Graphics Reference
In-Depth Information
Fig. 3.
The adding result of samples' projection onto the classification plane's normal vector
and offset for LBP
Table 2.
The cross validation result using the LBP feature
4
Conclusion
In this research paper we present a LBP-based automatic classification algorithm on
microscopic images of human embryos. It verified the validity of LBP as the feature
to do classification representing the local texture of embryo images, and it can be
make the two types of embryo images linearly separable; At the same time, when
combined with the SVM algorithm, the optimal projection direction for effectively
classifying LBP features was determined. The proposed algorithm is important for
computer-aided embryo transfer. In our further study, more training samples will be
collected for training purposes to aquire the best performance of the classifier.
References
1.
Santos Filho, E., Noble, J.A., Wells, D.: A Review on Automatic Analysis of Human Emb-
ryo Microscope Images. The Open Biomedical Engineering Journal 4, 170-177 (2010)
2.
Siristatidis, C., Pouliakis, A., Chrelias, C., Kassanos, D.: Artificial Intelligence in IVF: A
Need. Systems Biology in Reproductive Medicine 57, 179-185 (2011)
Search WWH ::
Custom Search