Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)

IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

Classification of Automatic Detection of Plant

Disease in Leaves and Fruits using Image
Processing with Convolutional Neural Network
2023 4th International Conference on Smart Electronics and Communication (ICOSEC) | 979-8-3503-0088-8/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICOSEC58147.2023.10275843

1S. 2Veeramreddy 3Bhukya

Muthuselvan, Rajasekhar, Shankar,
Department of Information Technology, Department of Computer Science Department of Electronics and
KCG College of Technology, Chennai, Engineering, Malla Reddy Engineering Communications Engineering,
Tamil Nadu, India College, Kompally Rangareddy Dt, CVR College of Engineering,
[email protected] Secunderabad, Hyderabad, India Ibrahimpatnam, Telangana, India,
[email protected] [email protected]
4P. Mohana,

Scientist - C, Centre for Remote

Sensing and Geoinformatics, 5Doddabasappa N, 6Badugu Suresh,

Sathyabama Institute of Science and School of Electrical and Electronic Department of Electronics and
Technology, Rajiv Gandhi Road, Engineering, REVA University, Communications Engineering,
Chennai - 119, Tamilnadu, India, Bengaluru, Karnataka,India, Koneru Lakshmaiah Educational
[email protected] [email protected] Foundation, Vaddeswaram, Guntur,
Andhra Pradesh, India
[email protected]

Abstract—Detection and identification of disease in plants is the

fundamental step to avoid various constraints and to enhance Farming is one of the important occupations in rural
productivity. In agricultural farm, the crops are prone to areas. It is done based upon the climatic circumstances and
various bacterial and fungal disease. It is initiated due to the soil conditions. It also includes various raw material for
varied ecological conditions and soil categories. The earlier
prediction of diseases helps to prevent the complete degradation
industries. In India, nearly 75% of population depend upon
of crops. The disease is detected through small patches on the agriculture both directly and indirectly. India is the second
leaves of the plant. The evolution of automation and innovations largest manufacturer of rice and wheat [2]–[4]. The
helps in the earlier detection of disease in the leaves and fruits of Government provides various schemes and policies to
the crops. It is achieved by artificial intelligence. This is done develop agriculture in a sustainable manner. The user friendly
through the image processing techniques. It uses the comparison manner of cultivation is implemented to maintain a balanced
of healthy plant leaves to that of the infected leaves and fruits. ecosystem. This helps to obtain the food crops with higher
The classification of disease in the fruits and leaves are done nutritious content with rich quality [5]–[7].
accurately through the convolution neural network. This helps
to detect the type and cause of the bacterial and fungal
infections. This includes image pre-processing, data acquisition
The Government forms rules to enact crop
and image classification with feature extraction. Thus the CNN management system with large scale ecosystem through
model helps to detect and classify the disease in plant and fruits innovative technologies. They are adopting digitalization in
in agricultural farm. agricultural sectors through various technologies are
advanced equipment’s to enhance the productivity yield.
Keywords—Productivity, bacterial infections, pest, artificial Numerous techniques are implemented such as monoculture
intelligence, deep learning, image processing, convolutional farming with reduced water consumption. The monoculture
neural network farming is referred as the farming and cultivation of a single
I. INTRODUCTION crop in a farm land to adopt increased productivity. The
economic conditions of one nation is dependent upon the flora
and fauna. This also includes the flowering and non-flowering
In India, the agriculture forms a prominent role in plants [8]–[11].
enhancing the GDP of the country. The agriculture is defined
as the art of cultivation, increasing livestock and production. Crops are cultivated based upon the climatic and
This includes adoption of plant animal foodstuffs to raise the economic conditions of the particular environment. The
living. This helps to improve the overall standard of a farmers are also adopting greenhouse farming to attain a
community [1].

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1408

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

greener ecosystem. There are various crops such as stable

crops and commercial crops. The stable crops are important
for livelihood and commercial crops increase the lifestyle of
a particular community [12]–[15].

