Sei sulla pagina 1di 3

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Under water Image Enhancement Using MATLAB


Asistant Prof.Kala H S, Mahesh N, Abhilash Nair N S, Ranjith C, Kantharaju P R
MVJ COLLEGE OF ENGINEERING, Dept. of ECE, Bengaluru 560067, India
abhilashnair409@gmail.comnmahesh846@gmail.com, kantharajupr@gmail.com, cranjith097@gmail.com
intensity values of RGB and Ycbcr almost lies in the same
range. Hence it is difficult to extract interested pixels, that is
Abstract—The study of underwater is very important to find
vehicles pixels. Though intensity values of and can be
the unfound many algorithms have been derived to enhance an
underwater image but no method gives a clear vision of human distinguished in Y, S, V but Y and V values cannot be
eye. Many methods found suffers a disadvantage that the clarity considered as these values will varies with climatic change
reduces as the depth of the water increases this is due to the and lightning conditions
fading of colours in underwater. To solve this problem, we get
into a non-reference underwater image enhancement method.
The three main attributes of an image are its colour contrast Underwater images are used to survey the ocean floor, much
and sharpness. In this method we modify the attributes of the often with acoustic sensors for their capability of remote sensing.
image step by step to attain a clear image. The results of this
experiment demonstrate clearly that this method produces an Optical sensors have been introduced into these vehicles and the
output image with a high information the measures are applied use of video is well integrated by the underwater community for
to underwater images to show their importance in practical short range operation.
application. This project solves this problem my a simple and
easy methodology with changing the three main attributes of an Over the last five years, many methods have been proposed to
image thus enhancing the information of the image. overcome traditional underwater imaging problems. This paper
Index Terms— Absorption, Scattering, Clarity,
Enhancement, Algorithms, Underwater image, Human eye,
aims to review the state-of-the-art techniques in underwater
Attributes, methodology. image processing by highlighting the contributions and
challenges presented in over 40 papers. We present an
I. INTRODUCTION overview of various underwater image-processing approaches,
Image processing is the field of signal processing where both such as underwater image de-scattering, underwater image
the input and output signals are images. Images can be colour restoration, and underwater image quality assessments.
thought of as two-dimensional signals via matrix
representation, and image processing can be understood as
applying standard one dimensional signal processing
techniques to two-dimensional signals. In this paper, we
introduce a new spatial domain color contrast enhancement
algorithm based on the alpha weighted quadratic filter. The
goal of this work is to utilize the characteristics of the
nonlinear filter to enhance contrast while recovering the color
information. The goal of color contrast enhancement is to
recover the color captured from image scenes and enhance
the contrast similar to the human visual system. Several color
contrast enhancement techniques have been proposed both in
the spatial domain and the transform domain.
The image processing technique we will be implementing
will be uiccs. As the board we have does not support a direct
connection for the input image, we will use MATLAB to
output the image as a matrix and store it in the data memory
of the DSP. To do this, we will use the parallel port
connection to get our input data into the board, scattering Fig4: Sample input image
always causes a blurring effect in underwater photography;
this rarely occurs in land photography. Second, wavelength Literature Survey
absorption usually causes a colour reduction in the captured
images, which rarely occurs in air. Third, except for • C.Gao, K.Panetta, and S.Agaian [1] Introduce a new
electronic noise, the sediments in the water also affect high spatial domain color contrast enhancement algorithm
dimensional imaging. Another problem occurs because based on the alpha weighted quadratic filter. The goal of
artificial lighting is widely used for underwater photography, this work is to utilize the characteristics of the nonlinear
and this non-uniform lighting causes vignetting in captured filter to enhance contrast while recovering the color
images. Image processing includes many techniques. information. [“A new COLOUR contrast enhancement
Intensity values of particular image is used to compare algorithm for robotic Applications”]
between RGB and Ycbcr is converted into R, G, B, Y, Cb, • S-D.Chen and A.R. Ramli [2] Introduces Histogram
Cr, H, S, V. Where R, G, B, Cb, Cr, H fails to distinguish Equalization (HE) is widely used for contrast
between original image and after image enhancement as the

Volume IX, Issue VI, JUNE/2019 Page No: 1948


International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

enhancement. However, it tends to change the 2. YCbCr COLOR MODEL


brightness of an image and hence, not suitable for
consumer electronic products, where preserving the The difference between YCbCr and RGB is that YCbCr
original brightness is essential to avoid annoying represent color as brightness and two color difference signals,
artifacts. Bi-histogram equalization (BBHE) has been while RGB represents color as red, green and blue.
proposed and analyzed mathematically that it can
preserve the original brightness to a certain RGB TO YCbCr CONVERSION
extends.[“Minimum mean brightness error bi-histogram
equalization in contrast enhancement” ]
• Z.Wang,A.C.Bovik,H.R.Sheik and E.P. Simoncelli
[3],has given a Objective methods for assessing
perceptual image quality traditionally attempted to
quantify the visibility of errors (differences) between a
distorted image and a reference image using a variety of
known properties of the human visual system. Under the
assumption that human visual perception is highly
adapted for extracting structural information from a
scene.[“Image quality assessment: From error visibility
to structural similarity”]

