| Summary: | Landuse and land-cover classification from high resolution imagery has been seen as challenging by theremote sensing society. The high variability from pixel to pixel makes the use of pixel-based classifiersobsolete. Object-based classifiers along with rule-based descriptors can be used to overcome theseproblems because they consider the spatial distribution and topological relationships of the pixels. Theidentification of the landuse/land-cover classes based on objects and their spatial relationships can lead toa better classification results. The combination through fusion techniques of images with differentresolutions can also be used to improve landuse/land-cover classification.The objective of this research is to evaluate pixel- and object-based approaches for landuse/land-coverclassification and to identify which approach gives better results for high resolution imagery.QuickBird imagery covering the town of Richmond Hill, in the Greater Toronto Area (GTA), Ontario,Canada were used for landuse/land-cover classification. The classes considered were: water, low-densityresidential, transportation, contruction site, forest, golf course, corn, wheat, fallow, rapeseeds, pasture,parks, new low-density residential, commercial and industrial. The fusion techniques were used to mergePan and MS images were performed as an initial step: RGB-HIS using PCI Geomatica and wavelettransform with IHS using Matlab and Erdas Imagery. Pixel-based classifiers, such as MLC and Contextualwere compared to the object- and rule-based approach implemented in eCognition. It was found that thebest pixel-based classification results were obtained from MLC using 1-4 channels (kappa coefficient0.80662 and overall accuracy 83.71%). However, the classification from Wavelet-IHS Transformationfusion result implemented in Matlab with MLC (kappa coefficient 0.77637 and overall accuracy 81.10%)showed a balance between low loss of spectral information and the improved classification for objects notclearly defined in the original MS imagery.For the object-based and rule-based approach, it was found that a segmentation of 4 levels with theidentification of major land-cover types in the smallest scale and the integration of rules for theidentification of landuse classes in the other levels led to the best classification result for high-resolutionimagery (kappa coefficient 0.8565, and overall accuracy 86.70%). However, rules that describe classessuch as ‘parks’, ‘commercial’ and ‘industrial’ sites have to be improved in order to increase theidentification of these areas
|
|---|
