Redukcija oblaka tačaka primenom SVM metode
Point cloud reduction using support vector machines
Abstract
Ovaj rad istražuje mogućnosti smanjenja oblaka tačaka pomoću regresivne metode potporna vektorima (ε-SVR). ε-SVR je tehnika koja može da sprovede regresiju koristeći različite kernel funkcije (sigmoid, radial basis function, B-spline, spline, itd.) i pogodna je za detekcije ravnih regiona i regiona sa visokim neravninama. Korišćenjem ε-SVR gustina sačuvanih tačaka je prilagodljiva - sačuvane tačke su gušće u zakrivljenom a retke u ravnim regionima skeniranih podataka. Podešavanje regresivne greške, broj sačuvanih tačaka se mogu podešavati.
This paper explores the possibilities of point cloud reduction using ε insensitive support vector regression (ε-SVR). ε-SVR is a technique that can carry out the regression using different kernel functions (sigmoid, radial basis function, B-spline, spline, etc.) and it is suitable for detection of flat regions and regions with high curvature in scanned data. Using ε-SVR the density of preserved points is adaptive - preserved points are denser at highly curved region and rare at flat regions. Adjusting the error cost in the regression, the number of preserved points can be fine tuned.
Keywords:
SVM metoda / reverzibilno inženjerstvo / redukcija oblaka tačaka / support vector machines / reverse engineering / point cloud reductionSource:
Journal of Production Engineering, 2012, 15, 2, 59-62Publisher:
- Univerzitet u Novom Sadu - Fakultet tehničkih nauka - Departman za proizvodno mašinstvo, Novi Sad
Collections
Institution/Community
Mašinski fakultetTY - JOUR AU - Jakovljević, Živana PY - 2012 UR - https://machinery.mas.bg.ac.rs/handle/123456789/1419 AB - Ovaj rad istražuje mogućnosti smanjenja oblaka tačaka pomoću regresivne metode potporna vektorima (ε-SVR). ε-SVR je tehnika koja može da sprovede regresiju koristeći različite kernel funkcije (sigmoid, radial basis function, B-spline, spline, itd.) i pogodna je za detekcije ravnih regiona i regiona sa visokim neravninama. Korišćenjem ε-SVR gustina sačuvanih tačaka je prilagodljiva - sačuvane tačke su gušće u zakrivljenom a retke u ravnim regionima skeniranih podataka. Podešavanje regresivne greške, broj sačuvanih tačaka se mogu podešavati. AB - This paper explores the possibilities of point cloud reduction using ε insensitive support vector regression (ε-SVR). ε-SVR is a technique that can carry out the regression using different kernel functions (sigmoid, radial basis function, B-spline, spline, etc.) and it is suitable for detection of flat regions and regions with high curvature in scanned data. Using ε-SVR the density of preserved points is adaptive - preserved points are denser at highly curved region and rare at flat regions. Adjusting the error cost in the regression, the number of preserved points can be fine tuned. PB - Univerzitet u Novom Sadu - Fakultet tehničkih nauka - Departman za proizvodno mašinstvo, Novi Sad T2 - Journal of Production Engineering T1 - Redukcija oblaka tačaka primenom SVM metode T1 - Point cloud reduction using support vector machines EP - 62 IS - 2 SP - 59 VL - 15 UR - https://hdl.handle.net/21.15107/rcub_machinery_1419 ER -
@article{ author = "Jakovljević, Živana", year = "2012", abstract = "Ovaj rad istražuje mogućnosti smanjenja oblaka tačaka pomoću regresivne metode potporna vektorima (ε-SVR). ε-SVR je tehnika koja može da sprovede regresiju koristeći različite kernel funkcije (sigmoid, radial basis function, B-spline, spline, itd.) i pogodna je za detekcije ravnih regiona i regiona sa visokim neravninama. Korišćenjem ε-SVR gustina sačuvanih tačaka je prilagodljiva - sačuvane tačke su gušće u zakrivljenom a retke u ravnim regionima skeniranih podataka. Podešavanje regresivne greške, broj sačuvanih tačaka se mogu podešavati., This paper explores the possibilities of point cloud reduction using ε insensitive support vector regression (ε-SVR). ε-SVR is a technique that can carry out the regression using different kernel functions (sigmoid, radial basis function, B-spline, spline, etc.) and it is suitable for detection of flat regions and regions with high curvature in scanned data. Using ε-SVR the density of preserved points is adaptive - preserved points are denser at highly curved region and rare at flat regions. Adjusting the error cost in the regression, the number of preserved points can be fine tuned.", publisher = "Univerzitet u Novom Sadu - Fakultet tehničkih nauka - Departman za proizvodno mašinstvo, Novi Sad", journal = "Journal of Production Engineering", title = "Redukcija oblaka tačaka primenom SVM metode, Point cloud reduction using support vector machines", pages = "62-59", number = "2", volume = "15", url = "https://hdl.handle.net/21.15107/rcub_machinery_1419" }
Jakovljević, Ž.. (2012). Redukcija oblaka tačaka primenom SVM metode. in Journal of Production Engineering Univerzitet u Novom Sadu - Fakultet tehničkih nauka - Departman za proizvodno mašinstvo, Novi Sad., 15(2), 59-62. https://hdl.handle.net/21.15107/rcub_machinery_1419
Jakovljević Ž. Redukcija oblaka tačaka primenom SVM metode. in Journal of Production Engineering. 2012;15(2):59-62. https://hdl.handle.net/21.15107/rcub_machinery_1419 .
Jakovljević, Živana, "Redukcija oblaka tačaka primenom SVM metode" in Journal of Production Engineering, 15, no. 2 (2012):59-62, https://hdl.handle.net/21.15107/rcub_machinery_1419 .