TY - GEN
T1 - Non-negative locality-constrained linear coding for image classification
AU - Liu, Guo Jun
AU - Liu, Yang
AU - Guo, Mao Zu
AU - Liu, Pei Na
AU - Wang, Chun Yu
N1 - Publisher Copyright:
© Springer International Publishing Switzerland 2015.
PY - 2015
Y1 - 2015
N2 - The most important issue of image classification algorithm based on feature extraction is how to efficiently encode features. Locality-constrained linear coding (LLC) has achieved the state of the art performance on several benchmarks, due to its underlying properties of better construction and local smooth sparsity. However, the negative code may make LLC more unstable. In this paper, a novel coding scheme is pro- posed by adding an extra non-negative constraint based on LLC. Generally, the new model can be solved by iterative optimization methods. Moreover, to reduce the encoding time, an approximated method called NNLLC is proposed, more importantly, its computational complexity is similar to LLC. On several widely used image datasets, compared with LLC, the experimental results demonstrate that NNLLC not only can improve the classification accuracy by about 1-4 percent, but also can run as fast as LLC.
AB - The most important issue of image classification algorithm based on feature extraction is how to efficiently encode features. Locality-constrained linear coding (LLC) has achieved the state of the art performance on several benchmarks, due to its underlying properties of better construction and local smooth sparsity. However, the negative code may make LLC more unstable. In this paper, a novel coding scheme is pro- posed by adding an extra non-negative constraint based on LLC. Generally, the new model can be solved by iterative optimization methods. Moreover, to reduce the encoding time, an approximated method called NNLLC is proposed, more importantly, its computational complexity is similar to LLC. On several widely used image datasets, compared with LLC, the experimental results demonstrate that NNLLC not only can improve the classification accuracy by about 1-4 percent, but also can run as fast as LLC.
KW - Image classification
KW - Locality-constrained Linear Coding (LLC)
KW - Non-negative constraint
KW - Spatial Pyramid Matching (SPM)
UR - https://www.scopus.com/pages/publications/84951776881
U2 - 10.1007/978-3-319-23989-7_47
DO - 10.1007/978-3-319-23989-7_47
M3 - 会议稿件
AN - SCOPUS:84951776881
SN - 9783319239873
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 462
EP - 471
BT - Intelligence Science and Big Data Engineering
A2 - He, Xiaofei
A2 - Zhou, Zhi-Hua
A2 - Gao, Xinbo
A2 - Liu, Zhi-Yong
A2 - Zhang, Yanning
A2 - Fu, Baochuan
A2 - Hu, Fuyuan
A2 - Zhang, Zhancheng
PB - Springer Verlag
T2 - 5th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2015
Y2 - 14 June 2015 through 16 June 2015
ER -