Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)

IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

Non-Destructive Approach for Detection of Food

Freshness using IoT
Bazaru Venkata Sai Nithin Banothu Sneha
Electronics and Communication Kandhada Sai Nithin Electronics and Communication
Electronics and Communication Gokaraju Rangaraju Institute of
Gokaraju Rangaraju Institute of
Engineering and Technology
Engineering and Technology Gokaraju Rangaraju Institute of
Hyderabad, India
Hyderabad, India Engineering and Technology
[email protected]
[email protected] Hyderabad, India
[email protected]

Vadladi Vijaya Kumar

2023 International Conference on Inventive Computation Technologies (ICICT) | 979-8-3503-9849-6/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICICT57646.2023.10134232

Electronics and Communication

Gokaraju Rangaraju Institute of
Engineering and Technology
Hyderabad, India
[email protected]

Abstract- Food and nutrition is essential for all li ving balanced diet with a higher intake of fiber, vitamins, and
beings to survi ve and perform vi tal acti vi ties like minerals. However, these sources of energy also attract
growth, development, and renewal of damaged cells. It is bacteria and environmental conditions such as humidity and
important to consi der various factors that can affect the temperature pro mote their growth, causing food spoilage
safety and quality of perishable foods, such as light, O2 and potential harm to human health. Food-borne illnesses
levels, temperature, humi dity, spoilage bacteria, and result in hospitalizat ion and death, affecting a large number
other environmental condi tions. When stored at room of individuals each year. Therefore, it is crucial to detect
temperature, food may undergo chemical reactions that food spoilage from the production to consumption stages.
affect its taste, texture, and aroma. Consuming s poiled
food can result i n foodborne illnesses, which can be Evaluating the freshness of food through manual methods,
dangerous. This research study presents an efficient such as smelling, tasting, and observing its color and
solution by utilizing IoT technol ogy to manage food texture, is time-consuming and susceptible to errors due to
storage. The proposed system is designed to address the human and environ mental factors. Therefore, it’s important
challenges associated wi th monitoring the quality of to progress in developing a system that can automat ically
stored food. The proposed solution combi nes vari ous IoT detect spoilage of food without human intervention. Digital
devices, cloud services, and web applicati ons. Computer image processing and machine learning techniques have
vision technol ogy offers a non-intrusive, automated, and recently been proposed as a means of detecting food
cost-effecti ve approach to eval uate food quality. This spoilage, offering significant potential for the food industry.
innovati ve soluti on can accurately assess the quality of Adopting a detection technology that is non-destructive and
food products without damaging them, allowing for a automatic can enhance food quality, increase production
more efficient eval uation process. The experimental efficiency, and reduce labor intensity, which is essential in
results demonstrate that the proposed system is effecti ve addressing growing concerns about food safety and quality.
in detecting food quality and promptly alerting users
about any degradation in food quality.

Keywords- Food Quality, Internet of Things, sensors, ripeness,

OpenCV

I. INTRODUCTION
In humans, food sources such as meat, meat products, fruits,
and vegetables provide the necessary energy. The
consumption of fresh fruits and vegetables has become Fig. 1 Fruit And Vegetable Spoilage Is A Hidden Contributor T o
increasingly popular worldwide as people aim for a mo re Underdevelopment

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1396

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)
IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

For the objective of maintaining the quality of food and these gases in a particular food item. The objective of the
preventing spoilage, it is crucial to mon itor various system is to detect spoiled food items and activate a buzzer
parameters such as pH, humidity, and o xygen levels. via NodeMCU, while also transmitting the collected data to
Changes in these parameters can signal the potential ThingSpeak for analysis in the cloud. To determine whether
spoilage of food products. Biosensors can also be used to a food item is spoiled or not, the authors utilized logistic
detect bacterial contamination in food samples. Our regression and trained the system using relevant data.
proposed system includes a monitoring device equipped
with gas sensors, pH sensors, and moisture sensors that
provide real-time data. The data collected is then transmitted
to a laptop and a Raspberry Pi mini-co mputer with a
mach ine-learning model that can analy ze the data and
determine the current state of the food. Based on this
analysis, the system can determine whether the food is safe
to consume or needs to be discarded. This approach is a
promising way to prevent food spoilage and guarantee that
food products are safe and of high quality.

