Global voting schemes based on the Hough transform (HT) have been widely usedto robustly detect lines in images. However, since the votes do not take lineconnectivity into account, these methods do not deal well with clutteredimages. In opposition, the so-called local methods enforce connectivity butlack robustness to deal with challenging situations that occur in manyrealistic scenarios, e.g., when line segments cross or when long segments arecorrupted. In this paper, we address the critical limitations of the HT as aline segment extractor by incorporating connectivity in the voting process.This is done by only accounting for the contributions of edge points lying inincreasingly larger neighborhoods and whose position and directional contentagree with potential line segments. As a result, our method, which we callSTRAIGHT (Segment exTRAction by connectivity-enforcInG HT), extracts thelongest connected segments in each location of the image, thus also integratinginto the HT voting process the usually separate step of individual segmentextraction. The usage of the Hough space mapping and a correspondinghierarchical implementation make our approach computationally feasible. Wepresent experiments that illustrate, with synthetic and real images, howSTRAIGHT succeeds in extracting complete segments in several situations wherecurrent methods fail.
展开▼
