Feature Point Detection for Real Time Applications

This paper presents a new feature point detector that is accurate, efficient and fast. A detailed qualitative evaluation of the proposed feature point detector for grayscale images is also done. Experiments have proved that this feature point detector is robust to affine transformations, noise and perspective deformations. More over the proposed detector requires only 22 additions per pixel to evaluate the interest point and its strength, making it one of the fastest detectors. The accuracy and speed of this algorithm makes it a strong contender for ASIC and hardware implementations and applications requiring e feature point abstraction.

• Type: Conference
• Year: 2008
• Where: IPCV
• Authors: Neeta Nain, Vijay Laxmi, Bhavitavya Bhadviya, Ameya Pandit, Chetna Goyal
• Keywords: Feature Point Abstraction, Feature Point Detector, Image Processing, IPCV 2008, Proposed Detector