Object-Oriented Information Extraction From High Resolution Remote Sensing Imagery

The development of the remote sensing technology makes us obtain very abundant information of the nature, especially with the appearance of high resolution remote sensing, it extends the visual field of the nature. Compared with the low or middle resolution image, the high resolution remote sensing image such as QuickBird and IKONOS has richer structure information and the texture information. It will result in not only reducing the accuracy of classification but also making the spatial data redundant and wasting the resource when the single traditional classification method based on spectrum of pixels is applied to the high resolution remote sensing image.A new information extraction method-object oriented information extraction—is introduced in extracting information from high resolution remote sensing image. Image segmentation and information extraction are the most important in the method. An objective function is introduced and used to select the optimal scale, and a segmentation hierarchical is established combined with the max-area method. Experiment is done based on high resolution image—QuickBird and IKONOS.The work is done in the word as following:1) In the multi-segmentation method, users are needed to supply suitable parameters to get a good segmented result. An objective function is proposed to measure the quality of the resulting segmentation, so a best segmentation result can be got. The function aims at maximizing intrasegment homogeneity an intersegment heterogeneity. Intrasegment homogeneity ensures the simple of image-object and intersegment heterogeneity ensures the separability of image-object.2) According to the character of study area, decision tree analysis is used to select the suitable feature for each class, and a classification system is founded based on this. A better result is got.3) Pixel-based and object-oriented image classification approaches are compared in three aspects: (1) compare of theory; (2) compare of eyes; (3) compare of classification accuracy.