Abstract:
Detecting Phases in Steel Microstructure is one of the interesting problem in
the field of computer vision. In this work, we discuss a pixel based classification
approach. A classifier is only as good as the information you give it.
On the other hand it may have a huge intrinsic disproportion the number of
examples in each class, Which hinder the classification performance. There
are many ways you can adjust how you’re representing your input data for
learning of model. In this paper, we propose ensemble method based on
outlier detection to comprehend better the data used as a part of learning of
model , which is random forest and discuss it merits and demerits of other
related method which we use.
