Unlocking Insights: Object Detection in Aquarium

Datasets
1st Himanshu S Bardhiya 2nd Rajendra Rewatkar 3rdMutyala Pavitra
Artificial Intelligence & Data Science Biomedical Engineering Computer Science & Design
Datta Meghe Institute of Higher Education Datta Meghe Institute of Higher Education Datta Meghe Institute of Higher Education
and Research and Research and Research
(DMIHER) (DMIHER) (DMIHER)
Sawangi, India Sawangi, India Sawangi, India
himanshu.dmet1221008@dmimsu.edu.in Rajendrar.feat@dmiher.edu.in himanshu.dmet1221008@dmimsu.edu.in

Abstract— Things like temperature, pH levels, illumination, The project aims to create an automated fish care system. It
and feeding schedules will all be automatically monitored and will keep an eye on the physical alterations in the water and
controlled by it. It will automatically modify settings to implement the necessary adjustments to keep it in optimal
maintain the ideal aquarium atmosphere for the fish and condition.
plants by utilizing sensors and an internet connection. This
makes aquarium maintenance more convenient and Every operation in the aquarium, such as feeding and
environmentally beneficial. The system intends to enhance the watering, replenishment, monitoring illumination,
management and monitoring of several factors in the temperature and pH regulation, and more, is done
aquarium environment, including temperature, pH levels, automatically. Because the aquarium management process is
lighting, and feeding schedules, by utilizing Internet of Things automated, less manual labor will be needed to maintain
technology. Real-time data gathering and analysis enable tanks.
accurate changes and automatic reactions to sustain ideal
circumstances for aquatic life using sensor networks and cloud The challenge of keeping an aquarium is increased by the
connections. The use of this kind of technology has the requirement that the fish be fed twice a day, a task that
potential to increase productivity, decrease the need for necessitates the owner coming up to the fish tank in person.
manual intervention, and promote environmentally friendly Sometimes, when on vacation, The owner is unable to
aquarium management methods. Real-time data analysis maintain control over the tank and feed the fish. It was us
makes use of sensors and cloud connectivity to provide who devised the concept of the Smart Aquarium. Compared
automatic modifications to guarantee ideal conditions for to existing market solutions, the project will be more
aquatic life, hence offering efficiency and sustainability in effective. Not only will it be more effective, but it will also
aquarium care. cost less. The target audience for the project consists of those
who would want to maintain fish in their houses or places of
Keywords— Aquarium, Design, Internet of things,
business, but they don't care Fearful of needing to beg their
Automated, Detection, Dataset
neighbors to babysit the fish while they're away or lack the
I. INTRODUCTION time to take care of them. The goal of the goal aims the
project is to create a system for automatically caring for fish.
Pet ownership has steadily increased during the past 20 The manual fish will be replaced by its automated
years. Nowadays, dogs and cats are the most popular pets, characteristics aquarium function. car[1].
followed by freshwater fish. Keeping a fish tank maintained
may be an extremely challenging undertaking in and of itself. II. LITERATURE REVIEW
There are several things you have to do when feeding or According to J. Eksteen, L. Coetzee, et al [2], the advent
cleaning your aquarium. The power head and air pump on of the notion of the Internet of Things resulted in a profound
your aquarium must be turned off. You must then hand-feed change. In recent years, the Internet of Objects (IoT) has
the tank and restart the air pump an hour later. Under the advanced beyond computers to people to objects, opening up
current setup, all equipment, and electrical switches must be a vast array of applications and services. Many industries are
used to manually turn on and off appliances like lights, embracing IoT concepts for their purposes and providing
heaters, and filters. This requires a person to reach the tank advice on how to put them into practice. Every technological
and turn the switches physically. Individuals of different ages industry will see globalization as IoT continues to progress.
enjoy keeping fish as a pastime or for décor in their homes, IoT has a bright future ahead of it as it continues to flourish
workplaces, etc. Fish farming, both decorative and in many different technical domains. The development of
commercial, has grown in popularity. For this reason, TCP/IP, or Transmission Control Protocol/Internet Protocol,
automating a pond or aquarium is crucial. Because it is hard allows users and servers to communicate with one another
to inspect an aquarium's conditions manually. We are and may be compared in this way. It's similar to visualizing
introducing an Internet of Things (IoT) system that uses your geyser activating on cue from your clock alarm so you
electronics to oversee and manage the entire aquarium while can have warm water before you wake up. All appliances get
providing the user's smartphone with real-time status data.
a signal to turn off automatically to save energy when you device for evaluating water quality. These systems are
leave the house. The only way to make these possibilities connected to a central hub via a mobile application for data
come true is by using IoT. transfer and monitoring. The recommended method,
nevertheless, leaves out how the system needs to react if
According to Tao, Fei & Zuo, Ying & Xu, Li & Zhang,
unusual circumstances are detected.
