63 
COMPARISON OF MORPHOLOGICAL PYRAMID AND LAPLACIAN 
PYRAMID TECHNIQUES FOR FUSING DIFFERENT FOCUSING IMAGES 
Jia Yonghong 3 * ,Fu Xiujun b , Yu Hongwei c 
School of Remote Sensing and Information Engineering , Wuhan University, Wuhan, China - yhjia2000@sina.com 
b Hubei Qingjiang Supervision Co. LTD, Bubei, China -hljyhw@vip.163.com 
'Heilongjiang Bureau of Surveying and Mapping,Heilongjian, Harbin , China -fuxiujun361@hotmail.com 
Commission VII, WG VII/6 
KEY WORDS: Fusion, Image, Multiresolution, Method, Morphology, Laplacian pyramid 
ABSTRACT: 
Image fusion attempts to combine complementary information from multiple images of the same scene, so that the resultant image is 
more suitable for human visual perception and computer processing tasks such as segmentation, feature extraction and object 
recognition. In this study, an efficient fusion method based on morphological pyramid decomposition is studied. As a non-linear 
filter, the morphological filter outperforms the linear Gaussian filter in precise edge localization, and better represents the detailed 
spatial information of images. Several source images with different focuses of the same scene are taken and processed with the 
fusion methods based on the morphological pyramid and the Laplacian pyramid. Experimental results show that the proposed 
method based on the morphological pyramid is superior to the Laplacian pyramid fusion method in both visual and objective 
evaluation. 
1. INTRODUCTION 
Optical lenses suffer from the problem of limited depth of field. 
According to the lens formula, only objects at one particular 
depth will be truly in focus. Consequently, if one object in the 
scene is in focus, another object at a different distance from the 
lens will be out of focus and thus blurred. The degree of this 
blurring is affected by a number of factors, including the object 
distance, the focal length and f-number of the lens, and the 
distance between the lens and the sensor plane [Pham 1999]. 
For the same object scene, two or more pictures (source images) 
are taken from the same camera position but with different 
focuses, such that each relevant object is in focus in at least one 
of them. How are all objects in an image in good focus from 
these pictures? 
Image fusion is a process of combining images of the same 
scene to form a composite image. The composite image is 
formed to improve image content and to make it easier for the 
user to detect, recognize, and identify targets and increase his 
situational awareness. We can follow an approach according to 
the image fusion. Multiresolution techniques have proven to be 
very useful for image fusion [Piella, 2002]. The algorithms 
based on multiresolution techniques found in the literature are 
pixel-based and region-based approaches, and region-based 
approaches may help to circumvent some of the well-known 
drawbacks of pixel-based techniques, such as blurring effects, 
high sensitivity to noise and misregistration. In this study a 
recent and efficient technique of fusion based on morphological 
Pyramid was attempted and its efficiency was compared with 
1 at of Laplacian pyramid fusion techniques. Accordingly, 
several source images with different focuses of the same scene 
are taken and processed with the morphological pyramid and 
the Laplacian pyramid fusion technique. The outputs were 
evaluated using visual comparison, statistical correlation and 
entropy. Compared with the Laplacian pyramid technique, the 
morphological pyramid techniques proved to be a better option 
since it preserved most of the spatial information and improved 
information content. 
2. METHODOLOGY 
2.1 Laplacian pyramid (LP) fusion 
The Laplacian pyramid is derived from the Gaussian pyramid 
(GP), which is a multi-scale representation obtained through a 
recursive reduction (low-pass filtering and decimation) 
[Burt, 1983]. 
Let Go(i, j), i=0,... ,M-1 ,and j=0,... ,N-1 ,be a grey scale 
image. The octave GP is defined as 
G*0‘J) =reduce 2 [G*.i] (Uj) (1) 
for £=1,...,K, in which k identifies the level of the pyramid, K 
being the top. 
From the GP, the enhanced LP (ELP) is defined, for k=\,..., K, 
as 
Lk(i,j) = G*0 J)-expand 2 [G* +1 ] (ij) (2) 
The project supported by the State Surveying and Mapping Fund of China.yhjia2000@sina.com