The paper is structured in the following manner:

• A summary of previous research conducted in the area is
Fig. 2 Harvested Fruits are alive
provided in section II.
• Section III exp lains the operation of our suggested storage The e-fresh device [2], ut ilizes biosensors capable of
model. measuring pH, mo isture, and ethanol levels in a given food
• The experiment procedure is detailed in Section IV. item. To analyze a food item, the user selects it via an
• The results obtained from the experiment are presented in Android mobile applicat ion, and this command is sent to the
Section V. Arduino Uno through a Bluetooth module. The Arduino
• Project's applications and implementations are discussed in Uno then gathers readings from the sensors and makes a
Section VI, and Section VII addresses future prospects. decision based on a predefined algorithm. Finally, the
• Sect ion VIII presents the conclusions drawn fro m the results are displayed on an LCD screen.
project.
In the study conducted by B. Sarmah et al. [4], Ultrav iolet
(UV) detectors were emp loyed to observe food quantity,
II. LITERATURE S URVEY while an MQ4 gas sensor was utilized to catch the presence
Detecting food spoilage through manual inspection using of CH4 gas in the air as an indicat ion of food freshness . This
methods such as observing color changes, tasting, and approach allowed for immediate alerts to be sent to users
smelling is very t iring, less effective, and expensive due to when food degradation begins, enabling pro mpt detection of
potential human slip-ups and environmental factors that can food quality.
lead to inaccurate results. To address this issue, various
technologies have been developed for monitoring food In a study by Keshavamurthy et al. [5], a system for
quality and detecting food spoilage. identifying and assessing the quality of fruits was proposed
using Python’s OpenCV library. To accurately assess the
The emergence of IoT technology has enabled the quality of fruits and determine their type, they emp loyed a
connection of ord inary objects to the Internet through the Convolutional Neural Network (CNN) that utilized
use of sensors, actuators, and other sensing technologies, numerical data. Their device consisted of a microcontroller
facilitating intelligent sensing, monitoring, detection, and with a sensor system and a conveyor belt design for fru it
other functionalities. In the development o f IoT solutions, separation. Environ mental conditions in the storage unit
intelligent sensors, and wireless communication are vital were mon itored using sensors to maintain fruit quality over
components, making sensors an essential part of the process. time, with the data transmitted to the cloud through Internet
As a result, IoT systems can be employed to develop of Things (IoT) technology.
visibility and traceability systems for the management of
perishable food. In their study, Shubham Pradhan et al. [6] emphasize on the
vital ro le that the agro-food sector plays in our life, as
In their work, Hebber et.al [1] proposed a system that people increasingly consume packaged foods fro m around
emp loys NH3 and O2 gas sensors to measure the levels of the world. To address issues in the fruit quality checking

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1397

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)
IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

process, the authors propose a simp le, portable solution built measure the pH of dairy p roducts, while gas sensors detect
on the Internet of Things (IoT) technology and a set of the amount of natural gas emitted by a particular food item.
apples. A microcontroller that’s based on Arduino is utilized This method of evaluating food quality is useful in
for analog operation, and the concept is validated through a situations where food is tightly packed in a container or
pre-calibrated indexing table. The study specifically when food degradation occurs due to a lack of fresh air
examines how the quality-checking process can be (o xygen) or due to the growth and activities of
conducted while monitoring the ripening of fruits. microorganis ms, particu larly bacteria , and yeasts. The real-
time data fro m these sensors is transmitted to a mini-
computer, such as a Raspberry Pi, for analysis and
III. PROPOS ED METHODOLOGY
processing. This subsystem can be used to mon itor fruits,
vegetables, meat, grains, and dairy products such as milk
Our project objective is to design a device that utilizes and curd.
sensors to detect the freshness and spoilage of food with
high accuracy. To increase detection accuracy and reduce pH, a measure of acid ity or alkalin ity, is determined by the
false negatives, we will use a combination of both sensors potential of hydrogen in a solution and is usually measured
and openCV. Co mbin ing IoT and machine learning will on a scale ranging fro m 0 to 14. In the case of dairy
enhance the device's interoperability and potential products such as milk and yogurt, it is crucial to maintain a
applications. The goal is to increase the shelf life o f foo d standard pH since even minor variations can have a
products by reducing or managing senescence, which is the
natural deteriorat ion of the product due to various factors
such as aging, ripening, and exposure to environ mental
stressors.

