ISSN (Online) 2581-9429

IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Impact Factor: 7.301 Volume 3, Issue 3, April 2023

Study on Underwater Image Processing Techniques

Prerana Agrawal1 and Dr. Pramod B. Bhalerao2
Student, Department of Computer Science and Engineering1
Guide, Department of Computer Science and Engineering2
Deogiri Institute of Engineering & Management Studies, Aurangabad, Maharashtra state, India

Abstract: Image processing is extensively used to improve the quality of photos with scattering and poor
contrast. Image processing is used in many fields of computer vision, including underwater image
processing. It is necessary to process underwater photographs in order to improve the poor image quality
caused by light scattering and refraction. As depth deepens, different colors are absorbed by the adjacent
medium specific wavelengths. This research describes an effective approach for enhancing underwater
photographs that have been deteriorated owing to medium absorption and scattering. Image processing has
been used to develop certain approaches for improving the quality of underwater photographs. This study
also discusses some of the algorithms for enhancing the quality and resolution of photographs. Using an
edge detection robustness criterion performance of filtering will be evaluated and also the color correction.

Keywords: Image quality, filtering, underwater image enhancement, contrast

I. INTRODUCTION
Underwater photography is widely used to observemarine resources and to observe the underwater infrastructure. Now
a day underwater image processing has very vast importance in computer vision. The images captured underwater are
suffered from so many disturbances. Scattering, refraction of light, and absorption are three major underwater image
impairments, and these impairments are responsible for low brightness, low contrast, and introduction of noise. As the
light travels deep into the water, the intensity of light gets poor, and so that the image gets affected by poor visibility
and contrast.
Visible red light contains considerably greater energy over invisible infrared radiation and is much more quickly
absorbed by water than some other visible wavelengths, according to research. When it reaches the depth of the water
level, it exhibits the impact of light color. At 20 meters, it reflects the color of the substances in the water, making
redfish seem nearly black. Longer wavelength light that is absorbed more rapidly than short wavelength light. Higher-
energy light having short - wavelength, like blue, can dive deeper as a result. Figure 1 represents the effect on light in
deep water level.

Figure 1. Effect of light color in deep water level.

With the advancement of image processing technology, the quality of underwater images is being improved. Picture
enhancement improves image quality while also helping to improve the overall appearance of the image. There are so
Copyright to IJARSCT DOI: 10.48175/IJARSCT-9206 276
www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Impact Factor: 7.301 Volume 3, Issue 3, April 2023

many algorithms are proposed and implemented on underwater image processing. Image processing is mainly
introduced in two ways is Image enhancement and Image restoration technique.
Throughout the image restoration process, the image enhancement approach increases the information quantity and
quality of the original data, as well as the noisy and fuzzy image. Marine biology can benefit greatly from underwater
picture processing. The focus of this research is on several picture enhancing approaches in underwater image
processing.

