Page 1 of 8

Journal for Studies in Management and Planning

Available at

http://edupediapublications.org/journals/index.php/JSMaP/

e-ISSN: 2395-0463

Volume 02 Issue 12

December 2016

Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 23

Perceptual evaluation of the multi-exposure quality of the

image Fusion

1

tummala Santosh Kumar,2 Mr.P.V.Vara Prasad Rao

1PG Scholar, Department of ECE, SLC's Institute of Engineering and Technology, Piglipur Village, Hayathnagar

Mandal, Near Ramoji Film City, Ranga Reddy District, Hyderabad, Telangana

2Assosciate Professor, Department of ECE, SLC's Institute of Engineering and Technology, Piglipur Village, Hayathnagar

Mandal, Near Ramoji Film City, Ranga Reddy District, Hyderabad, Telangana

Abstract— Multi-exposure image fusion

(MEF) is considered an effective quality

enhancement technique widely adopted in

consumer electronics, but little work has

been dedicated to the perceptual quality

assessment of multi-exposure fused images.

In this paper, we first build an MEF

database and carry out a subjective user

study to evaluate the quality of images

generated by different MEF algorithms.

There are several useful findings. First,

considerable agreement has been observed

among human subjects on the quality of

MEF images. Second, no single state-of-the- art MEF algorithm produces the best quality

for all test images. Third, the existing

objective quality models for general image

fusion are very limited in predicting

perceived quality of MEF images. Motivated

by the lack of appropriate objective models,

we propose a novel objective image quality

assessment (IQA) algorithm for MEF

images based on the principle of the

structural similarity approach and a novel

measure of patch structural consistency.

Our experimental results on the subjective

database show that the proposed model well

correlates with subjective judgments and

significantly outperforms the existing IQA

models for general image fusion. Finally, we

demonstrate the potential application of the

proposed model by automatically tuning the

parameters of MEF algorithms.

.I. INTRODUCTION

MULTI-EXPOSURE image fusion (MEF) is

considered an effective quality enhancement

technique that is widely adopted in

consumer electronics .MEF takes a sequence

of images with different exposure levels as

inputs and synthesizes an output image that

is more informative and perceptually

appealing than any of the input images.

MEF fills the gap between high dynamic

range (HDR) natural scenes and low

Page 2 of 8

Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 24

dynamic range (LDR) pictures captured by

normal digital cameras. Comparing with

typical HDR imaging techniques which first

construct an HDR image from the source

sequence and then tone-map it to an LDR

image, MEF bypasses the intermediate HDR

image construction step and directly yields

an LDR image that can be displayed on

standard viewing devices.

Since first introduced in 1980’s MEF has

been an active research topic and attracted

an increasing amount of attention in recent

years . With many MEF algorithms at hand,

it becomes pivotal to compare their

performance, so as to find the best algorithm

as well as directions for further

advancement. Because the human visual

system (HVS) is the ultimate receiver in

most applications, subjective evaluation is a

straightforward and reliable approach to

evaluate the quality of fused images

Although expensive and time consuming a

comprehensive subjective user study has

several benefits. First, it provides useful data

to study human behaviors in evaluating

perceived quality of fused images. Second, it

supplies a test set to evaluate and compare

the relative performance of classical and

state-of-the-art MEF algorithms. Third, it is

useful to validate and compare the

performance of existing objective image

quality assessment (IQA) models in

predicting the perceptual quality of fused

images. This will in turn provide insights on

potential ways to improve them. Over the

past decade, substantial effort has been

made to develop objective IQA models for

image fusion applications Most of them are

designed for generalpurpose image fusion

applications, not specifically for MEF, and

some of them can only work with the case of

two input images. Furthermore, little has

been done to compare them

with (or calibrate against) subjective data

that contains a wide variety of source

sequences and MEF algorithms. In this

work, we aim to tackle the problem of

perceptual quality assessment of MEF

images. We build one of the first databases

dedicated to subjective evaluation of MEF

images. The database contains 17 source

sequences with multiple exposure levels (≥

3) and the fused images generated by 8

classical and state-of-the-art MEF

algorithms. Based on the database, we carry

out a subjective user study to evaluate and

compare the quality of the fused images. We

observe considerable agreement between

human subjects, and not a single MEF

algorithm produces the best quality for all

test images. More importantly, we find that

existing objective quality models for general

Page 3 of 8

Available online: http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 25

image fusion are very limited in predicting

perceived quality of MEF images. This

motivates us to develop a novel objective

IQA model for MEF images. Our model is

inspired by the structural similarity (SSIM)

index whose philosophy is that the HVS is

highly adapted for extracting structural

information from natural scenes. To

compare the structures of multiple patches

from different exposures, we introduce a

novel measure of patch structural

consistency. Furthermore, to balance

between finer-scale detail preservation and

coarserscale luminance we adopt a multi- scale approach where with the scale shifting

from fine to coarse, SSIM-based structural

comparison captures image distortions from

fine details to large-scale luminance

variations. Experimental results show that

the proposed model well correlates with

subjective judgments and significantly

outperforms existing objective IQA models

for general image fusion. The value of

objective models are beyond measuring and

comparing MEF images and algorithms.

II. RELATED WORK

The problem of MEF can be generally

formulated as Y(i ) = K _ k=1 Wk (i )Xk(i ),

(1) where K is the number of multi-exposure

input images in the source sequence, Xk (i )

and Wk (i ) represent the luminance value

(or the coefficient amplitude in the

transform domain) and the weight at the i -th

pixel in the k-th exposure image,

respectively. Y denotes the fused image. The

weight factor Wk (i ) is often spatially

adaptive and bears information regarding the

relative structural detail and perceptual

importance at different exposure levels.

Depending on the specific models for

structural information and perceptual

importance, MEF algorithms differ in the

computation of Wk . A significant number

of MEF algorithms have been proposed,

ranging from simple weighted averaging to

sophisticated methods based on advanced

statistical image models. Local and global

energy weighting approaches are the

simplest ones, which employ the local or

global energy in the image to determine Wk

. Dated back to 1984, Burt first employed

Laplacian pyramid decomposition for

binocularimage fusion. Later in 1994, Burt

and Kolczynski applied thisdecomposition

to MEF, where they selected the local

energy of pyramid coefficients and the

correlation between pyramids within the

neighborhood to compute Wk . Laplacian

pyramid turns out to be an effective scheme

in image fusion to avoid unnatural

appearance and unwanted artifacts