Title :
Shape from stereo images for small objects
Author_Institution :
Fac. of Eng., Takushoku Univ., Hachioji, Japan
Abstract :
A novel method of shape estimation from stereo images is presented, which is based on geometrical relations targeting small objects. Each of left and right images consists of a number of images obtained with different focusing distances, taking a limited depth of focus into account. In the absence of enough textures, the object is made to have artificial textures, for example, by putting carbon powders on its surface. Shapes were experimentally obtained for an object of size 8.2 × 4.5 × 3.5 mm3 to show the procedure and to examine the method. The method may be useful to observe shapes for small objects.
Keywords :
estimation theory; image texture; stereo image processing; artificial textures; carbon powder; geometrical relations; shape estimation; stereo images; Cameras; Carbon; Estimation; Focusing; Geometry; Shape; Stereo vision; depth of focus; geometrical relations; shape estimation; small objects; stereopsis;
Conference_Titel :
Information Science, Electronics and Electrical Engineering (ISEEE), 2014 International Conference on
Conference_Location :
Sapporo
Print_ISBN :
978-1-4799-3196-5
DOI :
10.1109/InfoSEEE.2014.6947767