II. LITERATURE SURVEY

There are so many algorithms and techniques in underwater image processing to enhance underwater image quality.
Some methods are used to strengthen underwater images, such as the white balance method, color correction
algorithms, and contrast enhancement [1]. Underwater images are affected mainly by non-uniform lighting, so some
processes correct the non-uniform lightening effect [2]. As the seawater level's depth increases, the underwater light
attenuation leads to unusual color change, depending on the organic compound dissolved in water and on the
wavelength [3].
As they move further into the sea, the underwater sights usually become blue and greenish. As the light intensity drops,
underwater photographs are subjected to issues such as noise, distortion, poor contrast, and low visibility. Some
algorithms aid in the restoration of color, information loss, and difference [4, 5]. Image enhancing technologies are
divided into two categories: spatial domain and frequency domain. All image enhancing procedures in the frequency
domain approach are done on the Fourier transform. The image's Fourier transform is computed first, followed by the
inverse Fourier transform of the resulting picture. The pixel values of pictures are operated in the spatial domain
manner [6].
Some of the image processing methods utilized for underwater picture improvement include white balance, color
correction, and contrast enhancement [7].The contrast enhancement approach involves histogram equalization [8]. This
approach is based on color correction and is based on the retinex model underwater image improvement method; this
process applies a different strategy to strengthen the reflection and lighting components of the underwater picture [9],
and the results are integrated. A color correction approach is created based on the automated color equalization model
[10]. Unsupervised color correction approach underwater picture improvement based on color equalization and contrast
correction is suggested [11], which includes Histogram equalization. For underwater pictures, color casting is created
by the various absorption capacity for different wavelengths; the backward reflecting effect reduces the contrast in
images. The picture characteristics are often blurred as a result of the forward scattering phenomenon [12].
After a distance of 4-5meters, the clarity of underwater pictures rapidly declines as the depth of the water will increase.
Researchers have created a variety of hardware platforms including cameras [13, 14], which are currently being used in
a variety of industries. The majority of the approaches suggest modifying grayscale picture quality measures, while a
few apply by transforming a color image to a grayscale image [15] or combining the measure values with various
weights by assessing the quality in each color component separately[16]. The fundamental purpose of image processing
is to minimize noise and improve quality [17]. a noise removal method based on a standard median filter has been
proposed. A new technique provided a revolutionary denoising method and increasing nearly uniform illumination
homomorphic filtering to reduce the additive noise found in underwater pictures [18]. The laser imaging modality is
among the most modern imaging technologies for processing underwater photographs; it works by picking reflected
light out from object.

III. METHODOLOGY
Underwater image processing has many applications, such as underwater archeology, ocean engineering, and
underwater photography; therefore, it is necessary to have quality in an underwater scene's photos. The literature survey
focuses on underwater imaging enhancement, image analysis , and recent work on machine learning approach [20], as
stated by several authors.Generally, the medium of water and the particles which are suspended in the water have some
properties like scattering and absorption effect on the light. Because of these, some issues are there, like distortion and
low visibility of images underwater.

Copyright to IJARSCT DOI: 10.48175/IJARSCT-9206 277

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Impact Factor: 7.301 Volume 3, Issue 3, April 2023

One of the methods, which is dependent on local and global level histogram equalization and contrast enhancement
techniques, is beneficial for picture improvement. The Retinex-inspired approach is used to choose a picture first. To
produce the contrast-enhanced image, several color modifications are made using global equalization of the histogram,
followed by local equalization. Then, using a dual image multi-scale fusion approach [21], these two pictures are
fused.To evaluate the algorithm results, it is equally important to have some of the underwater image processing
datasets. Fish4Knowlege dataset is one of the databases for the detection and recognition of images. SUN is another
database for scene recognition and object detection. For autonomous marine robotics,the MARIS dataset is used. In the
sea-thru dataset, it contains 1100 underwater images are present [22].

IV. CONCLUDING REMARK

This work providesvarious underwater image processing techniques for image enhancement. All methods have given
proper justification for the strategies they are using to improve the image enhancement up to a better level. All the
researchers have addressed the issue related to uneven light distribution in the water. Various authors have worked on
underwater image processing for enhancing the image quality for the same they have given multiple algorithms.
However, still, because of the light effect, we did not get enough accuracy. Work done using local and global
equalization of histogram and duel image multi-scale fusion could notachieve consistency in the background color of
images taken underwater. Future work will focus on the effect of turbidity in underwater image processing at different
levels.

