TY - JOUR
T1 - Integrated visual vocabulary in latent Dirichlet allocation-based scene classification for IKONOS image
AU - Kusumaningrum, Retno
AU - Wei, Hong
AU - Manurung, Ruli
AU - Murni, Aniati
N1 - Funding Information:
The first author would like to acknowledge the funding support by the Directorate General Higher Education, Ministry of Education and Cultural, Indonesia, in part of Sandwich-Like Program 2012. We would also like to thank Diponegoro University for providing the research grant for the first author.
PY - 2014/1
Y1 - 2014/1
N2 - Scene classification based on latent Dirichlet allocation (LDA) is a more general modeling method known as a bag of visual words, in which the construction of a visual vocabulary is a crucial quantization process to ensure success of the classification. A framework is developed using the following new aspects: Gaussian mixture clustering for the quantization process, the use of an integrated visual vocabulary (IVV), which is built as the union of all centroids obtained from the separate quantization process of each class, and the usage of some features, including edge orientation histogram, CIELab color moments, and gray-level co-occurrence matrix (GLCM). The experiments are conducted on IKONOS images with six semantic classes (tree, grassland, residential, commercial/industrial, road, and water). The results show that the use of an IVV increases the overall accuracy (OA) by 11 to 12% and 6% when it is implemented on the selected and all features, respectively. The selected features of CIELab color moments and GLCM provide a better OA than the implementation over CIELab color moment or GLCM as individuals. The latter increases the OA by only ~2 to 3%. Moreover, the results show that the OA of LDA outperforms the OA of C4.5 and naive Bayes tree by ~20%.
AB - Scene classification based on latent Dirichlet allocation (LDA) is a more general modeling method known as a bag of visual words, in which the construction of a visual vocabulary is a crucial quantization process to ensure success of the classification. A framework is developed using the following new aspects: Gaussian mixture clustering for the quantization process, the use of an integrated visual vocabulary (IVV), which is built as the union of all centroids obtained from the separate quantization process of each class, and the usage of some features, including edge orientation histogram, CIELab color moments, and gray-level co-occurrence matrix (GLCM). The experiments are conducted on IKONOS images with six semantic classes (tree, grassland, residential, commercial/industrial, road, and water). The results show that the use of an IVV increases the overall accuracy (OA) by 11 to 12% and 6% when it is implemented on the selected and all features, respectively. The selected features of CIELab color moments and GLCM provide a better OA than the implementation over CIELab color moment or GLCM as individuals. The latter increases the OA by only ~2 to 3%. Moreover, the results show that the OA of LDA outperforms the OA of C4.5 and naive Bayes tree by ~20%.
KW - IKONOS
KW - bag of visual words
KW - integrated visual vocabulary
KW - latent Dirichlet allocation
KW - scene classification
UR - http://www.scopus.com/inward/record.url?scp=84897798634&partnerID=8YFLogxK
U2 - 10.1117/1.JRS.8.083690
DO - 10.1117/1.JRS.8.083690
M3 - Article
AN - SCOPUS:84897798634
SN - 1931-3195
VL - 8
JO - Journal of Applied Remote Sensing
JF - Journal of Applied Remote Sensing
IS - 1
M1 - 083690
ER -