Volume no :2, Issue no: 1, March (2015)

MULTI-FOCUS IMAGE FUSION BASED ON NONSUBSAMPLED SHEARLET TRANSFORM AND PULSE COUPLED NEURAL NETWORK WITH SELF-SIMILARITY AND DEPTH INFORMATION

Author's: Liu Shuaiqi, Shi Mingzhu, Zhao Jie, Geng Peng and Zhang Zhong
Pages: [47] - [65]
Received Date: January 26, 2015
Submitted by: Jianqiang Gao.
DOI: http://dx.doi.org/10.18642/ijamml_7100121449

Abstract

Combined with the shared similarity among multiple source images and depth of field in a camera, a new image fusion algorithm based on nonsubsampled shearlet transform (NSST) domain and pulse coupled neural network (PCNN) is proposed. First, NSST is utilized for decomposition of the source images, the low frequency coefficients are fused by weight votes in the structure-driven regions by using shared similarity and depth of field (SSSID), and the high frequency coefficients are fused by fired map of PCNN which motivated by larger sum-modified-Laplacian (SML) based on SSSID, finally, the fusion image is gained by inverse NSST. The algorithm can both preserve the information of the source images well and suppress pixel distortion due to nonlinear operations in transform domain. Experimental results demonstrate that the proposed method outperforms the state-of-the-art transform domain and PCNN fusion methods.

Keywords

image fusion, NSST, self-similarity, depth information, PCNN.