Detection of generalized principal axes in rotationally symmetric shapes
References (11)
- et al.
Contour registration by shape-specific points for shape matching
Comput. Vision Graphics Image Process
(1983) 3D reconstruction: the registration problem
Comput. Vision Graphics Image Process
(1988)- et al.
Automatic rotational symmetry determination for shape analysis
Pattern Recognition
(1987) - et al.
Recognitive aspects of moment invariants
IEEE Trans. Pattern Anal. Mach. Intell. PAMI-6
(1984) - et al.
Image normalization by complex moments
IEEE Trans. Pattern Anal. Mach. Intell. PAMI-7
(1985)
Cited by (70)
Robust detectors of rotationally symmetric shapes based on novel semi-shape signatures
2023, Pattern RecognitionShape measurement using LIP-signature
2018, Computer Vision and Image UnderstandingCitation Excerpt :The classical approach (Jain et al., 1995) defined orientation as the axis of the least second moment of inertia. Tsai and Chou (1991) proposed to detect principal axes of shapes having symmetry properties. Lin (1993, 1996) defined shape orientation using universal principal axes, formed of half lines starting from the shape centroid.
Orientation and anisotropy of multi-component shapes from boundary information
2011, Pattern RecognitionCitation Excerpt :It is worth mentioning that the latter method is analogous to the standard method which defines the shape orientation by the line that minimises the integral of squared distances of the points in the shape to the line. A use of a higher exponent than 2 (in the standard method) is studied in [31,36]. Namely, it is well known that the standard method should be modified in order to be applicable to a wider class of shapes.
Shape elongation from optimal encasing rectangles
2010, Computers and Mathematics with ApplicationsCurvature weighted gradient based shape orientation
2010, Pattern RecognitionCitation Excerpt :There are already several definitions and methods developed to determine and compute shape orientation. Different techniques have been used, including those based on geometric moments, complex moments, principal component analysis, etc., but, there is still an ongoing demand for the new methods; as well as previously established methods [5–8,10], there are also very recent ones [2,3,9,13,14], as well. In this paper we define a new boundary based method for computing the shape orientation.
A Hu moment invariant as a shape circularity measure
2010, Pattern RecognitionCitation Excerpt :Note also that, due to the diversity of shapes and the diversity of applications in different areas such as computer science, medicine, biology, robotics, etc., several methods are often developed for measuring the same shape property. As an illustration we list just a few known methods for measuring shape convexity: [1,11,19,21,26,37] and for computing the orientation of a shape: [3,8,13,27,28,33]. The latter is often used within a shape normalisation procedure that is sometimes required prior to further shape analysis.
- §
Sheng-Lin Chou is also with Advanced Technology Center, Electronics Research and Service Organization, Industrial Technology Research Institute, Chutung, Hsinchu, Taiwan 31015, R. O. C.