Use of normalized difference builtup index in automatically mapping urban areas from TM imagery. Use of probability entropy for the estimation and graphical representation of the accuracy of maximum likelihood classifications. Fuzzy ARTMAP supervised classification of multispectral remotelysensed images. When the objects in the scene become increasingly smaller relative to the resolution cell size, they may no longer be regarded as individual objects.
For example, cytochrome c is a protein found in almost all organisms and is used as a character (a feature defining a taxon, taxonomic category, or group of organisms). Remotely sensed data, including both airborne and spaceborne sensor data, vary in spatial, radiometric, spectral, and temporal resolutions. 1996, Jakubauskas 1997, Nyoungui et al.
Previous literature has reviewed the methods for integration of remote sensing and GIS (Ehlers et al. Parametric and nearestneighbor methods for hybrid classification: a comparison of pixel assignment accuracy. Reusing backpropagating artificial neural network for land cover classification in tropical savannahs. Using shade fraction image segmentation to evaluate deforestation in Landsat Thematic Mapper images of the Amazon region. 1988, Ekstrand 1996, Richter 1997, Gu and Gillespie 1998, Dymond and Shepherd 1999, Tokola et al. AVIRIS and EO1 Hyperion images with 224 bands). It evaluates each pixel spectrum as a linear combination of a set of endmember spectra (Adams et al. 1982, Civco 1989, Colby 1991, Meyer et al. Estimating pixelscale land cover classification confidence using nonparametric machine learning methods. A noisy classification result is often produced due to the high variation in the spatial distribution of the same class. 2000, Liu et al. Uncertainty and confidence in land cover classification using a hybrid classifier approach. The analyst is responsible for labelling and merging the spectral classes into meaningful classes. 1999, Congalton and Plourde 2002, Foody 2002b, 2004a). Fuzzy neural network models for supervised classification: multispectral image analysis. Classification of digital image texture using variograms. 2004). The interactive effect of spatial resolution and degree of internal variability within landcover types on classification accuracies. Evaluation of the greylevel cooccurrence matrix method for landcover classification using SPOT imagery. A review and analysis of back propagation neural networks for classification of remotely sensed multispectral imagery. Landsat TMbased forest damage assessment: correction for topographic effects. Identification of suitable textures involves determination of texture measure, image band, the size of moving window, and other parameters (Franklin et al. Comparing MODIS and ETM+ data for regional and global land classification. ICA mixture models for unsupervised classification of nongaussian classes and automatic context switching in blind signal separation. Another major drawback of the parametric classifiers lies in the difficulty of integrating spectral data with ancillary data. Improvement of forest type classification by SPOT HRV with 20m mesh DTM. Use of the average mutual information index in evaluating classification error and consistency. The fraction images are related to biophysical characteristics, and thus have the potential for improving classification (Roberts et al. Optimal classification methods for mapping agricultural tillage practices. The longwavelength radar data can penetrate the canopy structure to a certain depth and can provide information on vegetation stand structures (Leckie 1998, Santos et al. 2004, Olthof et al. 2004). You can see that the lumper ends up with two groups while the splitter ends up with several groups. Spectral shape classification of Landsat Thematic Mapper imagery. A segmentation and classification approach of IKONOS2 imagery for land cover mapping to assist flood risk and flood damage assessment. 2004). The study of uncertainty will be an important topic in the future research of image classification. 1997), and a combination of neural network and statistical approaches (Benediktsson and Kanellopoulos, 1999, Bruzzone et al. 1997, 1999) have been used for classification of multisource data.
Hence, the reflectance measured by the sensor can be treated as a sum of interactions among various classes of scene elements as weighted by their relative proportions (Strahler et al. Data fusion and multisource image classification. A review of current issues in the integration of GIS and remote sensing data. Pohl and Van Genderen (1998) provided a literature review on methods of multisensor data fusion. Dungan (2002) found that five types of uncertainties exist in remotely sensed data: positional, support, parametric, structural (model), and variables. Classification of remote sensing having high spectral resolution images. Improved urban land cover mapping using multitemporal IKONOS images for local government planning. 2001, Magnussen et al. Fisher (1997) summarized four causes of the mixed pixel problem: (1) boundaries between two or more mapping units, (2) the intergrade between central concepts of mappable phenomena, (3) linear subpixel objects, and (4) small subpixel objects. Hodgson et al.
GIS and remote sensing integration for environmental applications. Temporal resolution refers to the time interval in which a satellite revisits the same location. 2003). Making a definitive decision about the land cover class that each pixel is allocated to a single class.
An increase in spectral bands may improve classification accuracy, but only when those bands are useful in discriminating the classes (Thenkabail et al.
On the slopeaspect correction of multispectral scanner data. Soft classifications have been performed to minimize the mixed pixel problem using a fuzzy logic. Multisensor image fusion in remote sensing: concepts, methods, and applications. Accuracy assessment based on error matrix is the most commonly employed approach for evaluating perpixel classification, while fuzzy approaches are gaining attention for assessing fuzzy classification results. Dai and Khorram (1998) presented a hierarchical data fusion system for vegetation classification.
