2026-05-13T16:19:41.026Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:1:6372 2018-11-18T21:39:10Z 1

Conic properties of planar point sets 2008 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/1/6372 http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.714.5822&rep=rep1&type=pdf M. Stojmenović A. Nayak Basic shape descriptors such as linearity, circularity and ellipticity are useful tools in the field of pattern recognition. It is possible to identify groups of points as lines, circles or ellipses in an image. This is potentially a useful step in image based object detection. Our goal is to find algorithms that give shape measurements in the interval [0, 1], where values close to 1 indicate that the shape in question is a line, circle or ellipse, and values close to 0 indicate the opposite. We are interested in measures which are invariant to rotation, scaling, and translation. These measures should also be calculated very quickly and be resistant to protrusions in the data set. All of the measures surveyed here are shape boundary based which makes them applicable to point sets such as extracted edges from real world images. The main linearity measures are found in [13]. Circularity measures were discussed in [11], and an overview of such measures applied to unordered point sets is presented here. The circularity of unordered data is determined directly from the linearity measure, whereas the circularity of ordered data is derived by multiplying the unordered data circularity measure by a monotonicity factor. The proposed algorithms work on both open and closed curves. Direct ellipse fitting methods, discussed in [12], are guaranteed to specifically return an ellipse as the fit rather than any conic. The proposed algorithms work well on both open and closed curves. They are used to classify a set of 25 curves in 4 categories: line, circle, ellipse, or no shape. These three measures together could form a base for a larger shape detection system in images. conferenceObject Relations, Orders and Graphs: Interaction wtih Computer Science