%0 Journal Article
%A CHEN Bei-jing
%A FAN Chun-nian
%A SU Qing-tang
%A WANG Ding-cheng
%A YANG Jian-hao
%T Block-Wise Two Dimensional Kernel Quaternion Principal Component Analysis
%D 2019
%R 10.13190/j.jbupt.2018-045
%J Journal of Beijing University of Posts and Telecommunications
%P 53-60
%V 42
%N 1
%X Currently, kernel quaternion principal component analysis (KQPCA) has been proposed and successfully applied to process linear quaternion signals. However, two dimensional version of KQPCA (2DKQPCA) has not been successfully implemented due to the quite time-consuming problem for diagonalizing the high dimensional kernel matrix. So, using the block-based idea and the parallel computing idea, the block-wise 2DKQPCA (B2DKQPCA) is proposed to implement 2DKQPCA really. After the overall consideration of computational complexity, application performance and quaternion Hermitian block, B2DKQPCA mainly processes the blocks of three directions:main-diagonal direction, anti-diagonal direction and side-diagonal direction. Then, B2DKQPCA is applied into RGB-D object recognition by combining B2DKQPCA and quaternion representation of RGB-D images. Experimental results on two publicly available datasets demonstrate that the proposed RGB-D object recognition algorithm based on the column direction B2DKQPCA outperforms some existing algorithms using principal component analysis and some existing algorithms using convolutional neural network.
%U https://journal.bupt.edu.cn/EN/10.13190/j.jbupt.2018-045