Thus the sustainability and environmental factors

play an important aspect in increasing the agricultural
products. To overcome the consequences in the conventional
and traditional methods in agriculture, the innovative
technologies are implemented. Innovation and automation
plays an important role in agriculture. This helps to reduce
man power and improved through time saving parameters to
enhance higher efficiency. The automation helps to increase
the productivity. The important factor includes the selection
of crops based upon the prevailing conditions. This also
includes numerous challenges in cultivation. This leads to Fig 1: Various diseases in leaves and fruits
protection of crops from various external disturbances such
as insects, bacterial and fungal disease. These disease leads to The figure 1 represents the various diseases in leaves and
reducing the productivity which leads to huge loss. The initial fruits in agricultural farm land. The external appearance
prediction of these disease may lead to reduction of various denotes the type of bacterial and fungal infections. The
loss. This is done through artificial intelligence. The artificial various infections are early and late blight in potato leaves,
intelligence is the process of training the machines to function citrus disease, alternaria of pomegranate leaves and bacterial
as similar to that of human intelligence [16], [17]. blight [16].

They are programmed to function without the aid of These infections are detected through the black spots
human interaction. They help to achieve an optimum solution and immature leaves at the top. The infections are caused by
for the complex problems through optimization algorithms. insect’s fungi and bacteria. Controlling the infections may
These optimization algorithm helps to achieve the solution leads to improve the agricultural productivity [25]. The
through the computational process takes place in the hidden disease varies from plant to plant based upon the climatic and
layers through processing them with data pre-processing environmental factors. In apple, black spots, dark patches and
techniques. The extraction of original numerical data from mosaic and rust and the symptoms of infections. In citrus
raw data helps to achieve the optimum solution. These fruits, lesions and brown spot determines the effects of
artificial intelligence composition of significant part in the disease that leads to degrade the overall productivity.
advancement of crop yields [18]–[21].
Thus by using automation in the agricultural field
The deep learning is the subset of machine learning leads to the detection of diseases and infections at an earlier
that customs various layers to detect the solutions. It uses the stage. Thus it is a promising efficient way to detect the
convolution neural network to extract the optimum solutions. constraints. The initial detection of disease is done through
They are used to diagnose the disease in a faster way. The the analysis of leave spots and black lesions on the fruits. At
convolutional neural network helps to improve the efficiency an initial stage they are detected through the image processing
and performance parameters. This helps to detect the disease techniques. The convolutional neural network helps to
at an earlier stage. This helps to develop the productivity with identify the disease through involving various layers to detect
automatic decision making in production of agricultural the exact cause of the bacterial and fungal infections. The
products. Thus the detection of disease in the leaves and fruits recognition of disease at an earlier phase helps to except the
are detected through image processing techniques. The crops without any issues in the productivity. These
infected plant is identified through marks on the leaves and automations also help in the removal of weed in the
fruits. There are various kinds of bacterial infections and agricultural farm yard. The growth of weeds in the
disease happen in the agricultural farm [22]–[23]. agricultural land is also a driving factor that affects the
productivity. This leads to wastage of water to the unwanted
These diseases are detected through comparing with plants in the agricultural land. They are done through the
the healthy plant individual to that of the infected part of the image processing techniques. There are various infections
plants. The fundamental symptoms of disease occur on the and disease occur in the agricultural crops. They must be
leaves and fruits. This indicates the initial condition of the evaluated keenly to solve the issues.
infections to take place. The detection and identification of
disease are done on the leaves. Identification of the infections Thus the automatic detection of plant disease
with naked eye are impossible and thus the automations with through artificial intelligence techniques helps in versatile
intelligence are adopted to accurately find out the important ways. Due to the advancement of technological innovations,
cause for the disease. the e-commerce is implemented in which the farmers can
purchase the pesticide to control the disease through online
platform. This helps them to save time and cost.

