
184
4 CONCLUSION
Local gradient thresholds are usually needed in the detection of interested boundary surfaces
from 3D images, but their computation is a very complex problem. Inspired by the adaptive
contrast detection mechanism of human vision, this paper proposes a feasible method to
compute local gradient thresholds in 3D images. Experimental results show that the pro-
posed method is effective for step-like boundary surfaces in many different 3D images.
ACKNOWLEDGEMENT
This work was supported in part by the NSFC of China (61375020), 973 program of
China (2013CB329401) and Cross Research Fund of Biomedical Engineering of SJTU
(YG2013ZD02, Y