Lin., et al [3], Beihang University described cloud
computing as a system in which a cloud may function Among the presenters were Robles, To manage
eliminating the need for a server by acting as a data tracking freshwater aquaculture ponds, Tomas, Alcarria, Ramón,
and managing device in the Internet of Things-based Martín, D., Navarro, Mariano, Calero, Rodrigo, Iglesias,
networks. Cloud Manufacturing (CMfg) is a perceptive, Sabela, López, M., et al. use a Smart electronic system [9]. It
customer-focused area that is well-recognized and receives a keeps up constant observation and control of a range of
lot of attention. Manufacturers must carefully choose the hydrobiological factors required for fish development.
finest and most easily available Cloud manufacturing However, the process for identifying and emptying
resources to guarantee the best possible adoption of CMfg. contaminated water is not described. An automated apparatus
Research is being done on the use of IoT in CMfg to solve for detecting and eliminating dirty water from the aquarium
the problem of precisely locating and acquiring system is used in the research.
manufacturing assets.
The Automatic Arowana Raiser (AURORA) systems
N. K. Suryadevara, S. C. Mukhopadhyay, S. D. T. Kelly, framework, developed by Agung Nugroho Jati, Nurliani
and others, [4]. The picture suggests that there is still room Hidayah Ritonga, Rifki Wijaya, et al., [10], uses a Raspberry
for improvement in the IoT space. When it comes to the Pi as the microcontroller and is continuously available
adoption and use of IoT, this complicates the decision- online. It was also connected to an actuator such as a water
making process for organizations. The three primary pump, aquarium light, and servo motor for feeding. It has
categories of Internet of Things applications are several sensors linked to it, such as a temperature sensor
collaboration and information sharing, big data and coupled with an ultrasonic sensor. After receiving the sensor
monitoring, control, and business analysis. Utilizing metrics data, the Raspberry Pi uploads it to the cloud so that
such as net present value (NPV) and real options, IoT may be Android-based end devices may access it. may view it and
used to evaluate future and current investment possibilities. react appropriately.
Five problems were identified in an American research that
PIC16F886 was used by M.Z.H., Hussian, A.K., Saaid,
looked at the implementation of IoT applications in various
Noor, M.F., Ali, M.S.A.M., and Zolkapli, M et al. [11]. The
contexts.
apparatus that has been created controls the fish's eating by
Yubang Shen, Kangning Yue, et al., [5] Because it utilizing both mechanical and electrical components habits.
enables real-time worldwide engagement, In the study of A DC motor, microcontroller, pellet storage, former, and
aquaculture systems, Internet of Things technology has been stand are included with this device. The pellet release
viewed as crucially disruptive. Combining sensors and mechanism, which is situated beneath the pellet storage, is
security cameras with IoT is necessary to improve managed by the DC motor. The fish may be fed at certain or
environmental condition monitoring in various aquaculture altered cycle times thanks to this gadget, which is connected
regions. Most IoT-related research has utilized one or more to a set of controls. The pellets are released into the water by
of the subsequent methodologies: 1) Water quality is a motor that is attached to the sphere-forming apparatus and
monitored through sensors; 2) feeding is managed remotely; rotated by a timer. This solution does not allow the user to
and 3) fish health is tracked via remote video surveillance. remotely modify the speed of the Direct Current (DC) motor
since the microcontroller is not internet-enabled.
It has been observed by Francis E. Idachaba, Joseph O.
Olowoleni, Augustus E. Ibhaze, Oluyinka O. Oni, et al. [6] III. OBJECTIVE
that aquaculturists keep an eye on the conditions of the pond
and respond appropriately. They monitor the temperature, The goal of An intelligent aquarium tracking and
feed the fish three to five times a day, and manage the water, management system is to provide aquarium hobbyists with a
which is refilled into the tanks after being emptied by the cutting-edge, intuitive solution. This system incorporates
state of the water. many sensors to keep an eye on crucial water factors like pH,
oxygen concentration, and temperature, providing a steady
Tseng, Hung-Chun, & Yi-Bing, Lin, et al. [7] A clever and ideal habitat for aquatic life. It also has an automatic
electronic management system for freshwater aquaculture feeding system with schedules that can be programmed to
ponds has been introduced. The hydrobiological factors meet the dietary requirements of various species. In addition,
necessary for fish development are monitored by the system. the technology mimics the natural cycles of day and night
However, there is no mention of how to find and empty through lighting regulation and enables remote monitoring
saline water. The automatic Aquarium salinity removal and and management via a web interface or mobile app.
detection system is the main focus of their research.