The presented system incorporates two sub-systems: a

computer vision system and a food storage unit monitoring
system. Various sensors are utilized to measure different
food parameters, including gas level, pH, mo isture, and
others. Raspberry pi mounted with the pi cam is used for
getting a continuous feed fro m the storage unit. The
workflow is discussed further in this section.

Fig. 3 Experimental Setup

Fig. 4 Architecture of our project

A. Food Storage Unit Subsystem

significant impact due to the inverse exponential nature of
the measurement. For instance, milk has a standard pH of
This subsystem is equipped with a variety of sensors, approximately 6.5-6.7. Any changes in pH could alter the
including pH sensors, gas sensors, humid ity/mo isture palate, taste, and storage life of dairy products.
sensors, and more. These sensors monitor real-time Consequently, pH sensors are the best choice for mon itoring
parameters such as temperature, moisture, and natural gas pH in dairy products.
emissions from food items. For examp le, p H sensors

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1398

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)
IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

Moisture is the presence or amount of water or other liquid IV. RES ULTS
substance within a material, either as a component of its
This section will present the results and findings obtained
composition or as a coating on its surface. In the case of fro m the e xperiments conducted. 75 tests were perfo rmed
Meat and poultry, they are comprised of natural water, using various types of apples, bananas, and oranges. The
muscle tissue, bone, and fat. The muscle tissue accounts for confusion matrix is shown in Tab le 2 belo w. An accuracy of
around 75% of the meat, and the water content varies 96% is calculated from the matrix.
depending on the type of meat. Appropriate water content is
T ABLE 1 SHOWING DET ECT ION BY PROPOSED SYST EM
crucial to maintain meat freshness, as excessively wet or dry
meat is unsuitable for consumption. Excessive water content Type of Fresh/Rotten Detected by
can modify the pH level and chemical co mposition of meat, S. No Fruit System
ultimately affecting its freshness. Therefore, a mo isture
sensor is the most effective way to measure the moisture 1 Apple Fresh Fresh
content of chicken and beef.
2 Banana Rotten Rotten
Ethanol is a natural substance, common ly referred to as
3 Tomato Fresh Fresh
alcohol, that has several uses. It can be utilized as a method
to artificially locate ripe fruit, and lower amounts of ethanol 4 Orange Fresh Fresh
found in fruits may serve as a feeding catalyst. During fru it
5 Apple Rotten Rotten
maturation, co lor, taste, sugar, and ethanol content are
modified. As bananas ripen, they emit various gases,
including ethylene and small quantities of ethanol. As time T ABLE 2 CONFUSION MAT RIX
passes, ethanol production increases. To monitor the
production, presence, and concentration of ethanol in fruits,
gas sensors are employed. Actual

B. Computer Vision Subsystem Fresh Rotten

Food processing industries can utilize co mputer vision Predicted
technology as a non-destructive and practical solution for Fresh 35 1
evaluating the quality of their food products. An array of
images of oranges, apples, and bananas with differing
shapes and sizes is captured by employing a camera to Rotten 2 37
construct the dataset. This subsystem consists of a mini-
computer called raspberry pi wh ich is interfaced with a
camera. Raspberry Pi remains a top choice in the industry as
an affordable, versatile single -board computer with GPIO
pins for physical computing and IoT exploration, running
Linu x OS. Raspberry pi analyses the data it gets from
OpenCV analysis and the data from the sensors to make a
decision on whether the placed item is fresh or decaying.

