Research of Multi-focus Image Fusion Algorithm Based on Sparse Representation and Orthogonal Matching Pursuit - Advances in Image and Graphics Technologies - page 57

Research of Multi-focus Image Fusion Algorithm
Based on Sparse Representation and
Orthogonal Matching Pursuit

Li Xuejun 1,2 and Wang Minghui 1

1 Sichuan University, College of Computer Science, Chengdu, 610065
2 Southwest University of Science and Technology, Mianyang, 621010, China
[e-mail], [e-mail]

Abstract. Due to the unideal effects of those common multi-source focus image
fusion algorithms, in this essay we propose a multi-focus image fusion
algorithm based on sparse representation and orthogonal matching pursuit
(OMP), and demonstrate the results of the corresponding multi-source focus
image fusion experiments by MATLAB. Compared with the fused images of
the above several common algorithms by evaluating subjectively and
objectively, the results suggest that the multi-focus image fusion algorithm
based on sparse representation and orthogonal matching pursuit (OMP) present
higher mutual information, minimum distorted values and higher Q ab / f values
which indicate that the fused image by this algorithm can obtain more image
information with a smaller distortion from the original (image?), so as to get a
better image but cost much more time.

Keywords: Multi-focus image fusion, sparse representation, orthogonal
matching pursuit and performance evaluation.

1 Introduction

The clarity of optical lens imaging relies on its depth of field in an optical imaging
system. If the object distance of a target in the scene goes beyond the depth of field to
the optical lens, the image will be fuzzy, whereas a clear one can be assured. In
reality, the targets in the same scene produce different clarities on the impact of
external factors such as the distance and the light intensity when imaging. Therefore,
it is quite difficult to make an accurate and comprehensive interpretation from the
information of a single image in the same scene. Multi-focus image fusion technique
is used to deal with two or more images (abandoning the fuzzy parts and keeping the
clear parts), which are shot by different focus objects with different depths of field in
the same scene. It will finally form a new image to be perceived more suitably and
processed more easily. Multi-focus image fusion is a research branch of multi-source
image fusion [1].
Suppose that there are K images

{}

K

which describe the same scene with
subjects in different depths of field, but focus on different objects respectively, the
problem to be solved by multi-focus image fusion is how to recover the image F that

I

i

i1 =