Open Access
ARTICLE
Image Classification Using Optimized MKL for SSPM
Lu Wu, Quan Liu, Ping Lou
School of Information Engineering, Key Laboratory of Fiber Optic Sensing Technology and Information Processing, Wuhan University of Technology, Wuhan, Hubei, China
* Corresponding Author: Lu Wu,
Intelligent Automation & Soft Computing 2019, 25(2), 249-257. https://doi.org/10.31209/2018.100000010
Abstract
The scheme of spatial pyramid matching (SPM) causes feature ambiguity near
dividing lines because it divides an image into different scales in a fixed manner.
A new method called soft SPM (sSPM) is proposed in this paper to reduce
feature ambiguity. First, an auxiliary area rotating around a dividing line in four
orientations is used to correlate the feature relativity. Second, sSPM is
performed to combine these four orientations to describe the image. Finally, an
optimized multiple kernel learning (MKL) algorithm with three basic kernels for
the support vector machine is applied. Specifically, for each level, a suitable
kernel is selected to map the data that fall within the corresponding
neighbourhood. In addition, a mixed-norm regularization formulation is
optimized using MKL to solve the classification problem. The method proposed
in this paper performs well when applied to the Caltech 101 and Scene 15
datasets. Experimental results are collected under various conditions. The
results of sSPM are improved by nearly 4% compared with the existing
experimental results.
Keywords
Cite This Article
L. Wu, Q. Liu and P. Lou, "Image classification using optimized mkl for sspm,"
Intelligent Automation & Soft Computing, vol. 25, no.2, pp. 249–257, 2019. https://doi.org/10.31209/2018.100000010