Fig.5 Example images used in the project that consists of different stages of Fig. 6 pH sensor placed in milk
fresh and rotten apple, banana, and orange

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1399

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)
IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

The sensors can detect spoilt food items, and management

can be quickly alerted. Moreover, p rioritizing food items
with shorter storage lives to sell first can assist in stock
management, leading to more efficient inventory practices
and less waste. By implement ing the proposed system, staff
members can promptly remove spoiled food items fro m
inventory and replace them with fresh ones, leading to
enhanced productivity and efficiency.

• Household: A possible rephrased version of the original

sentence could be: In households, a set of sensors can be
placed inside refrigerators to identify any spoiled food
items. Once a spoiled food item is detected, the system
triggers a buzzer sound and sends a notification to the
owner, instructing them to discard the affected food item.

Fig. 7 T he graph plotting of the pH values of fresh milk over time showing
a gradual decline in pH as the milk becomes increasingly sour VII. CONCLUS ION
This paper suggests a non-destructive approach for
As milk spoils, it undergoes a chemical change that results in increased
acidity and a decrease in its pH value. The bacterium Lactobacillus found identifying bru ises in food products by utilizing co mputer
in milk breaks down lactose sugar into lactic acid, causing the pH level to vision, thereby ensuring their quality and safety.
decline. The rate of pH decrease would accelerate as the milk spoils further, Additionally, it proposes an IoT-based system that monitors
eventually falling below 4.6 - the pH level at which milk is no longer safe
and maintains food storage rooms to preserve the quality of
for human consumption.
food items for a more extended period. The comb ination of
sensors, IoT, and mach ine learning can revolutionize the
V. FUTURE WORK food industry, reduce manual inspection tasks, and
At present, we are testing different dairy and fruit products minimizes human errors. This will create an environment
to expand the capabilities of our device by incorporating where food manufacturers are encouraged to s ell healthier
new sensors and optimizing the current ones. Additionally, food products and increase awareness of healthy food
we plan to integrate an embedded system into the device, options among consumers. As a result, the food processing
allo wing for data transmission to a central server that can be industry, which manufactures various food products will
used for research purposes, as well as providing users with
have to prioritize food quality detection and management.
access to their data history. In a bid to enhance the accuracy
of results, we intend to integrate an Artificial Neural
Network(ANN) into the central server to analy ze the data
VIII. REFERENCES
and convey/communicate the findings to end-user
contraptions for practical application. [1] N. Hebbar, "Freshness of Food Detection using IoT and Machine
Learning," 2020 International Conference on Emerging T rends in
Information Technology and Engineering (ic-ETITE), 2020, pp. 1-3, doi:
10.1109/ic-ETITE47903.2020.80.
VI. APPLICATIONS
[2] Naveed Shahzad, Usman Khalid, Atif Iqbal, Meezan-UrRahman,
• Industry: The implementation of this model in food “eFresh-A device to detect food freshness” International Journal of Soft
Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-8 Issue-3,
processing industries can lead to more efficient mon itoring September 2018.
of different categories of food items and reduce the
[3] E. Henrichs and C. Krupitzer, "Towards Adaptive, Real-T ime
consumption of t ime and manpower. As a large nu mber of Monitoring of Food Quality Using Smart Sensors," 2022 IEEE
food items are processed on a conveyor belt, s ensors placed International Conference on Autonomic Computing and Self-Organizing
across the belt can detect spoilage and provide management Systems Companion (ACSOS-C), CA, USA, 2022, pp. 70-71, doi:
10.1109/ACSOSC56246.2022.00034.
with data on when and where most occurrences of spoilt
food items occur. Th is data can be used to take necessary [4] B. Sarmah and G. Aruna, "Detect ion Of Food Quality and Quantity at
Cold Storage using IoT," 2020 International Conference on Wireless
action to prevent further spoilage, resulting in imp roved
Communications Signal Processing and Networking (WiSPNET), 2020, pp.
food quality and reduced waste. 200-203, doi: 10.1109/WiSPNET48689.2020.9198348.