Correction of atmospheric and topographic effects for high spatial resolution satellite imagery. 1999, DeFries and Chan 2000, Lawrence et al. Integrating contextual information with perpixel classification for improved land cover classification. The output of SMA is typically presented in the form of fraction images, with one image for each endmember spectrum, representing the area proportions of the endmembers within the pixel. For example, forest distribution in mountainous areas is related to elevation, slope, and aspects. Texture, shape, and context information are currently most frequently used. 1986). Multiresolution wavelet decomposition image merger of Landsat Thematic Mapper and SPOT panchromatic data. The use of different seasons of remotely sensed data has proven useful for improving classification accuracy, especially for crop and vegetation classification (Brisco and Brown 1995, Wolter et al. 2003, Magnussen et al. Texture classification using features derived from random field models.
Objectbased classification of remote sensing data for change detection. Ancillary data, such as topography, soil, road, and census data, may be combined with remotely sensed data to improve classification performance. Previous literature has defined the meanings and provided computation methods for these elements (Congalton and Mead 1983, Hudson and Ramm 1987, Congalton 1991, Janssen and van der Wel 1994, Kalkhan et al. Fuzzyset classifiers, subpixel classifier, spectral mixture analysis.
A critical step is to develop suitable rules to combine the classification results from different classifiers. 2002, Neville et al. These disadvantages may lower classification accuracy if classifiers cannot effectively handle them (Irons et al. The difficulty in handling the dichotomy between vector and raster data models affects the extensive use of the perfield classification approach. Quality assessment of image classification algorithms for landcover mapping: a review and a proposal for a costbased approach. Realtime constrained linear discriminant analysis to target detection and classification in hyperspectral imagery. In many cases, a hierarchical classification system is adopted to take different conditions into account. Spectral features are the most important information for image classification. Nonparametric classifiers are thus especially suitable for the incorporation of nonspectral data into a classification procedure. 2001, Lu et al. The emphasis is placed on the summarization of major advanced classification approaches and the techniques used for improving classification accuracy. The major steps of image classification may include determination of a suitable classification system, selection of training samples, image preprocessing, feature extraction, selection of suitable classification approaches, postclassification processing, and accuracy assessment. In previous research, hyperspectral data have been successfully used for landcover classification (Benediktsson et al. 2002, Lucieer and Kraak 2004). Detecting subpixel woody vegetation in digital imagery using two artificial intelligence approaches. Data fusion involves two major procedures: (1) geometrical coregistration of two datasets and (2) mixture of spectral and spatial information contents to generate a new dataset that contains the enhanced information from both datasets. 1999) and have been used for image classifications (Gordon and Phillipson 1986, Franklin and Peddle 1989, Marceau et al. Although spatial information is remarkably useful for fine spatial resolution data, how to effectively derive and use it in image classification remains a research topic. 2003) are the most popular approaches used to overcome the mixed pixel problem.
Different approaches have been used to derive a soft classifier, including fuzzyset theory, DempsterShafer theory, certainty factor (Bloch 1996), softening the output of a hard classification from maximum likelihood (Schowengerdt 1996), IMAGINE's subpixel classifier (Huguenin et al. 1989, Ehlers 1990, Trotter 1991, Hinton 1996, Wilkinson 1996).
The huge amount of data storage and severe shadow problems in fine spatial resolution images lead to challenges in the selection of suitable imageprocessing approaches and classification algorithms. Cladistics tries to use as many derived characteristics as possible, including molecular distinctions.
A critical evaluation of the normalized error matrix in map accuracy assessment. Wavelet transform and spectral mixture analysis have also been used in recent years (Roberts et al. In general, a classification system is designed based on the user's need, spatial resolution of selected remotely sensed data, compatibility with previous work, imageprocessing and classification algorithms available, and time constraints. Sufficient reference data are available and used as training samples. Comparison of three different methods to select features for discriminating forest cover types using SAR imagery. Gaussian mixture discriminant analysis and subpixel land cover characterization in remote sensing.
MultiSpeca tool for multispectralhyperspectral image data analysis. Possible sampling designs include random, stratified random, systematic, double, and cluster sampling. Classification of multisource and hyperspectral data based on decision fusion.
1993, Roberts et al. Radar and optical data comparison/integration for urban delineation: a case study. Maximizing land cover classification accuracies produced by decision trees at continental to global scales. 2003). 2001, Rashed et al. Topographic normalization of Landsat Thematic Mapper digital imagery. Hard and soft classifications by a neural network with a nonexhaustively defined set of classes. Mapping land use/cover distribution on a mountainous tropical island using remote sensing and GIS. 5 Howick Place | London | SW1P 1WG. Spectral unmixing of hyperspectral imagery for mineral exploration: comparison of results from SFSI and AVIRIS. 2003, Magnussen et al. A comparison of methods for multiclass support vector machines. 1990, Kartikeyan et al. 2003). (1999), and Foody (2002b), have conducted reviews on classification accuracy assessment. 2001, Du et al. mean vector and covariance matrix) generated from the training samples are representative. The methods, including colourrelated techniques (e.g. Bolstad and Lillesand (1992) found that a rulebased classification with Landsat TM, soil, and terrain data yielded higher landcover classification accuracy than a standard spectralbased classification. neural network, decision tree), have their own strengths and limitations (Tso and Mather 2001, Franklin et al.