477

1 INTRODUCTION

Remotely Operated Vehicles (ROVs) are tethered to

and controlled from surface vessels, enabling the

exploration and monitoring of underwater

environments without the physical presence of human

operators within the vehicle [13]. The application of

ROVs in the technical inspection of submerged

components of marine and coastal infrastructure is of

significant importance. By means of video recording,

digital imaging, technical data acquisition, and seabed

topographic surveys, ROVs facilitate the collection of

critical information necessary for the design,

construction, maintenance, and rehabilitation of these

structures. Routine inspections are essential to

ensuring the operational integrity and longevity of any

underwater infrastructure [12]. In particular, the

inspection of port infrastructure plays a pivotal role in

maintaining safe, efficient, and sustainable port

operations.

In the context of Polish seaports, the frequency and

scope of quay inspections are governed by port

regulations established by the directors of maritime

offices with jurisdiction over specific coastal areas.

These regulations stipulate the obligations of quay

operators regarding the upkeep and monitoring of port

infrastructure. Specifically, operators are required to

conduct periodic hydrographic soundings and seabed

cleanliness assessments in the zones adjacent to quays

and port basins.

The utilization of ROVs in underwater surveying

and inspection is a relatively recent advancement,

which contributes to the limited volume of academic

and technical literature addressing this methodology

[18]. The most pertinent publication in this domain is

[19], originating from Lithuania, which details the

deployment of a domestically developed ROV

comparable to the one employed in the present study.

However, the Lithuanian research primarily focused

Use of an ROV with Modulated Lighting for Diagnosing

the Technical Condition of Submerged Port Wharf

Structures

A. Kaizer, B. Lednicka, A. Tessmer, W. Freda, J. Soszyńska- Budny & E. Ziajka

Gdynia Maritime University, Gdynia, Poland

ABSTRACT: The technical condition of underwater port structures is critical to the safety and longevity of

maritime infrastructure. Traditional inspection techniques, such as diver-assisted surveys, are constrained by

safety risks, limited duration, and reduced efficiency in turbid waters. This study explores the application of a

remotely operated vehicle (ROV) equipped with modulated lighting systems to enhance visibility and facilitate

high-resolution imaging in optically complex underwater environments. The experiment employed a CHASING

M2 ROV, modified with red, green, and white lighting configurations, to inspect the quay wall of a port. The

impact of lighting color on image quality was evaluated. Results indicate that modulated lighting tailored to the

optical properties of turbid coastal waters can improve image contrast and facilitate defect detection. The findings

highlight the potential for advanced ROV systems to augment underwater inspection protocols in challenging

optical environments.

http://www.transnav.eu

the International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 19

Number 2

June 2025

DOI: 10.12716/1001.19.02.17

478

on the inspection of a vessel’s submerged hull and the

development of a related database. A comprehensive

review of existing literature indicates that future port

infrastructures are expected to integrate advanced

automation technologies, electrification, and intelligent

energy management systems [7].

The deployment of Remotely Operated Vehicles

(ROVs) for the technical inspection of underwater port

infrastructure represents a promising and increasingly

viable alternative to traditional diver-based methods.

This approach offers the potential to substantially

reduce operational costs, inspection timeframes, and

risks to human safety. As diagnostic platforms, ROVs

can be outfitted with a range of sensor systems and

imaging technologies, facilitating the collection of

high-resolution visual data and comprehensive

technical information.

The objective of the present study is to assess the

feasibility of utilizing ROVs equipped with modulated

lighting systems for the visualization and diagnostics

of the technical condition of submerged components of

port wharfs. The goal is to determine whether this

approach can serve as a cost-effective yet equally

reliable alternative to conventional underwater

inspection techniques.

The successful planning and execution of

underwater engineering, inspection, and maintenance

activities depend critically on the optical characteristics

of the aquatic environment. In particular, two key

parameters—remote sensing reflectance (Rrs) and

turbidity—are instrumental in characterizing water

clarity, light propagation conditions, and the overall

suitability of a given water body for vision-based

operations. These parameters are especially pertinent

in port environments, where the presence of

suspended particulate matter (SPM), dissolved organic

and inorganic substances, and hydrocarbon-based

pollutants such as oil significantly affects underwater

visibility and image quality [9].

Among these factors, high turbidity constitutes the

most prominent challenge to effective underwater

imaging. Turbidity results primarily from elevated

concentrations of suspended particulate matter, which

scatter incident light and thereby impair visibility and

reduce the quality of acquired imagery. This scattering

is quantitatively described by the Volume Scattering

Function (VSF) (Formula 1), which represents the

angular distribution of scattered light intensity dI per

unit incident irradiance E in a given volume dV.

