Subject: CGAL users discussion list
List archive
- From: <>
- To: "" <>
- Subject: [cgal-discuss] Some questions regarding optimal alpha value
- Date: Thu, 23 Feb 2017 16:34:39 +0800
- Authentication-results: mail2-smtp-roc.national.inria.fr; spf=None ; spf=Pass ; spf=None
- Importance: normal
- Ironport-phdr: 9a23:gi4V4RP3C5hn1Wc/+VAl6mtUPXoX/o7sNwtQ0KIMzox0K/TyrarrMEGX3/hxlliBBdydsKMZzbOJ+Pm5BCRAuc/H6y9SNsQUFlcssoY/oU8JOIa9E0r1LfrnPWQRPf9pcxtbxUy9KlVfA83kZlff8TWY5D8WHQjjZ0IufrymUt2as8Pineu98pmWbwRTjyemeptzKg+3pEPfrINe1YBtI6J0xhrSqWZTYMxXw3lpLBSdhUCvyN23+cs34SRZveI8/MlGXo31eq05SfpTCzFwYDN939HiqRSWFVjH3XAbSGhDyhc=
Dear all,
I am new to this library and computational geometry and I am trying to use CGAL for my project. The problem is that I would like to know the shape of a 2D random point cloud’s boundary on a 2D surface. Thus I tried to get the alpha shape for this point cloud with some alpha value. Intuitively, I tried to used the optimal alpha computed by the following method.
Alpha_shape_2 A(points.begin(), points.end(), FT(100000), Alpha_shape_2::REGULARIZED); A.set_alpha(*A.find_optimal_alpha(1));
The result can be seen in the following picture:
As can be seen in the picture, it contains many small holes. In order to get rid of these holes, I am thinking to increase the alpha value a little bit. However, that does not guarantee a optimal result in other cases. Therefore, I would like to ask for any suggestions and help to eliminate these small holes. Also I am curious about what the value “1” in the “A.set_alpha(*A.find_optimal_alpha(1))” means and how exactly this “find_optimal_alpha” method works. Any explanation would be appreciated. Thank you very much!
Best regards, ZHAN, Xucong |
- [cgal-discuss] Some questions regarding optimal alpha value, zhanxucong, 02/23/2017
Archive powered by MHonArc 2.6.18.