Automatic building extraction is a bottleneck of automatic surveying and mapping of big scale urban image. Because of the complexity of urban imagery, automatic building extraction has many problems to resolve. And there is still a long way to go to put the automatic mapping of building to use. A semiautomatic building extraction is more practical in production. In the paper, the semiautomatic building extraction of big scale urban imagery has been studied, and the main works are summarized as below:1. Introducing the background of research and the development trend of building extraction.2. Studying the essential arithmetic of image processing and selecting the valid methods of edge detection and line extraction to extract buildings in big scale urban image.3. By means of Perceptual Organization Method, arranging the line segments of building's edges to reconstruct the structure of the roof.4. Presenting two methods of building extraction. One is based on region segmentation another is based on Hough Transform action. Using the methods, we can extract building automatically and the result is perfect. |