TY - GEN
T1 - Semi-supervised image classification in likelihood space
AU - Duan, Rong
AU - Jiang, Wei
AU - Man, Hong
PY - 2006
Y1 - 2006
N2 - This paper studies the problem of using limited amount of labeled data and large amount of unlabeled data in the training of a generative model for image classification, and proposes a likelihood space approach to improve the classification performance. Frequently when labeled data is limited, unlabeled data can help to improve classification performance if the assumption of the generative model structure in the classifier is correct. But classification accuracy can be degraded if the model structure assumption is incorrect. In this paper, we compare raw data space classification and likelihood space classification in semi-supervised learning framework, and we show that the classification performance can be improved in likelihood space when model is misspecified. We apply this likelihood space semi-supervised learning method in automatic target recognition on SAR images, and experimental results demonstrate the effectiveness of this proposed approach.
AB - This paper studies the problem of using limited amount of labeled data and large amount of unlabeled data in the training of a generative model for image classification, and proposes a likelihood space approach to improve the classification performance. Frequently when labeled data is limited, unlabeled data can help to improve classification performance if the assumption of the generative model structure in the classifier is correct. But classification accuracy can be degraded if the model structure assumption is incorrect. In this paper, we compare raw data space classification and likelihood space classification in semi-supervised learning framework, and we show that the classification performance can be improved in likelihood space when model is misspecified. We apply this likelihood space semi-supervised learning method in automatic target recognition on SAR images, and experimental results demonstrate the effectiveness of this proposed approach.
KW - Image classification
KW - Pattern recognition
KW - SAR
KW - Target recognition
KW - Unsupervised learning
UR - http://www.scopus.com/inward/record.url?scp=78649810407&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78649810407&partnerID=8YFLogxK
U2 - 10.1109/ICIP.2006.312634
DO - 10.1109/ICIP.2006.312634
M3 - Conference contribution
AN - SCOPUS:78649810407
SN - 1424404819
SN - 9781424404810
T3 - Proceedings - International Conference on Image Processing, ICIP
SP - 957
EP - 960
BT - 2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings
T2 - 2006 IEEE International Conference on Image Processing, ICIP 2006
Y2 - 8 October 2006 through 11 October 2006
ER -