Segmentation of range images using superquadric entities has been pointed out by a number of researchers as a powerful approach towards object recognition. Problems exist in finding an unbiased decision function for the assignment of a superquadric representation with an object prototype. This paper discusses current distance measures between recovered models and prototypes and presents a novel method for classification of uncertain superquadric representations using a maximum likelihood criterium and incorporating the range image characteristics of an active optical triangulation sensor. The main advantage of this probabilistic method is its inherent minimisation of the classification error rate. The approach is applied to recognition of electronic components for printed circuit board waste management.
|Title of host publication||Proceedings of MVA '96|
|Subtitle of host publication||IAPR Workshop on Machine Vision Applications : November 12-14 1996, Tokyo, Japan|
|Place of Publication||Tokyo, Japan|
|Publication status||Published - 12 Nov 1996|
|Event||IAPR Workshop on Machine Vision Applications, MVA 1996 - Tokyo, Japan|
Duration: 12 Nov 1996 → 14 Nov 1996
|Workshop||IAPR Workshop on Machine Vision Applications, MVA 1996|
|Period||12/11/96 → 14/11/96|
van Dop, E., & Regtien, P. (1996). Object recognition from range images using superquadric representations. In Proceedings of MVA '96 : IAPR Workshop on Machine Vision Applications : November 12-14 1996, Tokyo, Japan (pp. 267-270). Tokyo, Japan: Keio University.