( )

VSF m sr

E dV





−−







(1)

This formulation accurately characterizes the

phenomenon of single scattering. In marine

environments, the Volume Scattering Function (VSF) is

known to exhibit significant asymmetry. Specifically,

forward scattering may exceed backscattering by up to

six orders of magnitude [15], a disparity that

substantially degrades the clarity of images acquired in

turbid waters. This pronounced anisotropy in

scattering behavior has been confirmed by the

pioneering measurements conducted by Petzold [14],

as well as by subsequent studies carried out in the

Baltic Sea [3, 4].

The integral of the VSF over the full solid angle

domain yields the scattering coefficient b(λ), which

quantifies the total amount of light scattered per unit

length and is expressed in units of m

-1

( ) ( )

2,b VSF sin d m



     

−



=





(2)

In axially symmetric systems, where particle

orientation is random, the scattering phase function—

defined as the Volume Scattering Function (VSF)

normalized by the scattering coefficient b(λ)—

describes the angular probability distribution of

scattered light. In highly turbid waters, where the

mean free path of photons is reduced to the order of

centimeters, multiple scattering effects become

predominant. Under such conditions, the single-

scattering approximation is no longer valid, resulting

in glare and diminished image contrast, particularly

when conventional, non-collimated lighting is

employed.

Turbidity represents a direct measure of water

clarity, quantifying the cloudiness or haziness caused

by the presence of numerous light-scattering and light-

absorbing particles. It is typically expressed in

Nephelometric Turbidity Units (NTU) and shows a

strong positive correlation with the scattering

coefficient b(λ), while correlating inversely with

underwater visibility [5]. Beyond its influence on light

propagation, turbidity imposes a fundamental limit on

the effective imaging range of optical sensors.

Turbidity levels are typically elevated in coastal and

port environments due to anthropogenic activities

such as dredging, vessel movement, and industrial

discharges.

In oligotrophic oceanic waters, turbidity is minimal,

allowing light penetration depths to exceed 30 meters.

Conversely, in turbid port environments, visibility

may be restricted to less than 1 meter due to high

concentrations of suspended particulate matter (SPM).

The particle size distribution (PSD), refractive index,

and angular scattering properties of the water are

strongly influenced by the composition and origin of

the particulates, all of which modulate light–matter

interactions within the medium.

SPM in port waters is predominantly generated by

sediment resuspension processes and typically consists

of silts, clays, and other mineral particles. These

particulates are highly efficient at scattering light,

particularly in the forward direction. In port areas,

sediment resuspension is frequently induced by vessel

propeller wash, cargo handling operations, and

construction-related activities. The backscattering

coefficient bb(λ) increases proportionally with mineral

content, contributing to elevated remote sensing

reflectance (Rrs) values, especially within the green and

red spectral bands.

Absorption is another critical factor affecting

underwater visibility. In clear, oligotrophic waters,

blue light exhibits the greatest penetration depth,

owing to minimal absorption by both water molecules

and colored dissolved organic matter (CDOM). In

contrast, turbid coastal waters—characterized by

elevated concentrations of CDOM and SPM—exhibit

absorption minima in the mid-visible range (500–600

479

nm), resulting in the characteristic green coloration of

such waters [6; 8; 17].

Remote Sensing Reflectance (Rrs) (Formula 3)

describes the spectral ratio of water-leaving radiance

(Lw) to downwelling irradiance (Ed).

( )



(3)

Remote sensing reflectance (Rrs) is significantly

influenced by both the absorption and backscattering

coefficients, which are themselves governed by the

concentration, size distribution, and composition of

suspended particulate matter. In underwater

environments where visual tasks such as crack

detection, biofouling assessment, and infrastructure

mapping are of critical importance, Rrs functions as a

real-time, non-invasive indicator of water clarity. As

such, it can inform both lighting modulation strategies

and image post-processing techniques employed by

Remotely Operated Vehicles (ROVs). Rrs typically

exhibits peak values at wavelengths where absorption

is minimal and backscattering is relatively high. This

spectral behavior has direct implications for the design

of ROV lighting systems: rather than relying on

broadband white light, the use of spectrally optimized

light sources—particularly within the green region

(500–600 nm)—can enhance image contrast and depth

perception under turbid conditions.

