Accurate and robust line segment extraction by analyzing distribution around peaks in Hough space

Hough transform (HT) is a well-known technique for extracting lines. However, it is difficult for most existing HT methods to extract line segments robustly from complicated images, mainly because the influence from various objects other than line segments are not taken into account. This paper proposes an accurate and robust evaluator that dynamically removes contributions of backgrounds and analyzes voting patterns around peaks in the accumulator space. In the experiments, four peak detection algorithms are tested against seven images completely automatically. Results show that our method is superior to existing methods in terms of accuracy and robustness while there are no clear differences in execution time. The proposed evaluator detects peaks after the HT and hence it can be applied to any HT that keeps the basic characteristics of the voting process.