Moment-Based Methods of Identification and Localization of Partially Visible Objects
The paper contains the brief description of identification of partially occluded 2-D objects and 3-D objects represented by their silhouettes (which also can be occluded). The algorithms are based on moment expressions. For occluded objects the moments of contours are applied instead of the moments of regions.
Unable to display preview. Download preview PDF.
- 1.Hu, M.K.: Pattern recognition by moment invariants, IRE Trans. Inf. Theory, IT-8 (1962), 179–187.Google Scholar
- 2.Cyganski, D. et al.: A tensor operator method for identifying the affine transformation relating image pairs, Proc. IEEE Conf. Comp.Vision and Patt. Rec., 1983, 361–363.Google Scholar
- 3.Pinjo, Z. et al.: Determination of 3-D object orientation from projections, Patt. Rec. Lellers, 3 (1985), 351–356.Google Scholar
- 4.Śluzek, A.: Identification of 2-D objects in 3-D space, in: Signal Processing III (Ed. I.T.Young et al.), North-Holland, Amsterdam 1986, 921–924.Google Scholar
- 5.Scdjadi, F.A. and E.L. Hall.: Three-dimensional moment invariants, IEEE Trans.PAMI,PAMI-2 (1980), 127–136.Google Scholar
- 6.Śluzek, A.: Using moment invariants to recognize and locate partially occluded 2-D objects, Patt. Rec. Letters, 7 (1988), 253–257.Google Scholar