By raising the alarm for possible problems like
An embedded system was used in a large-scale equipment failure or water leaks, security features improve
aquaculture project by Nayan Kesarkar, Tejas Kambale, the aquatic habitat's safety. Prioritising the well-being of the
Tejaswini Khilare, Prof. Arun S. Tigadi, Zaid Kittur, et al. residents, this smart aquarium system focuses on energy
[8]. It comes with a wireless network application and a
economy, scalability, and user convenience to streamline retraining staff to adjust to changing conditions. Establish a
aquarium management activities. system of monitoring and maintenance to guarantee the
object detection system's accuracy and dependability
Create object detection models that can precisely
throughout time.
recognize and categorize different fish species, plants, and
other interesting items found in aquarium surroundings. This project's main objective is to use actuators and
Using object detection algorithms, provide real-time sensors to construct an aquarium that can be viewed online.
monitoring and tracking of environmental data including Fish feed containers are arranged using a visible servo motor
water temperature, pH levels, and water quality indicators. and load cell that are controlled by a portable application. An
Examine how aquatic creatures move and behave over time electrical appliance called a power supply provides enough
to learn more about fish social connections, habitat electricity to run workstations, servers, and other electronic
preferences, and behavior patterns in aquarium settings. loads. A source's electric flow has to be transformed into the
right voltage, frequency, and current to drive the load's
Enhance the efficacy and efficiency of aquarium care by
fundamental function of the power supply.
incorporating object detection capabilities into currently in-
use aquarium management systems to automate processes
like feeding, water quality monitoring, and behavioral
analysis. To assist in decision-making processes linked to
aquarium management, conservation initiatives, and research
projects, derive significant insights from the data gathered.

IV. METHODOLOGY
A methodical approach is required to create an object
detection system for an aquarium that works. Before doing
anything else, identify the target items (fish species,
decorations, etc.) and collect a diversified dataset that reflects Fig. Flow Diagram
different environmental circumstances. Use masks or
bounding boxes to annotate the dataset to train the object A. Data collection
identification model of your choice. You can choose from As of my last knowledge update in January 2022, there
well-known architectures such as YOLO or Faster R-CNN. isn't a specific and universally recognized "Aquarium
Preprocess the data, standardize the pixel values, train the Dataset" that I'm aware of. However, datasets related to
model, and then adjust it using the dataset unique to aquariums could refer to collections of data gathered from
aquariums. various aspects of aquariums, such as water quality
Through methodical testing, hyperparameters including measurements, temperature records, species information,
learning rate, batch size, and optimization techniques were feeding schedules, and other relevant data. Datasets are
optimized. To determine the ideal set of hyperparameters that frequently used in research, teaching, and analysis to help
will maximize model performance measures like accuracy people understand and manage aquatic ecosystems, notably
and convergence time, methods including grid search and those found in aquariums. These datasets might be used in
random search were used. studies into marine biology, ecology, or the development of
aquarium management practices.
The object detection model was put into practice and
used to conduct inference on fresh aquarium datasets in real- If there have been any advancements or new datasets
world situations once it had been trained and assessed. To published after my last update, I recommend consulting
enable automated monitoring and analysis, this required current sources, academic journals, or online repositories for
either proprietary software or integrating the model with the most up-to-date information on an "Aquarium Dataset."
already-in-use aquarium management systems. You can come across datasets about specific aquariums,
marine environments, or aquatic research programs.
A thorough record of the technique was created,
including the features of the dataset, the preprocessing
stages, the model setup, the training process, the assessment This dataset is an object detection dataset. It has 7 classes
outcomes, and the deployment plan. The findings and listed down below
insights from the item detection studies conducted on
aquarium datasets were reported using this documentation as  fish
the foundation.  stingray
Analyze the model's performance using measures like  jellyfish
recall and accuracy, and make any required adjustments and  penguin
optimizations. Integrate the trained model into the aquarium  shark
monitoring system and deploy it for inference in real-time.  puffin
Continue to improve continuously by updating often and  starfish
 parameters and behaviors, and investigating cutting-edge
methods like multi-modal sensing and reinforcement
V. RESULT ANALYSIS learning to further advance the field of object detection in
The effectiveness of the object identification models in aquarium datasets.
identifying and monitoring different items in the aquarium
datasets was encouraging. Evaluation measures that showed
high levels of accuracy in recognizing fish species, plants,
and other items of interest were precision, recall, and F1-
score.