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1409

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

II. PROPOSED SYSTEM

In this work, the proposed system uses deep learning with

convolutional neural network to analyse and detect the
diseases in leaves and fruits. This is implemented to save cost
and time when compared to the existing system. It helps to
improve the quality of the agro products. This helps in the
recognition of the infections and diseases at an earlier period.
They are done through determining efficient training and
learning stages with optimization. This helps in residual
interconnections. The conventional method is replaced
through the depth wise convolutional methods. This helps to
detect the root cause of the disease accurately. The detection
of disease involves various stages. These stages helps to
determine the cause for the attack of bacterial and fungal
infections.

Input image collection

Pre-processing Fig 3: Convolution neural network

The figure 3 represents the functioning of convolutional

Segmentation of image neural network.

Feature extraction and classification Thus the spreading and allowing the data to one
layer to another layer in the neural network are denoted as the
feed forward network. Thus by using the weights and bias the
Optimum results cost parameter can be minimized. The extraction of
information is done through the input images and decreased
Fig 2: Stages in plant disease detection through the pooling layer. They are implemented through k-
means algorithm in the convolutional neural network.
The figure 2 represents the various periods in the plant disease
identification system. This includes the collection of images
and processing them to extract the optimum output.

III. IDENTIFICATION OF DISEASE THROUGH

CONVOLUTIONAL NEURAL NETWORK

The convolutional neural network is a subset of

artificial neural network. Artificial intelligence is used to
perform the task through extracting the knowledge through
human intelligence. The CNN includes image processing and
feature extraction techniques. The extraction of information
through computer vision is termed as the feature extraction.
The convolutional neural network comprises of three layers
namely the input layer, hidden layer and the output layer. The
information is processed in the hidden layer to produce the
optimum results. They are done with the weights and bias.
These values are multiplied and proceeded to obtain the
activation function. Determining the initial clusters of citrus Fig 4: K-means algorithm
plant diseases involves identifying and grouping citrus plants
that exhibit similar symptoms. The figure 4 represents the k-means algorithm
flowchart

(i)Input image collection

The image collection is the important step in the pre-

processing and validation of output. The collection of data

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1410

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

includes both the healthy and infected individuals of the The table I represents the Collection of dataset in agricultural
plants. This helps to differentiate the infected part through farm land. These images are obtained in pixels’ value. The
various stages. This includes input images proceeded with dimension of the obtained images is deducted through the
testing and training the data. These dataset is used in down sampling values.
augmented images that are trained with the healthy individual
with classification of classes [23]. The duplicate images are (iii)Segmentation of image and classification
rejected through comparison of size, shape and color.
The gathering of accurate information through the
(ii)Pre-processing and feature extraction images are done through segmentation process. This helps in
obtaining a detailed information. This is the important part in
the feature extraction process. The important process
involved in the segmentation of image includes the k-means
clustering process. This helps to obtain a clear analysis of
information from the image. This is done through the k-means
algorithm.

Fig 5: Data pre-processing

The figure 5 represents the data pre-processing

through k-means algorithm. The pre-processing the process
of preliminary analysis for the feature extraction process. This Fig 6: Segmentation of image
helps to classify and neglect the duplicate images. The
segmentation process is a fundamental task in image The figure 6 demonstrates the segmentation of image through
processing and computer vision. It involves partitioning an k-means algorithm. The k-means clustering is an
image into multiple regions or segments to extract meaningful unsupervised learning algorithm. They does not have the
information from the image. division of data and functions through the division of images
obtained. This is extracted through the non-overlapping sub
These images are decoded to obtain the exact category of images termed as clusters. Thus by repeating the
images. The feature extraction is the process of obtaining centroid with multiple iterations helps to obtain the optimum
information from the raw data. They are used to format the outcome.
images for further training and testing process [24]. This
includes the classification of color, shape and texture of
leaves and fruits. The feature maps are obtained through the
pooling layers. The rectified linear units and the flattening
techniques are used to achieve and modify the negative values
to zero.

TABLE I. Dataset obtained from farm

Section Infection Count

Citrus leaves and Greening 240
fruits Black spot 290
Canker 564
Melanoses 140
Healthy 650
Early and late Greening 340
blight of potato Dark patches 863
leaves Brown patches 250
Healthy 760
Alternaria of Brown and white 150
pomegranate patches Fig 7: CNN model
leaves
Bacterial blight Dark brown 289 The figure 7 represents the convolutional neural network
of pomegranate patches model to determine the infected leaves and fruits at an earlier
leaves stage through k-means algorithm.

