| |
Abstract:
We propose a new and efficient technique to incorporate
contextual information into object classification. Most of the
current techniques faces the problem of exponential computation
cost, which is dealt with by either limiting to local searching
space, which might cause incomplete context, or simplifying the
problem using specific domain knowledge. In this paper, we propose
a new general framework that incorporates
partial
context at a linear cost. This technique is applied to microscopic
urinalysis image recognition, resulting in significant improvement
of recognition rate than context free approach, which would have
been impossible by conventional context incorporating
techniques.
|