• Retail Stores: The suggested model can be advantageous [5] Keshavamurthy, & Mariyam, Steffi & Meghamala, M & Meghashree,
M & Neha, (2019). Automatized Food Quality Detection and Processing
in supermarkets and retail stores by placing an array of
System Using Neural Networks. 1442-1446.
ammon ia and oxygen sensors inside shelves and containers. 10.1109/RTEICT46194.2019.9016919.

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1400

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Inventive Computation Technologies (ICICT 2023)
IEEE Xplore Part Number: CFP23F70-ART; ISBN: 979-8-3503-9849-6

[6] Partha Pratim Ray “IOT based fruit quality measurement

system”.Online International Conference on Green Engineering and
Technologies (IC-GET) 2016, IEEE. ISBN: 978-1-5090-4557-0, 19 Nov.
2016

[7] IoT-based milk monitoring system for detection of milk adulteration,

January 2018, International Journal of Pure and Applied Mathematics
118(9):21-32

[8] D. Karakaya, O. Ulucan, and M. T urkan, "A Comparative Analysis on

Fruit Freshness Classification," 2019 Innovations in Intelligent Systems
and Applications Conference (ASYU), 2019, pp. 1-4, doi:
10.1109/ASYU48272.2019.8946385.

[9] A. Prajwal, P. Vaishali, Z. Payal, and D. Sumit, "Food Quality

Detection and Monitoring System," 2020 IEEE International Students'
Conference on Electrical, Electronics and Computer Science (SCEECS),
Bhopal, India, 2020, pp. 1-4, doi: 10.1109/SCEECS48394.2020.175.

[10] M. V. C. Caya, F. R. G. Cruz, C. M. N. Fernando, R. M. M. Lafuente,

M. B. Malonzo and W. -Y. Chung, "Monitoring and Detection of Fruits and
Vegetables Spoilage in the Refrigerator using Electronic Nose Based on
Principal Component Analysis," 2019 IEEE 11th International Conference
on Humanoid, Nanotechnology, Information Technology, Communication
and Control, Environment, and Management ( HNICEM ), 2019, pp. 1-6,
doi: 10.1109/HNICEM48295.2019.9072715.

[11] N.Usharani, D.Suruthi, V.Sangeetha, L.Punitha, et al.,” Arudino-based

smart IOT food quality detection technology”, International Research
Journal of Engineering and T echnology, Volume- 07, March 2020.

[12] R. Raju, G. E. Bridges, and S. Bhadra, "Wireless Passive Sensors for

Food Quality Monitoring: Improving the Safety of Food Products," in IEEE
Antennas and Propagation Magazine, vol. 62, no. 5, pp. 76-89, Oct. 2020,
doi: 10.1109/MAP.2020.3003216.

[13] A. Mostaccio, G. M. Bianco, S. Amendola, G. Marrocco, and C.

Occhiuzzi, "RFID for Food Industry 4.0 - Current T rends and Monitoring
of Fruit Ripening," 2022 IEEE 12th International Conference on RFID
Technology and Applications (RFID-TA), Cagliari, Italy, 2022, pp. 109-
112, doi: 10.1109/RFID-TA54958.2022.9924104.

[14] J. M. Talavera, L. E. Tobón, J. A., Gómez, M. A., Culman, J. M.,

Aranda, D. T ., Parra, ... and L. E. Garreta, “Review of IoT applications in
agro-industrial and environmental fields.” Computers and Electronics in
Agriculture, vol. 142, pp. 283-297, 2017.
[15] F. Ying and L. Fengquan, "Application of Internet of Things to the
Monitoring System for Food Quality Safety," 2013 Fourth International
Conference on Digital Manufacturing & Automation, Shinan, China, 2013,
pp. 296-298, doi: 10.1109/ICDMA.2013.71.

979-8-3503-9849-6/23/$31.00 ©2023 IEEE 1401

Authorized licensed use limited to: The Technology Library. Downloaded on December 14,2024 at 12:45:05 UTC from IEEE Xplore. Restrictions apply.