Finding Spatial Units for Land Use Classification Based on Hierarchical Image Objects
نویسندگان
چکیده
Remote sensing in urban areas has been a challenger for quite some time due to their complexity and fragment with combination of man-made features and natural features. High-resolution satellite images and airborne laser altimetry data offered potential possibilities for feature extraction and spatial modelling in urban areas. Land use classification of urban areas may become possible by exploiting current high-resolution sensor data. The proposed approach incorporates spectral information from multi-spectral IKONOS images and height information from laser scanning data in hierarchical image segmentation based on semantically meaningful thresholds. By image segmentation, we obtain image objects at several levels with certain properties, which make it possible to include the spatial relations between adjacent image objects. Land cover classification and identification of image objects can be carried out mainly according to their properties. Land use classification at a higher level need to be inferred based on land cover objects and structural information at lower levels. We use Delaunay triangulation for deriving spatial relations between image objects and for structural analysis. Based on adjacency relationships of image objects, human settlements and other urban spaces are formed that create a base for land use identification as well as for structural analysis of urban areas. In this paper, the hierarchical image segmentation schema and the corresponding semantic-based thresholds are presented. To test the approach, we selected a site in a suburban area in Amsterdam, the Netherlands. The experiments show that hierarchically formed image objects are useful tools for image analysis and spatial modelling as compared to pixel-based approaches. Structural information can be derived based on hierarchical image objects, which plays an important role in land use classification in urban areas. We could cluster 704 image objects (buildings) into 82 spatial units for land use classification based on the measurement of shortest distance between adjacent buildings.
منابع مشابه
Hierarchical Image Object-based Structural Analysis toward Urban Land Use Classification Using High-resolution Imagery and Airborne Lidar Data
High-resolution remotely sensed imagery and airborne laser altimetry data offer exciting possibilities for feature extraction and spatial modelling in urban areas. In this study, hierarchical image objects have been generated by image segmentation based on IKONOS imagery and laser scanning data using semantically meaningful thresholds. Delaunay triangulation and morphological image analysis tec...
متن کاملMicro-classification of orchards and agricultural croplands by applying object based image analysis and fuzzy algorithms for estimating the area under cultivation
Remote sensing technology is one of the most efficient and innovative technologies for agricultural land use/cover mapping. In this regard, the object-based Image Analysis (OBIA) is known as a new method of satellite image processing which integrates spatial and spectral information for satellite image process. This approach make use of spectral, environmental, physical and geometrical characte...
متن کاملChange Detection Gamasiab River Margins in Kermanshah by Comparison Pixel Base and Object Orientd Algorithms
Introduction Land use reflects the interactive characteristics of humans and the environment and describes how human exploitation works for one or more targets on the ground. Land use is usually defined on the basis of human use of the land, with an emphasis on the functional role of land in economic activities. Land use, which is associated with human activity, is undergoing change over time....
متن کاملSpectral-spatial classification of hyperspectral images by combining hierarchical and marker-based Minimum Spanning Forest algorithms
Many researches have demonstrated that the spatial information can play an important role in the classification of hyperspectral imagery. This study proposes a modified spectral–spatial classification approach for improving the spectral–spatial classification of hyperspectral images. In the proposed method ten spatial/texture features, using mean, standard deviation, contrast, homogeneity, corr...
متن کاملRobust Method for E-Maximization and Hierarchical Clustering of Image Classification
We developed a new semi-supervised EM-like algorithm that is given the set of objects present in eachtraining image, but does not know which regions correspond to which objects. We have tested thealgorithm on a dataset of 860 hand-labeled color images using only color and texture features, and theresults show that our EM variant is able to break the symmetry in the initial solution. We compared...
متن کامل