The improvement of spatial resolution of remote sensing images has greatly enriched the information content of ground objects, but also posed great challenges to some traditional target extraction methods. This paper combines the object-based idea with the Boosting algorithm to propose a new method for automatic extraction of buildings in high-resolution remote sensing images. This method effectively solves the problem of poor performance when using sliding windows of predefined shapes and sizes to detect targets in general methods by constructing object networks to associate image segmentation and recognition. Then, an effective feature classifier is trained for the target characteristics of the building, and the label confidence is used to comprehensively analyze various types of image information to complete target extraction and subsequent processing. Experimental results show that this method can be used to extract buildings of various types and structures with high accuracy and good robustness, and has high application value.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore