use of AI in post harvest storage of fruit crops.pptx

Hemwati Nandan Bahuguna Garhwal University , Srinagar
Uttrakhand
School of Agriculture and Allied
Sciences
Department of Horticulture
SOA/HS/PG0: Self Study Courses (Project Preparation and
Presentation on Advances in Horticulture)
TOPIC: Use of AI in postharvest storage of fruit crops.
Submitted to: Dr. T.S. Bhist
Submitted by : Akshita Sharma
Roll No.: 22134338001
MSc. (Ag.) Horticulture 3rd sem.

Sr.No. Content
1. Introduction
2. Objective of postharvest management.
3. Challenges in postharvest Management.
4. Over view of AI and its application in horticulture.
5. Role of AI in postharvest management of fruits.
6. AI-based techniques for fruit quality assessment.
7. Computer vision system
8. Sensor technologies.
9. Machine learning algorithms.
10. AI-based predictive models for fruit storage and transportation
11. Case study-1
12. References

Fruits and vegetables are important supplement to the nutritional
requirements in the human diet as they provide the essential
minerals, vitamins and fiber required for maintaining health.
They provide energy, help in body building and protecting body
from many malnutritional problems.
There is huge production of Fruits and Vegetables in India which
is sufficient for all but an estimated 30-40% of fruits and vegetables
are lost between harvest and consumption due to several factors
including inadequate infrastructure, subpar transportation, limited
knowledge of post-harvest handling, market inefficiencies , and
technological gaps.

INTRODUCTION TO POST HARVEST
MANAGEMENT OF FRUIT CROP
It is an inter-disciplinary science and techniques applied to agricultural
or horticultural commodities after harvest for the purpose of
preservation, conservation, quality control/enhancement, processing,
packaging, storage, distribution, marketing, and utilization to meet the
food and nutritional requirements of consumers in relation to their
needs.

 Ensuring Quality and self
life
Maintaining fruit quality and extending
edible life after harvest is crucial for
marketability and reduce wastage
 Market Demand & Consumer
Expectations
Meeting market demand and consumer
expectations for fresh, high-quality
produce is essential.
 Preservation of Freshness
Post-harvest management aims to
maintain the freshness and quality
of fruit crops after harvesting.
Importance

CHALLENGES IN POST HARVEST
MANAGEMENT
ADDRESSING KEY HURDLES IN FRUIT CROP MANAGEMENT
• Infrastructure and labour Limitations
Inadequate storage and transportation infrastructure pose significant challenges in
post-harvest management. Skilled labour is required which is not generally
available.
• Pest and Disease Control
Managing pests and diseases during storage and transportation is critical for quality
preservation. Due to human error infected fruit may be ignored.
• Temperature and Humidity Control
Maintaining optimal storage conditions in varying climates is a major challenge for
fruit crop management. As manually it is difficult to maintain this at a large scale.

 Market Volatility
Adapting to fluctuating market demands, price variations, and
consumer preferences.
 Quality ensurance :
It is not possible to maintain quality of fruits over a large scale. Due to
human error there can be ignorance to quality of some fruits.

The postharvest stage is the final and most critical in
agriculture and requires close attention because time
and money have been used to cultivate food products.
An ineffective postharvest stage or negligence may
result in severe postharvest losses and consequent
financial loss (Prusky, 2011).

Overview of AI
and Its
Applications in
HORTICULTURE
AI is a general term that
includes machine learning (ML),
Computerized study and
sensory technology
The term ‘Artificial Intelligence’ was
coined by John McCarthy in 1950
The field of computer science known
as artificial intelligence (AI), also
referred to as machine intelligence,
teaches machines how to mimic human
physical movements and respond in
human-like ways.

o AI in Precision Agriculture
AI enables precision farming, optimizing resource use and
enhancing crop yield and quality.
o Predictive Analytics
AI algorithms predict crop growth, disease outbreaks, and
optimize harvest timings, shelf life of fruits.
o Robotic Automation
AI-powered robots assist in planting, harvesting, and sorting
fruits, improving efficiency.

use of AI in post harvest storage of fruit crops.pptx

AI tools can be used individually but are increasingly used in various
combinations to tackle more complex tasks.
Availability of smart machines, IoT, and blockchain technologies support AI
analytics implementation in agri-food systems.
The huge data volumes are collected in various formats, kinds, and content. The
data that has to be analyzed onsite can be heterogeneous (or multivariate) and
have noise and redundancy.
Organizing and cleaning data is necessary for traditional data analytics and
requires time. Nowadays, AI tools that are automated can save time for these
tasks.

