A Comparison of the Performance of Pixel Based and Object Based Classifications over Images with Various Spatial Resolutions

In the last three decades, advances in computer technology, earth observation sensors and GIS science, led to the development of “Object-based Image Analysis” as an alternative to the traditional pixel-based image analysis. Many studies have shown that traditional pixel-based image analysis is limited because it uses only spectral information of single pixels and this approach produces poor results especially with high spatial resolution satellite images. By contrast, object-based image analysis works on (homogeneous) objects which are produced by image segmentation and allows using more elements in the classification. As an object is a group of pixels, object characteristics such as mean value, standard deviation of spectral values, etc. can be calculated; besides shape and texture features of the objects are available and can be used to differentiate land cover classes with similar spectral information. These extra types of information give object-based image analysis the potential to produce land cover thematic maps with higher accuracies than those produced by traditional pixel-based method. In this study, we look at the performance of object-based image analysis in classifying satellite images with different spatial resolutions; comparing the classification results with those produced by the pixel-based method, we intend to find out how spatial resolution of satellite images influences the performance of object-based image analysis. The experiment showed that with relatively high spatial resolution images such as SPOT-5 and Landsat-7 ETM+, object-based image analysis obtained higher accuracy than that by the pixel-based one; while for coarse resolution images with 100 and 250 m spatial resolution, object based image analysis did not obtain higher accuracy. This study shows that the object-based image analysis has advantage over the pixel-based one, however the advantage was only found in the higher spatial resolution images.