Corner Detection Are Useful for Computer Vision Applications Essay

Best Essays
Corner detection and its parameters: position, model and orientation are useful for many computer vision applications, such as object recognition, matching, segmentation, 3D reconstruction, motion estimation [2, 3, 4, 34.] indexing, retrieval, robot navigation and in our case edge tracking from geometry design. This need has driven the development of a large number of corner detectors [1, 5, 6, 7, 8, 9, 10, 11, 12, 13.]. Other methods for corner detection are described in [14, 15]. These detectors compete with each other in terms of precision localization, accuracy, speed, and information they provide. Model classification and orientation are the most interest information needed in process of edge tracking.
For some of these approaches,
…show more content…
Corner strength has been first defined by Noble [12] from which a slightly different version has been proposed by Harris and Stephen [11]: (1.2)
The role of the parameter k is to remove sensitivity to strong edges.
The Plessey operator uses estimates of the variance of the gradient of an image in a set of overlapping neighborhoods. This detector, which produced much interest, was extended by including local gray-level invariants based on combinations of Gaussian derivatives [17].
One of the earliest detectors [16], which was based on the Moravec operator, defines corners to be local extrema in the determinant of the Hessian Matix, H=M.
The Kitchen and Rosenfeld operator [5] uses an analysis of the curvature of the grey-level variety of an image. The SUSAN operator [18] uses a form of grey-level moment that is designed to detect V- corners, and which is applied to other model of corner.
The earlier Forstner [22] algorithm is easily explained in terms of H (Hessian Matrix). For a more recently proposed detector [20], it has been shown [21] that under affine motion, it is better to use the smallest eigenvalue of H as the corner strength function.
Recently, George Azzopardi and Nicolai Petkov [36] propose a trainable filter which we call Combination Of Shifted FIlter REsponses (COSFIRE) and use for keypoint detection and pattern recognition.
2) Contour based methods
These methods extract contours and then
Get Access