Role of AI in Post Harvest Management of Fruit
Crops
•AI enables precise and rapid assessment of fruit quality,
identifying defects and maturity.
•AI-powered image recognition systems can be employed to
assess the quality of fruits based on their size, color, shape,
and external defects.
•Machine learning algorithms can learn from vast datasets to
identify defects that may not be easily detectable by human
eyes.
1. Smart Quality Assessment
Artificial Intelligence (AI) can play a significant role in improving post-harvest
management of fruits by enhancing efficiency, reducing waste, and ensuring
better quality.

2. Automated Sorting and Grading
•AI-based systems classify fruits based
on size, color, and quality, streamlining
the sorting process.
•This ensures that fruits are distributed
accurately, meeting market demands
and reducing the chances of spoilage.

3. Predictive Storage Models
•AI develops predictive models for optimal storage
conditions, reducing spoilage and wastage.
•AI can analyze historical data, weather conditions,
and other relevant factors to predict optimal harvest
times.
•This helps in planning harvest schedules, ensuring
fruits are picked at the right time for maximum
quality and shelf life.

4. Inventory Management:
•AI can be used for real-time monitoring of inventory levels and
predicting demand patterns.
•This helps in preventing overstocking or understocking issues,
reducing waste and ensuring a steady supply of fresh fruits.
5. Quality Monitoring during Storage:
•Continuous monitoring using sensors and AI can detect signs of
decay, ripening, or spoilage during storage.
•Automated alerts can be generated to take corrective actions,
such as adjusting storage conditions or expediting the distribution
process.

6.Smart Packaging:
• AI can be used to develop smart packaging solutions equipped with
sensors and RFID tags that monitor the condition of fruits during
storage and transportation.
• These smart packages can provide real-time data on temperature,
humidity, and other environmental factors, ensuring the quality and
freshness of the fruits until they reach the consumer.

AI-based Techniques for Fruit Quality
Assessment
o Computer Vision Systems AI-driven systems categorize
fruits based on size, color, and external quality attributes.
o Sensor Technologies AI-based grading assigns fruit
quality grades based on predefined parameters and
standards.
o Machine Learning Algorithms AI enables real-time
decisions for sorting and grading, optimizing the process
efficiency.

Machine vision system
Quality check of Mango

Computer Vision Systems
There are lots of techniques to identify diseases in fruits in its early stages. The old
method of disease detection in fruit is naked eye observation and it‘s not effective.
Using digital method, the disease detection can efficient, accurately, time consuming
is less, saves time.
Image Recognition: Computer vision techniques, often powered by deep learning algorithms,
can analyse images of fruits to assess their quality based on factors such as size, colour, shape,
and external defects.
Convolutional Neural Networks (CNNs): CNNs are effective in feature extraction from images,
making them suitable for tasks like fruit quality classification and defect detection.

BASIC STEPS OF IMAGE PROCESSING
Step1: Image Acquisition: This is the first step of image
processing in which camera is used for capturing fruits images in digital form
and store in any digital media.
Step2: Image Pre-processing: This section removes noise,
smoothen the image also perform resizing of images. RGB images are
converted to the grey images also contrast of image is increased at certain
level.
Step3: Image Segmentation: Segmentation is used for
partitioning an image into various parts.
Step4: Feature Extraction: This section is used for obtaining
features like color, texture and shape which reduce resources to describe large
set of data before classification of image.
Step5: Classification: This section analyzes numerical
property of image features and organize its data into categories. It use neural
network which performs training and classification of fruits diseases.

Different image processing techniques and lots of algorithms
have been developed by researchers with the help of
MATLAB software for accurate fruit disease identification.
There are a lot of models use for classification :
Fuzzy Logic, Artificial Neural Network, Support Vector
Machine and Adaptive Network-based Fuzzy Interference
System.

A.Fuzzy Logic model :
Fuzzy systems provide the means of translating the expert knowledge of
humans about the process in terms of fuzzy (IF–THEN) rules. A fuzzy rule is
the basic unit to gain knowledge in fuzzy systems.
• Kavdir et al. proposed a method of apple grading in colour,size
and defects of apples are extracted. These features were
gathered and evaluated using fuzzy system and this gave 89%
accuracy in classification.
• Date fruits were graded using Fuzzy by extracting some features
like quality of juice, size and freshness and it gave 86% accuracy.
• Rokunuzzman et al. presented an algorithm to classify tomatoes
and it gave 84%accuracy rate.

B. Artificial Neural Network (ANN) :
ANN is massively parallel distributed information processing system that is
made up of artificial neurons has certain performance characteristics
resembling biological neurons of the human brain.
• Mustafa et al. proposed a method to determine the size and
ripeness of banana. Shape and colour features were extracted.
Then ANN was used for classification and it gave accuracy of 79-
90%.
• Rokunuzzman et al. presented an algorithm to classify tomatoes
ANN also and it gave 87.5% accuracy rate which was more than
the accuracy rate given by Fuzzy logic.
• Alipasandi et al. proposed a method in which peach was classified
and it gave 99.3% accuracy rate.

C. Adaptive Neural Fuzzy Interference System:
ANFIS is a fusion of artificial neural network and fuzzy
interference system.
Nozari et al. presented an algorithm for grading of
Mozafatidates and classified these based on weight, length,
width and thickness. These fruits were graded using both
ANFIS and human experts for comparison and ANFIS
showed 93.5%accuracy as compared to human experts.

D. Support Vector Machine:
SVM has the highest accuracy rate as compared to other technique.
Suresha et al. proposed an Apple classifier using SVM and it gave
100% accuracy when only colour features are compared.
Zheng et al. presented an algorithm of mango grading. Fractal
dimension and L*a*b* colour model are used for grading purposes.
Using SVM, fractal dimension and colour gave 85.19%and 88.89%
accuracy respectively.

Internet of Things (IoT): Smart sensors can be deployed in storage facilities to
monitor various parameters like temperature, humidity, and gas levels. AI
algorithms analyze the sensor data in real-time to ensure optimal storage
conditions and prevent quality deterioration
Sensor Technologies
In the fruit cold chain field, sensing technology typically involves monitoring and
controlling the entire fruit supply chain by sensing multiple source parameters using
rigid sensors.
However, traditional rigid sensors face various limitations.
Monitoring the environment in the cold chain using sensor technology has become
increasingly popular, for example, freshness monitoring of blueberries in cold chain
logistics (Huang, Wang, Zhang, Xia, & Zhang, 2023), wireless monitoring of post-
harvest peach quality (X. Wang, Fu, Fruk, Chen, & Zhang, 2018), non-destructive
wireless real-time monitoring of grape quality (X. Wang, He, Matetic, Jemric, & Zhang,
2017).

Here are some key sensor technologies commonly used in post-harvest management:
1. Temperature and Humidity Sensors:
- Purpose: Monitor and control the storage environment to prevent temperature fluctuations
and humidity levels that could lead to premature ripening, decay, or fungal growth.
- Application: Cold storage facilities, shipping containers, and transportation vehicles.
2. Ethylene Sensors:
- Purpose: Detect and measure ethylene gas, a plant hormone responsible for fruit ripening.
Monitoring ethylene levels helps in managing the ripening process and optimizing storage
conditions.
- Application: Ripening rooms, storage facilities.
3. Gas Sensors:
- Purpose: Measure the concentration of gases such as oxygen and carbon dioxide.
Monitoring these gases is critical for managing the respiration rate of fruits and preventing
anaerobic conditions.
- Application: Controlled atmosphere storage, modified atmosphere packaging.

4. Weight Sensors:
- Purpose: Monitor the weight of fruit batches to track inventory levels and
assess changes in moisture content or freshness.
- Application: Conveyor belts, packaging lines.
5. Acoustic Sensors:
- Purpose: Detect internal quality attributes by analyzing the sound produced
when tapping or vibrating a fruit. This can help identify issues like internal
bruising or ripeness.
- Application: Quality assessment during sorting processes.
6. Color Sensors:
- Purpose: Measure the color of fruits to assess ripeness and external quality
attributes.
- Application: Sorting and grading machines.

7. Moisture Sensors:
- Purpose: Measure the moisture content of fruits to prevent issues like
dehydration or excessive moisture, which can lead to decay.
- Application: Storage facilities, packaging lines.
8. RFID (Radio-Frequency Identification):
- Purpose: Enable real-time tracking and traceability of individual fruit
batches throughout the supply chain. RFID tags can store information
such as origin, harvest date, and storage conditions.
- Application: Inventory management, traceability systems.
9. Image Sensors (Cameras):
- Purpose: Capture high-resolution images for computer vision
applications, allowing for the visual inspection and quality assessment of
fruits.
- Application: Sorting machines, quality control processes.

10. Vibration Sensors:
- Purpose: Monitor vibrations in storage or transportation equipment to
identify potential issues, such as excessive mechanical stress or damage to
fruits.
- Application: Conveyor systems, transportation vehicles.
11. UV and NIR Sensors:
- Purpose: Measure ultraviolet (UV) and near-infrared (NIR) light absorption
to assess internal fruit quality attributes, such as sugar content and ripeness.
- Application: Spectroscopy systems, non-destructive quality assessment.

New sensor system developed by scientists at the
Leibniz Institute of Agricultural Engineering and
Bioeconomics can automatically and continuously
measures the oxygen consumption and carbon
dioxide production of fresh products in the
warehouse or in the packaging.
Some Sensors are there that look like a
fruit and act like real one to monitor
storage and predict storage life.

Machine Learning Algorithms
Machine learning can also monitor environmental conditions such as temperature
and humidity and adjust the environment to optimize fruit preservation. ML helps to
predict the optimal time for harvest, packaging, and distribution of the fruit to ensure
it is fresh for the customer.

The basic principle of machine learning applied to ready-to-eat fruits is
to use large amounts of data to identify patterns, apply algorithms, and
then create models that can be used to make predictions.
This model can then be applied to data gathered from numerous sources
to make better decisions about the quality of the fruit or even the best
time of year to purchase it.
ML is used in fresh-cut fruits in various ways.
It has the potential to monitor the quality of fresh-cut fruit by detecting
and analyzing any visible defects.
It can also be used to determine the physical properties of the fruits and
their freshness. It can also be used to track the freshness of stored fruit
and alert the user when it is no longer in optimal condition.
This technology is used to forecast the life span of fresh-cut fruit based
on its current condition.

Taxonomical hierarchy of machine learning.

The future of machine learning in the food industry is very
promising. ML helps to improve food production, increase efficiency,
and reduce waste. It can also be used to create new recipes and
improve existing recipes.
By analyzing customer data, companies can identify trends and
create personalized menus and recipes tailored to the customer’s
tastes.
It is also used to increase the accuracy of food quality control,
optimize food packaging and storage, and help reduce food waste

AI-based Predictive Models for Fruit Storage and
Transportation
Climate Control Systems
AI models optimize storage conditions by
regulating temperature, humidity, and gas
levels
Route Optimization
AI algorithms determine the most
efficient transportation routes and
conditions for fruit transit.
Real-time Monitoring
AI-enabled monitoring systems
track and adjust storage and
transit parameters in real time.

Smart Packaging
Fruits packaging is a critical step toward the transport and
sale of produce, and it can be categorized into two main types
based on its objectives: consumer packaging and
transportation packaging.
Consumer packaging is the individually packaged collection of
fruits that are sold as a single unit, which are usually made of a
plastic film or container.
On the other hand, transport packaging is designed to allow
large quantities of produce to be transported easily, efficiently,
and safely.

Consumers can make informed decisions about the ripeness
state of the fruit by simply noting the color of the sensor.
different stages of fruit ripeness may also release volatile
compounds that have been studied using e-noses.

(A) Reaction setup for the
detection of apple flavor using
a sensor label
(B) degree of sensory
ripeness and RGB index plot
vs storage time, and
(C) color changes in sensor
label after exposure to apple
flavors, reproduced with
permission from ref

Improvement of Sensor Technology.
One of the main limiting factors for the further
development of smart packaging is the limitation in sensory
technology. While there are many sensors that could give
accurate readings for different freshness parameters
observed in produce and other food products, their usability,
specifically size and cost, makes it difficult to incorporate
them in large-scale packaged containers. The balance
between functionality, portability, and size has always been
an area of interest in technology.

CASE STUDY -1
NEOLITHICS LTD. (U.S.) USE AI TO SCAN WIDELY
CONSUMED PRODUCE LIKE APPLES,
• A single person inspects 15 to 20 tons of produce per hour by hand. Sorting through over a
hundred tons of produce a day at a rapid pace is a dull job, and workers are understandably
prone to making mistakes and missing defects.
• “Packing houses are the last systematic link to look at quality before products reach consumers,
but their analyses are not thorough enough,” says Amir Adamov, CEO of Neolithics.
• The Herzliya-based company develops software that is implemented in sorting and automation
equipment in warehouses and packing houses. Its AI measures the physical appearance of the
product and whether it suits what the market wants in terms of look, shape and size, as well as
its nutritional parameters, including starch, fats, sugar, biomatter and fibers.
• The software has a 90 per cent accuracy, though it varies with the product.