II. COLOR PROCESSING IV. PROPOSED BLOCK DIAGRAM


1. RGB COLOR MODEL
Input image Pre-processing Local Colour
Working directly on RGB color model often results in stretching transfer
error, as the color properties of the object of interest
changes based on the brightness.
However in few cases, working individually on Post Hist equa pyramids
Colour transfer
Red/Green/Blue image yields good results. Also, working processing
on individual color is better due to memory efficiency.
Each color requires 8bits/pxl, where as RGB pxl requires
24bits Output performance
image

Figure2shows the proposed system architecture to enhance the


underwater image. Underwater image is considered as an in put
from the database. The remove noise, different types of filters are
applied, here Gaussian and isotropic filters are applied. The
filtered image contrast is enhanced to increase the intensity value.
The RGB colour image is converted to CyBC pane. „component
from theYCbCr is considered for applying pyramid. Gaussian and
Laplacian pyramid are applied to luminance component. The
histogram equalization is applied using CLAHE.
The YCbCr colour plane is converted to RGB colour plane. In the
post processing the image sharpening is applied to sharpen the
image. Capturing images underwater is difficult, due to
attenuation caused by light that is reflected from a surface and is
deflected and scattered by
particles, and absorption substantially reduces the light energy.
The random attenuation of the light is mainly cause of the haze
appearance while the fraction of the light scattered back from the
water along the sight considerable degrades the scene contrast. In
particular ,the objects distance of more than 10meters are almost
in distinguishable while the colours are faded due to the
characteristic wavelengths are cut according
to the water depth.

III PROGRAM USED FOR IMAGE RESTORATION clear all;


clc; warning off all;
close all; Repeat = 1;

Volume IX, Issue VI, JUNE/2019 Page No: 1949


International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

while(Repeat) We use the DLSR CAMERA to capture the underwater image


choice = menu('Underwater Image and to clear the noises we uses the MATLAB Software to
Enhancement','selectQuery','Img Enhancement',... remove the noise present in the original image
'Exit');
V. CONCLUSION
In this paper, Each attribute measure can be used separately
switch choice for specific underwater image processing tasks. Several
properties, such as luminance and contrast masking, the
colour perception property, and relative contrast sensitivity,
case 1 are incorporated in the formulations of the measures.
disp('selectQuery'); Therefore, comparing with other quality measures used in
FiltImg = selectQuery; existing underwater image processing algorithms This
method is a simple method which gives 70 percent of
information from the raw image only by adjustments of
case 2 colour, sharpness and contrast. A first-of-kind underwater
disp('imgenhancement'); colour image quality evaluation metric is proposed.
encImg=ImgEnhancement(FiltImg);
VI. RESULT
case 3
close all; The results indicate that the proposed metric has fast
processing time, which makes it applicable for real-time
clear all;
image processing. It is able to successfully predict the
clc; relative distortion with similar scenes and the difference
Repeat = 0WA between enhancement results. The measure helps to classify
when an image is underexposed. Due to the simplicity and
end
efficiency,
end The measure helps to classify when an image is
underexposed. Due to the simplicity and efficiency, The
ultimate goal behind the MMBEBHE is to allows maximum
level of brightness preservation in Bi-Histogram
Equalization to avoid unplesant artifacts and unnatural
enhancement due to excessive equalization while enhancing
the constrast of a given image as much as possible. This
paper has also formulated an efficient,

REFERENCES
[1] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli
“Image quality assessment: From error visibility to
structural similarity,” IEEE Trans. Image Process., vol.
13, no. 4, pp. 600–612, Apr. 2004.

[2] S.-D. Chen and A. R. Ramli, “Minimum mean brightness


error bi-histogram equalization in contrast enhancement,”
IEEE Trans. Consum. Electron., vol. 49, no. 4, pp. 1310–
1319, Nov. 2014.

[3]C. Gao, K. Panetta, and S. Agaian, “A new COLOUR


contrast enhancement algorithm for robotic Applications,” in
Proc. IEEE Conf. Technol. Practical Robot Appl. (TePRA),
pp. 42–47 Apr, 2012.

Fig 7: Hardware and software implementation of


underwater image restoration

Volume IX, Issue VI, JUNE/2019 Page No: 1950

Potrebbero piacerti anche