Multispectral and panchromatic image fusion using chaotic Bee Colony optimization in NSST domain

Affiliation of Author(s):电子信息工程学院

Journal:Yaogan Xuebao/J. Remote Sens.

Abstract:The rapid development of remote sensing technology provides an effective technical approach for humans to understand living environments and to utilize natural resources. Various remote sensing sensors exist, and the images formed by different image sensors have various characteristics, thereby resulting in multi-source remote sensing images, such as multi-spectral and panchromatic images. Fusing the multi-source remote sensing images of the same scene is necessary to efficiently and comprehensively deal with these image data. The multispectral image has high spectral resolution and rich spectral information; however, the spatial resolution of this image is low due to the limitations of physical devices. The panchromatic image has high spatial resolution and clear spatial detail; however, its spectral resolution is low. The fusion of multi-spectral and panchromatic images is the integration of the spatial detail information of the panchromatic image into the multi-spectral image to generate an image with high spatial resolution and spectral resolution, which benefit subsequent image processing. A method for the fusion of multi-spectral and panchromatic images using chaotic artificial bee colony optimization and improved Pulse Coupled Neural Network (PCNN) in a Non-Subsampled Shearlet Transform (NSST) domain is proposed. First, Intensity Hue Saturation (IHS) transform is performed on the multi-spectral image. The histogram of the panchromatic image is matched to the histogram of the intensity component of the multi-spectral image. The intensity component of the multispectral image and the new panchromatic image are then decomposed by NSST. Next, the low-frequency component is fused with the improved weighted fusion algorithm. Recently, artificial bee colony algorithm is one of the effective swarm intelligence optimization algorithms, which can adaptively determine the weighted coefficient. Chaotic bee colony optimization algorithm is designed by introducing the tent mapping chaotic sequence to avoid premature phenomena. The chaotic bee colony optimization algorithm has high convergence precision and rapid convergence speed in global optimization. By exploiting this property, the optimal improved weighted coefficient is determined by the chaotic artificial bee colony optimization algorithm. Mutual information is used as the fitness function. The improved PCNN method is adopted for the fusion of high-frequency components. Finally, the fused image is obtained by inverse NSST and inverse IHS transform. Many multi-spectral and panchromatic remote sensing images from LANDSAT TM, IKONOS, and SPOT 4 satellites are tested. Qualitative and quantitative evaluation results are obtained to verify the feasibility and effectiveness of the proposed method. The proposed method outperforms five other kinds of fusion methods: the IHS method, the method of Non-Subsampled Contourlet Transform (NSCT) combined with Non-negative Matrix Factorization (NMF),the method of NSCT combined with PCNN in the subjective visual effect, and the objective quantitative evaluation indexes, such as information entropy and spectral distortion. The proposed method can effectively preserve the spectral information of the multispectral image, while the details of panchromatic images are injected into the fused image as much as possible, effectively improving the spatial resolution of the fused image. © 2017, Science Press. All right reserved.

ISSN No.:1007-4619

Translation or Not:no

Date of Publication:2017-07-25

Co-author:王志来

Correspondence Author:wyq

Date of Publication:2017-07-25