USING HOUGH TRANSFORM IN LINE EXTRACTION
Efstratios Stylianidis, Petros G. Patias · ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences · 2000
In close-range images, normally a large number of geometrical features is available. For photogrammetrists, the detection of shapes such as straight lines is very useful. These features are very helpful for further photogrammetric work, such as sensor calibration, image orientation, DTM generation etc. Hough Transform (HT), is such a powerful tool for detecting predefined shapes (i.e. lines, ellipses). HT has been used for more than three decades in the areas of image processing, pattern recognition and computer vision. However, in digital close range photogrammetry HT has only rarely been used. This paper is a contribution on how HT can be used as a powerful technique for line extraction in close range photogrammetric problems.