Design of Corner Detection Techniques Based on Computational Intelligence

• Main Authors: Ming-Tsung Liu, 劉明宗
• Other Authors: Pao-Ta Yu
• Format: Others
• Language: en_US
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/32513425919650400629

博士 === 國立中正大學 === 資訊工程所 === 98 === Two fast two-stage corner detectors with noise tolerance are presented in this dissertation. At the first stage, candidate-corner pixels are selected by the proposed candidate pruning approaches. At the second stage, real corners are recognized by the Harris detector among the candidate-corner pixels. In general, corners are considered as the junction of edges. Therefore, edge pixels with a high gradient in more than one direction can be selected as candidate-corner pixels. Meanwhile, impulse noise often corrupts digital images while images are captured by using a camera with faulty sensors or are being transmitted over an unreliable channel. Noisy pixels always cause false detections in most corner detectors. Those noisy pixels thus must be excluded from candidate-corner pixels to enhance the noise tolerance capability. In these two-stage corner detectors proposed in this dissertation, candidate-corner pixels are selected based on local features that are extracted from a sliding observation window. These features including gradient, edge, and impulse noise are regarded as pieces of evidence and are further combined by using Dempster-Shafer evidence theory or support vector machines to yield a final belief value. Through the well-selection of candidate-corner pixels, the proposed candidate pruning approaches can 1) enhance the noise tolerance capability, and 2) reduce the computational cost of the proposed corner detector. Experimental results show that the proposed methods outperformed other well-known corner detectors.