In addition to particulate-induced turbidity, the

presence of oil and hydrocarbon derivatives introduces

further complexity to underwater optical assessments.

This is particularly relevant in port and harbor

environments, especially those in proximity to

petrochemical terminals, where surface oil films,

dispersed oil droplets, and emulsified hydrocarbons

are frequently observed. These substances induce

nonlinear optical effects due to their complex refractive

indices and strong wavelength-dependent absorption

and reflection properties [9; 11].

Oil contamination in water typically leads to

increased absorption at shorter wavelengths (blue–

violet) and generates specular reflections at air–oil–

water interfaces. Furthermore, emulsified oil droplets

can modify the scattering phase function by enhancing

mid-angle and backscattering components. This optical

alteration results in elevated Rrs values in the red and

near-infrared spectral regions, while simultaneously

diminishing spectral contrast. The resulting flattening

of the reflectance spectrum reduces the effectiveness of

color-based segmentation and object detection

algorithms commonly used in underwater computer

vision systems.

Consequently, ROVs operating in port

environments must be capable of adapting to

dynamically variable optical conditions, which may

arise from fluctuating concentrations of suspended

particles, oil contamination, and inputs from urban

runoff.

2 METHODOLOGY

The ROV used in this experiment is CHASING M2

(figure 1), which is a compact-sized, professional

underwater ROV designed for professional users and

industrial applications. Its compact aluminium alloy

body allows for single-person operation and quick

deployment. The M2 features omni-movement in all

directions, and thanks to its depth-lock function, it can

hover precisely in any position. This enables

underwater photography, observation, and

manipulation at any angle. The ROV is compatible

with various attachments such as a Robot Claw, GoPro

camera, external LED lights, and a laser scaler, with a

maximum mounting payload of 1.5 kg. The default

battery provides up to 4 hours of continuous operation.

CHASING M2 is capable of recording underwater

temperature and depth, providing professional

underwater data support. Its maximum working depth

is 100 meters, and the operating temperature ranges

from -10 ℃ to 45 ℃. The device achieves a maximum

speed of 1.5 m/s (3 knots). Additionally, the M2 is

equipped with two LED lights with a total brightness

of 4000 lumens (three gears), which compensate for the

darkness of deep water.

They improve underwater distance and visibility

clarity, significantly restoring natural underwater

colors. It is also equipped with a 4K video resolution

camera and the ability to observe during investigation

by connecting a phone to the remote controller [2]. A

GoPro camera was used, which was mounted on the

vehicle.

Figure 1. Chasing M2 - ROV with GoPro camera used in

experiment

In the conducted experiment, white, red, and green

light were used to investigate how the colour of light

affects the image observed beneath the water's surface

(figure 2,3,4).

Figure 2. White light

480

Figure 3. Red light

Figure 4. Green light

3 RESULTS

The technical condition survey of the port's quay wall

was conducted using an ROV. The inspection work

commenced with scanning the vertical face of the quay

wall along the entire length of the examined section

(figure 5,6).

Figure 5,6. Visibility of quay wall with daylight

The ROV, equipped with a high-resolution video

camera and LED lighting, moved along the structure,

recording real-time imagery. The first series of

observations was conducted using standard white

light (figure 7,8), which served as a baseline for

assessing overall visibility and the characteristics of the

underwater image. Subsequently, red (figure 9,10) and

green (figure 11,12)light were used.

Figure 7,8. Visibility near the quay wall with white light

Figure 9,10. Visibility near the quay wall with red light

Figure 11,12. Visibility near the quay wall with green light

Under daylight conditions, visibility was low due to

ambient light scattering and absorption (Figures 5, 6).

Switching to ROV-mounted illumination improved

image acquisition. For white light (Figures 7 and 8) it

provided balanced spectral coverage, but succumbed

from glare and low contrast in turbid water due to

multi-wavelength backscatter.

In case of red light (Figures 9 and 10) scattering was

minimized but was rapidly absorbed, limiting

visibility to a short range. Figures 11 and 12 show that

for green light optimal visibility has been delivered.

The wavelength (~550 nm) aligns with the absorption

minimum in CDOM-rich waters, improving depth

penetration and structural detail clarity.

4 DISCUSSION

Quantitative image analysis revealed that green light

increased the signal-to-noise ratio (SNR) of structural

features by approximately 25% compared to white

light and 40% compared to red light. Metrics such as

edge contrast (Ce), grayscale variance (σ2), and

histogram entropy (H) were computed to evaluate

image quality under each spectral condition (Table 1).

