Performance assessment of feature detection algorithms: a methodology and case study on corner detectors

• Published in: IEEE transactions on image processing Vol. 12; no. 12; pp. 1668 - 1676
• Main Author: Rockett, P.I.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.12.2003; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Appraisal; Computer aided software engineering; Computer vision; Corners; Detection algorithms; Detectors; Discrimination; Exact sciences and technology; Image edge detection; Image processing; Information, signal and communications theory; Labeling; Marking; Methodology; Performance assessment; Robustness; Signal processing; Telecommunications and information theory; Testing; Performance evaluation; feature detection; Angle; Image processing; Object detection; Feature extraction; Shape detection; Corner detection; ROC curves
• ISSN: 1057-7149; 1941-0042
• DOI: 10.1109/TIP.2003.818041

We describe a generic methodology for evaluating the labeling performance of feature detectors. We describe a method for generating a test set and apply the methodology to the performance assessment of the three well-known corner detectors of L. Kitchen and A. Rosenfeld (1982), of K. Paler et al. (1984), and of C. Harris and M. Stephens (1988). The labeling deficiencies of each of these detectors is related to their discrimination ability between corners and various of the features which comprise the class of noncorners.