REFERENCES
[1]. E. Y. Lam, "Combining gray world and retinex theory for automatic white balance in digital photography,"
in Proceedings of the Ninth International Symposium on Consumer Electronics,2005. (ISCE 2005), pp. 134–
139, Macau, Macau, June 2005.
[2]. R. Garcia, T. Nicosevici and X. Cufi, "On the way to solve lighting problems in underwater imaging," IEEE
OCEAN 02, 1018-1024, October 2002.
[3]. S. Q. Duntley, "Light in the sea,"J. Opt. Soc. Amer., vol. 53, no. 2,pp. 214–233, 1963
[4]. Iqbal, K., Odetayo, M., James, A., Salam, R.A. and Talib, A.Z., 2010. Enhancing the low quality images
usingunsupervised color correction method. International Conference on System Man and Cybernatics
(SMC), Istanbul, 10-13 October 2010, pp.1703-1709.
[5]. Iqbal, K., Salam, R.A., Osman, A. and Talib, A.Z., 2007. Underwater image enhancement using integrated
color model.IAENG International Journal of Computer Science, 34(2), pp.529-534.
[6]. Bhabatosh Chanda and Dwijest Dutta Majumder, 2002,Digital Image Processing and Analysis.
[7]. E. Y. Lam, "Combining gray world and retinex theory for automatic white balance in digital photography,"
in Proceedings ofthe Ninth International Symposium on Consumer Electronics, 2005. (ISCE 2005), pp. 134–
139, Macau, Macau, June 2005.
[8]. R. Hummel, "Image enhancement by histogram transformation," Computer Graphics and Image Processing,
vol. 6, no. 2,pp. 184–195, 1977.
[9]. X. Fu, P. Zhuang, Y. Huang, Y. Liao, X.-P. Zhang, and X. Ding, "A retinex-based enhancing approach for
single underwaterimage," in 2014 IEEE International Conference on Image Processing(ICIP), pp. 4572–
4576, Paris, France, Oct. 2014.
[10]. A. Rizzi, C. Gatta, and D. Marini, "A new algorithm for unsupervised global and local color
correction,"Pattern Recognit. Lett.24(11),1663–1677 (2003). 14. L. A. Torres-Méndez
[11]. K. Iqbal et al., "Enhancing the low quality images using unsupervisedcolour correction method," in IEEE Int.
Conf. on Systems Man andCybernetics, pp. 1703–1709 (2010).
[12]. S. Raimondo and C. Silvia, "Underwater image processing: State of the art of restoration and image
enhancement methods,"EURASIP J.Adv. Signal Process., vol. 2010, 2010, DOI: 10.1155/2010/746052.
[13]. B. J. Boom et al., "A research tool for long-term and continuous analysis of fish assemblage in coral-reefs
using underwater camera footage,"Ecol. Informat., vol. 23, no. 1, pp. 83–97, Sep. 2014.

Copyright to IJARSCT DOI: 10.48175/IJARSCT-9206 278

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Impact Factor: 7.301 Volume 3, Issue 3, April 2023

[14]. T. Fukubaet al., "A new drifting underwater camera system for observing spawning Japanese eels in the
epipelagic zone along theWest Mariana Ridge,"Fisheries Sci., vol. 81, no. 2, pp. 235–246, Mar. 2015.
[15]. Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image quality assessment: From error visibility
to structural similarity,"IEEETrans. Image Process., vol. 13, no. 4, pp. 600–612, Apr. 2004.
[16]. C. Gao, K. Panetta, and S. Agaian, "A new color contrast enhancement algorithm for robotic applications," in
Proc. IEEE Conf. Technol. Practical Robot Appl. (TePRA), Apr. 2012, pp. 42–47.
[17]. C. C. Chang, J. Y. Hsiao, and C. P. Hsieh, "An AdaptiveMedian Filter for Image Denoising," in 2008 Second
International Symposium on Intelligent Information Technology Application,pp. 346–350, Shanghai, China,
Dec. 2008.
[18]. C. J. Prabhakar and P. U. P. Kumar, "Underwater imagedenoising using adaptive wavelet subband
thresholding," in2010 International Conference on Signal and Image Processing,pp. 322–327, Chennai,
India, December 2010.
[19]. KocakD, DalgleishF, CaimiF, SchechnerY(2008) A focus on recent developments and trends in underwater
imaging. Marine Technology Society Journal, 42(1):52-67.
[20]. Y. Wang, J. Zhang, Y. Cao, Z. Wang, A deep CNN method for underwater image enhancement, in: IEEE
International Conference on Image Processing, 2017, pp. 1382–1386.
[21]. L. Bai, W. Zhang, X. Pan and C. Zhao, "Underwater Image Enhancement Based on Global and Local
Equalization of Histogram and Dual-Image Multi-Scale Fusion," in IEEE Access, vol. 8, pp. 128973 128990,
2020,

Copyright to IJARSCT DOI: 10.48175/IJARSCT-9206 279

www.ijarsct.co.in