Edge detection was first performed using a Sobel filter

[1], and then average contrast was computed across a

representative sample of edge segments. Moreover

grayscale variance was calculated for each image after

converting to grayscale. Histogram entropy H was

calculated using the Shannon entropy formula [16]. In

table 1 summarizes the results of image quality

assessment for three spectral lighting conditions

(white, red, and green) used during ROV inspections.

Table 1. The quantitative evaluation of image quality under

different lighting conditions

light

Ce (mean)

white

0.42

1056

6.72

red

0.31

678

5.88

green

0.54

1420

7.12

where edge contrast was calculated using the

difference in pixel intensity across detected edges:

max min

−

(4)

where:

481

− Imax - maximum pixel intensity on one side of an

edge.

− Imin - minimum pixel intensity on the other side.

Whereas grayscale variance was computed using:

( )



=−



(5)

where:

− Ii - intensity of the i-th pixel,

−

- intensity of the image,

− N - number of pixels.

The histogram entropy H was calculated using

formula:

H p log p

=−



(6)

where:

− pi - probability of the i-th grayscale level.

− L - total number of intensity levels.

Figures from 5 to 12 show that it is the green light

that consistently produced images with higher contrast

and lower noise levels. Green lighting outperformed

red and white in all three metrics, indicating better

contrast, structural detail, and texture preservation.

Red lighting, while reducing glare, low in entropy and

variance due to loss of detail. White lighting offered

balanced performance but introduced substantial

backscatter, diminishing edge clarity. Red lighting,

while effective at minimizing glare, was found

insufficient for identifying fine surface features beyond

1 meter of standoff. This is attributed to high

absorption of red wavelengths in water. Conversely,

white light produced adequate results at close range

but was impaired by multi-directional scattering

artifacts.

The use of modulated lighting markedly improved

the ability to discern surface textures, cracks, and

biofouling on the quay wall. These study's findings

align with established principles of underwater optics,

where water's absorption and scattering characteristics

vary across different wavelengths. Green light's

optimal balance between absorption and scattering

makes it particularly effective for underwater imaging

in turbid conditions. The rapid absorption of red light

and the high scattering associated with white light

limit their utility in such environments.

The integration of modulated lighting systems,

particularly those utilizing green LEDs, can

significantly enhance the capabilities of ROVs in

conducting underwater inspections. By improving

visibility and image clarity, these systems enable more

accurate assessments of structural conditions,

facilitating timely maintenance and repair

interventions. The findings underscore the importance

of tailoring ROV lighting configurations to the specific

optical properties of the inspection environment (local

approach).

Advancements in adaptive lighting technologies

and real-time image processing algorithms hold

promise for further enhancing ROV-based inspections.

The development of systems capable of dynamically

adjusting lighting parameters in response to

environmental conditions could optimize image

quality across a range of turbidity levels. Additionally,

integrating machine learning techniques for automated

defect detection could streamline the inspection

process and reduce reliance on manual analysis.

5 CONCLUSIONS

The results presented in this work may be used in the

future to provide a quantitative foundation for

selecting the optimal lighting spectrum for underwater

ROV inspections. It has been shown that modulated

lighting, particularly green light at approximately

550 nm, significantly improves image clarity and

defect detection during ROV-based underwater

inspections in turbid water conditions. Moreover red

light's rapid absorption and white light's high

scattering limit their effectiveness in enhancing

visibility in such environments. Therefore, tailoring

ROV lighting configurations to the specific optical

properties of the inspection environment is crucial for

optimizing image quality and inspection accuracy.

Future research should focus on the development of

dynamic lighting control systems and the

incorporation of automated image analysis tools to

streamline the inspection process and improve the

reliability of defect detection. The integration of

advanced lighting systems and adaptive technologies

can further enhance the capabilities of ROVs, enabling

more efficient and effective underwater inspections.

Moreover, without real-time knowledge of Rrs and

turbidity, visual navigation and inspection become

compromised, leading to false negatives in defect

detection or navigation errors. Therefore, deploying in

situ sensors capable of measuring Rrs and turbidity

profiles at multiple depths can inform lighting

configuration (e.g., selecting appropriate LED spectra)

and camera exposure settings.