1st stage
Pictures and analysis of
morphological characters

Challenges in use of AI storage of fruit crops
Data Quality and Availability:
AI models rely heavily on high-quality and diverse
data for training. In the post-harvest phase,
obtaining comprehensive and accurate data can
be challenging due to variations in crops, weather
conditions, and storage environments.
Adaptability to Diverse Crops and Conditions:
•Different crops and conditions require specific post-
harvest treatments. Developing AI models that can
adapt to the diversity of crops, storage facilities, and
environmental factors is a complex task.
Ethical and Social Concerns:
•The use of AI in agriculture, including post-harvest
technology, raises ethical considerations such as
data privacy, ownership of data, and potential
displacement of labor. Ensuring fair and responsible
AI practices is crucial.

Interdisciplinary Collaboration:
•Successful AI applications in post-harvest
technology often require collaboration between
agronomists, data scientists, engineers, and other
experts. Bridging the gap between these disciplines
and fostering effective communication can be a
challenge.
Resource Constraints:
•Small-scale farmers, who make up a significant
portion of the agricultural sector, may face
resource constraints in terms of finances,
technology literacy, and access to modern
equipment. Implementing AI solutions may
require significant investments that are beyond
the means of these farmers.

FUTURE OF AI IN STORAGE OF FRUITS CROPS:
• Because of high investment still lower farmers are not opting these so
working on it can give chance to be used by them.
• There is great future of AI in storage of fruits crops as reduction of
losses manually is not possible.
• AI systems can continuously learn and adapt to the specific
requirements of different fruits. By analyzing data on how different fruits
respond to varying storage conditions, adaptive control systems can
optimize parameters to extend shelf life and maintain quality.

Ongoing research and collaboration between agricultural
scientists, engineers, and AI experts will likely lead to the
development of more sophisticated and specialized AI
applications for fruit crop storage, addressing specific
challenges and improving overall outcomes.

Reference
• Ahmad, I.; Benjamin, T.A. Application of Artificial Intelligence and Machine Learning
to Food Rheology. Adv. Food Rheol. Appl. 2023, 8, 201–219.
• Arif U. Alam , Pranali Rathi, Heba Beshai , Gursimran K. Sarabha and M. Jamal
Deen Fruit Quality Monitoring with Smart Packaging Sensors 2021, 21, 1509.
https://doi.org/10.3390/s21041509.
• Aherwadi N, Mittal U, Singla J, et al (2022) Prediction of Fruit Maturity, Quality, and
Its Life Using Deep Learning Algorithms. Electronics (Switzerland)
11:. https://doi.org/10.3390/electronics11244100
• J. K. Patil, R. Kumar, ―Advances in Image Processing for Detection of Plant
Diseases,‖ in Journal of Advanced Bio informatics Applications and Research ISSN,
vol2(2), pp.135-141, June 2011..
• Kumar Vipin and singh Sudhanshu Artificial intelligence in horticulture crops, Annals
of Horticulture September 2023 DOI: 10.5958/0976-4623.2023.00014.2
• Monika Jhuria, Rushikesh borse, Ashwani Kumar ―Image Processing for Smart
Farming: Detection of Disease and Fruit Grading‖ Proceeding of the IEEE Second
International Conference on Image Information Processing, pp. 978 -1-4673-6101,
2013.

• Raja Sekar L, Ambika N, Divya V and Kowsalya T,- Fruit Classification System
Using Computer Vision: A Review. InternationalJournal of Trend in Research
and Development, Volume 5(1), ISSN: 2394-9333 www.ijtrd.com IJTRD | Jan -
Feb 2018.
• Sa I, Ge Z, Dayoub F, et al. Deepfruits: a fruit detection system using deep
neural networks. Sensors. 2016. https://doi.org/10.3390/S16081222
• Sherlin Varughese, Nayana Shinde, Swapnali Yadav, Jignesh Sisodia
―Learning-Based Fruit Disease Detection Using Image Processing‖
International Journal of Innovative and Emerging Research in Engineering
Volume 3, Issue 2, p-ISSN: 23945494,2016.

use of AI in post harvest storage of fruit crops.pptx

More Related Content

What's hot (20)

Similar to use of AI in post harvest storage of fruit crops.pptx (20)

Recently uploaded (20)

use of AI in post harvest storage of fruit crops.pptx