Multi-resolution segmentation for object-based classification and accuracy assessment of land use/land cover classificat
- PDF / 5,630,190 Bytes
- 13 Pages / 612 x 792 pts (letter) Page_size
- 71 Downloads / 188 Views
189
Photonirvachak
J. Indian Soc. Remote Sens. (June 2008) 36:189-201
RESEARCH ARTICLE
Multi-resolution Segmentation for Object-based Classification and Accuracy Assessment of Land Use/Land Cover Classification using Remotely Sensed Data Md. Rejaur Rahman . S.K. Saha
Received: 20 August 2007 / Accepted: 30 November 2007
Keywords Image segmentation . Object based classification . Pixel based classification . Land use / Land cover
Introduction In recent years, the significance of spatial data technologies, especially the application of remotely sensed data and geographic information systems (GIS) has greatly increased. In the classic image classification approach the unit is a single pixel. This approach utilizes spectral information of the pixels to classify the image, and the ability of this method to classify Md. Rejaur Rahman1 . S.K. Saha2 () 1 Dept. of Geography and Environmental Studies, University of Rajshahi, Rajshahi-6205, Bangladesh 2 Agriculture and Soils Division, Indian Institute of Remote Sensing (IIRS), Dehradun – 248001, India
images is limited when objects have similar spectral information. In most cases, information needed for image analysis and understanding is not represented in pixels, but in meaningful image objects and their mutual relations. Therefore, to partition images to sets of useful objects is a fundamental procedure for successful image analysis as part of image interpretation (Gorte, 1998; Baatz and Schape, 2000; Blaschke et al., 2000). The two most evident differences between pixel-based image analysis and object-oriented image analysis are: (i) in object-oriented image analysis, the basic processing units are image objects (segments), not single pixels; (ii) object-oriented image analysis uses soft classifiers that are based on fuzzy logic, not hard classifiers. In these respects, image segmentation is critical for subsequent image analysis, and even for further image understanding.
e-mail: [email protected]
123
190
Object-oriented approach and multi-resolution segmentation Object-oriented approach takes the form, texture and spectral information into account. Its classification phase starts with the crucial initial step of grouping neighboring pixels into meaningful areas, which can be handled in the later step of classification. Such segmentation and topology generation must be set according to the resolution and the scale of the expected objects. By this method, single pixels are not classified but homogenous image objects are extracted during a previous segmentation step. This segmentation can be done in multiple resolutions, thus allowing differentiation several levels of object categories. In remote sensing, the process of image segmentation is defined as: “....the search for homogenous regions in an image and later the classification of these regions” (Mather, 1999). The eCognition software offers a segmentation technique called Multiresolution Segmentation. This is a bottom-up region merging technique and is regarded as a region-based algorithm. It starts by considering ea
Data Loading...
