantic SegmentationVisual SemanticsMultiresolution SegmentationObject-based Image AnalysisSemantic Network AnalysisSemantic Network ModelingSemantic Network Structure
This work presents a high resolution image classification based on object oriented. The objects are derived by means of multiresolution segmentation. It allows a creation of different levels of segments supporting a hierarchy structure, generating spatial relations between objects and sub-objects. This hierarchy is the bedding for the semantic network. The knowledge is the basis of the semantic. The classification is based on fuzzy rules by the means of descriptors such as: form, texture and relations between objects and sub-objects. Different approaches of classification are assessed: semantic network, selective and context change classification. The tested site is an agricultural area in the city of Nova Esperança-Pr, wherein intended to map the riparian vegetation along of River Porecatú.
1