REFERENCES

[1] Jason D. Bakos, Chapter 4 - Memory optimization and

video processing, Editor(s): Jason D. Bakos, Embedded

Systems, Morgan Kaufmann, 2016, Pages 147-185, ISBN

9780128003428, doi: 10.1016/B978-0-12-800342-8.00004-3

[2] https://www.chasing.com/en/chasing-m2.html

[3] Freda W. Comparison of the spectral-angular properties

of light scattered in the Baltic Sea and oil emulsions. J

Europ Opt Soc Rap Public. 2014;9:14017. doi:

10.2971/jeos.2014.14017

[4] Freda W, Piskozub J. Revisiting the role of oceanic phase

function in remote sensing reflectance. Oceanologia.

2012;54:29–38. doi: 10.5697/oc.54-1.029

[5] Haule K, Kubacka M, Toczek H, Pranszke B, Freda W.

Correlation between Turbidity and Inherent Optical

Properties as an Initial Recognition for Backscattering

Coefficient Estimation. Water (Switzerland).

2024;16(4):594.

[6] Kowalczuk P, Olszewski J, Darecki M, Kaczmarek S.

Empirical relationships between coloured dissolved

organic matter (CDOM) absorption and apparent optical

properties in Baltic Sea waters. Int J Remote Sens.

2005;26(2):345–370. doi: 10.1080/01431160410001720270

[7] Kosiek J, Kaizer A, Salomon A, Sacharko A. Analysis of

Modern Port Technologies Based on Literature Review.

TransNav, the International Journal on Marine

Navigation and Safety of Sea Transportation.

2021;15(3):667–674. doi: 10.12716/1001.15.03.22

482

[8] Lednicka B, Kubacka M, Freda W, Ficek D, Sokólski M.

Multi-Parameter Algorithms of Remote Sensing

Reflectance, Absorption and Backscattering for Coastal

Waters of the Southern Baltic Sea Applied to Pomeranian

Lakes. Water (Switzerland). 2023;15(15):2843.

[9] Lednicka B, Otremba Z, Piskozub J. Light Penetrating the

Seawater Column as the Indicator of Oil Suspension

Monte Carlo Modelling for the Case of the Southern Baltic

Sea. Sensors. 2023;23(3):1175.

[10] Lednicka B, Otremba Z, Piskozub J. Modelling the

upwelling radiance detected in a seawater column for oil-

in-water emulsion tracking. Scientific Reports.

2023;13(1):23098.

[11] Lednicka B, Otremba Z, Piskozub J. Vector irradiance

modelling in a seawater column for assessing the

detection capabilities of an oil-in-water emulsion. Optics

Express. 2024;32(17):29424–29435.

[12] Lousada SA, Camacho RF, Palacios JS. Underwater

Technical Inspections Using ROV Applied to Maritime

and Coastal Engineering: The Study Case of Canary

Islands. 2021 Jan 7. In: IntechOpen [Internet]. Available

from: https://www.intechopen.com/chapters/74726

[13] NOAA Ocean Explorer. [Internet]. Available from:

https://oceanexplorer.noaa.gov

[14] Petzold TJ. Volume scattering functions for selected

ocean waters. Scripps Institution of Oceanography

Report. 1972; SIO 72-78.

[15] Piskozub J, McKee D. Effective scattering phase

functions for the multiple scattering regime. Optics

Express. 2011;19(5):4786–4794. doi: 10.1364/OE.19.004786

[16] Tingting Tao, Cheng Ji, Chengyu Han, Jingde Wang, Wei

Sun, Study on the noise contents of different

measurements in industrial process and their impact on

process monitoring, Computer Aided Chemical

Engineering. 2022; 51:1057-1062. doi: 10.1016/B978-0-323-

95879-0.50177-6

[17] Woźniak SB, Meler J, Stoń-Egiert J. Inherent optical

properties of suspended particulate matter in the

southern Baltic Sea in relation to the concentration,

composition and characteristics of the particle size

distribution; new forms of multicomponent

parameterizations of optical properties. Journal of Marine

Systems. 2022;229:103720.

[18] Venkatesh V, Kodoth K, Jacob AA, Upadhyay V,

Jhunjhunwala T, Rajagopal P, Ali MN, Balasubramaniam

K. Non-Destructive Testing of Quay Walls Using

Submersible Remotely Operated Vehicles (ROV) In

Waterways Around the North Sea Coast. OCEANS 2022

- Chennai, 2022:1–6, doi:

10.1109/OCEANSChennai45887.2022.9775419

[19] Žaglinskis J, Noreikaitė E. The Application of an

Underwater Robot for the Creation of a Database for the

Technical Inspection of the Underwater Components of

Water Transport and Infrastructure. TMT. 2024; 3:182-96.

doi: 10.56131/tmt.2024.3.1.259