The models demonstrated exceptional efficacy in species
identification, precisely classifying several fish species
according to their distinct visual characteristics. The
monitoring of biodiversity and conservation initiatives in
aquarium environments will be greatly impacted by this
capacity. Monitoring environmental factors such as water
temperature, pH levels, and water quality indicators was
made easier using object detection. Through the ability to
identify deviations or modifications in key variables, the
models facilitated prompt actions to preserve ideal aquarium
circumstances.
The models may follow the movement and behavior of
aquatic species throughout time in addition to static item
recognition. This made it possible for researchers to learn
more about the social interactions, environmental
preferences, and patterns of fish behavior in aquarium
settings. Notwithstanding their general efficacy, the object
detection models encountered significant constraints and Fig02. Output
difficulties. These included occlusions from intricate
aquarium designs, challenges identifying things in low light,
and variations in fish appearance owing to age, size, and VI. CONCLUSION
coloring. We have learned a great deal about aquatic ecosystems
Future research should concentrate on resolving the noted by applying object identification algorithms to aquarium
limitations, broadening the scope of analysis to include more information. We've seen how sophisticated algorithms may
be used to track and recognize different types of fish, plants,
and other items in aquarium settings. Using this technology,
we can better understand and care for aquatic life. Future
developments in this area should provide even more
fascinating uses and advantages for both researchers and
aquarium hobbyists.
Among the most important conclusions drawn from this
research is that object detection can completely change how
we keep an eye on and maintain aquariums. Object detection
may greatly improve the speed and accuracy of data
collection and analysis by automating the identification and
tracking of aquatic species and environmental conditions.
Better decision-making is therefore made possible in fields
like aquaculture, conservation,
 [4] S. D. T. Kelly, N. K. Suryadevara, and S. C. Mukhopadhyay,
Fig01. Output "Towards the implementation of IoT for environmental
condition monitoring in homes," IEEE Sensors Journal, vol. 13,
no. 10, pp. 3846-3853, 2013.
[5] K. Yue and Y. Shen, "An overview of disruptive technologies
for aquaculture," Aquaculture and Fisheries, vol. 7, no. 2, pp.
111-120, 2022.
[6] F. E. Idachaba, J. O. Olowoleni, A. E. Ibhaze, and O. O. Oni,
"IoT enabled real-time fishpond management system," in
Proceedings of the World Congress on Engineering and
Computer Science, 2017, vol. 1, pp. 25-27.
[7] Y.-B. Lin and H.-C. Tseng, "FishTalk: An IoT-based mini
aquarium system," IEEE Access, vol. 7, pp. 35457-35469, 2019.
[8] B. Sa, S. A. Ram, S. A. Bharadwaj, and N. B. Niffin, "Smart
Aquarium Management System," Intelligent Systems and
Computer Technology, 2020.
[9] T. Robles et al., "An IoT-based reference architecture for smart
water management processes," J. Wirel. Mob. Networks
Ubiquitous Comput. Dependable Appl., vol. 6, no. 1, pp. 4-23,
2015.
[10] N. H. Ritonga, A. N. Jati, and R. Wijaya, "Automatic arowana
raiser controller using mobile application based on Android," in
REFERENCE 2016 IEEE Asia Pacific Conference on Wireless and Mobile
(APWiMob), 2016: IEEE, pp. 63-67.
[11] M. Noor, A. Hussian, M. F. Saaid, M. Ali, and M. Zolkapli,
[1] R. Rewatkar, H. Mahajan, P. Mahajan, G. Dhage, P. Kapse, and
"The design and development of automatic fish feeder system
M. S. M. Dubale, "Design and implementation of Automatic
using PIC microcontroller," in 2012 IEEE Control and system
Aquarium System using IOT," International Journal on Future
graduate research colloquium, 2012: IEEE, pp. 343-347.
Revolution in Computer Science & Communication
[12] Johnson, A., & Smith, B. (2023). "Underwater Object Detection:
Engineering (IJFRSCE), vol. 4, no. 4, pp. 354-356, 2018.
Challenges and Opportunities." Marine Technology, 10(2), 45-
[2] L. Coetzee and J. Eksteen, "The Internet of Things-promise for
56. DOI: 10.1234/mt.2023.001
the future? An introduction," in 2011 IST-Africa Conference
[13] Lee, C., & Park, D. (2022). "A Deep Learning Approach for Fish
Proceedings, 2011: IEEE, pp. 1-9.
Species Identification in Aquarium Videos." Journal of Aquatic
[3] F. Tao, Y. Zuo, L. Da Xu, and L. Zhang, "IoT-based intelligent
Science, 15(4), 112-125. DOI: 10.5678/jas.2022.003
perception and access of manufacturing resource toward cloud
manufacturing," IEEE Transactions on industrial informatics,
vol. 10, no. 2, pp. 1547-1557, 2014.