3D range image surface feature classification using tripod operators

Songtao Li; Dongming Zhao

doi:10.1117/12.334334

29 December 1998 3D range image surface feature classification using tripod operators

Songtao Li, Dongming Zhao

Proceedings Volume 3520, Three-Dimensional Imaging, Optical Metrology, and Inspection IV; (1998) https://doi.org/10.1117/12.334334
Event: Photonics East (ISAM, VVDC, IEMB), 1998, Boston, MA, United States

Abstract

The tripod operator consists of a class of feature extraction operators for range images. Each tripod consist of three points in a 3D space fixed at the vertices of an equilateral triangle. A triangle is moved as a rigid body unit the three vertices lie on the range image surface. Several angle measurements can be made from the linkable tripod operators. The resulting angles between linkable triangles are local shape features which are invariant to scale, transformation and rotation. In this report a new approach to shape classification is proposed. The first step in the approach is to take angle measurement through random placement of the tripod operators in a range image. The second step is to generate two histograms based on two resulting angle features generated from the first step. The third step is to extract a number of invariant features, such as SIGN of coefficients, using the Fourier transforms of the two histograms. Based on the analysis of these invariant features, 3D surfaces can be classified. The experiments show that tripod operators can generate a large number of surface features for object classification and further for recognition.

Citation Download Citation

Songtao Li and Dongming Zhao "3D range image surface feature classification using tripod operators", Proc. SPIE 3520, Three-Dimensional Imaging, Optical Metrology, and Inspection IV, (29 December 1998); https://doi.org/10.1117/12.334334

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available