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1411

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

TABLE II. CNN model detection

Species Data source Network Accuracy in

for detection model detection
(part)
Apple Fruit part CNN 94%
Wheat Leaves CNN 96%
Rice flower DCNN 92%
Tomato Fruit DCNN 98%
Maize Leaves CNN 90%
Guava Leaves MCNN 94% Fig 9: Data pre-processing

Sugarcane Stem CNN 95% The figure 9 demonstrates the data pre-processing in matlab.
This includes the training of dataset through the stored
images.
The table II determines the CNN models for the detection of
disease in the various parts of plant. These are classified and
detected at an early stages based upon the previous raw data.
This shows the accuracy of outcome.

90
80
70
60
50 Images Fig 10: Image recognition and classification
40 Healthy ones
The figure 10 demonstrates the image recognition and
30 Infected images classification through the dataset. This includes the various
20 images in pixels with augmented form.
10
0
Class 1 Class 2 Class 3 Class 4

Fig 8: Classification of images

The figure 8 demonstrates the classification of images based

upon healthy and infected images. These infected images are
detected in the dataset. The earlier prediction of infected ones
can prevent the healthier ones. These are analysed through the
accuracy with fine tuning parameter. The augmented images
tend to perform numerous iterations to obtain higher
accuracy. The number of iterations helps to obtain the Fig 11 : Segmentation
optimum results.
The figure 11 represents the progression of segmentation with
feature extraction. The segmentation is done to obtain the
IV. SIMULATION RESULTS clear augmented images in the dataset.

The detection and recognition of infected leaves and fruits are

detected through convolutional neural network. This is
simulated in matlab Simulink. This is accompanied with
image processing techniques.

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1412

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

REFERENCES
[1] N. Sajitha, S. Nema, K. R. Bhavya, P. Seethapathy, and K. Pant,
“The Plant Disease Detection Using CNN and Deep Learning
Techniques Merged with the Concepts of Machine Learning,”
2022 2nd International Conference on Advance Computing and
Innovative Technologies in Engineering, ICACITE 2022, pp.
1547–1551, 2022, doi: 10.1109/ICACITE53722.2022.9823921.
[2] Y. Sun, Y. Liu, G. Wang, and H. Zhang, “Deep Learning for Plant
Identification in Natural Environment,” Computational
Intelligence and Neuroscience, vol. 2017, 2017, doi:
10.1155/2017/7361042.
[3] Y. Guo et al., “Plant Disease Identification Based on Deep
Learning Algorithm in Smart Farming,” Discrete Dynamics in
Nature and Society, vol. 2020, 2020, doi: 10.1155/2020/2479172.
[4] S. S. Chouhan, A. Kaul, U. P. Singh, and S. Jain, “Bacterial
Fig 12: Prediction of disease in leaves foraging optimization based radial basis function neural network
(BRBFNN) for identification and classification of plant leaf
diseases: An automatic approach towards plant pathology,” IEEE
The figure 12 represents the prediction of diseases through Access, vol. 6, pp. 8852–8863, 2018, doi:
iterations. This is detected at the earlier stage of bacterial or 10.1109/ACCESS.2018.2800685.
fungal infections. [5] E. Fujita, Y. Kawasaki, H. Uga, S. Kagiwada, and H. Iyatomi,
“Basic investigation on a robust and practical plant diagnostic
system,” Proceedings - 2016 15th IEEE International Conference
on Machine Learning and Applications, ICMLA 2016, pp. 989–
992, 2017, doi: 10.1109/ICMLA.2016.56.
[6] J. Kolli, D. M. Vamsi, and V. M. Manikandan, “Plant Disease
Detection using Convolutional Neural Network,” 2021 IEEE
Bombay Section Signature Conference, IBSSC 2021, pp. 1473–
1476, 2021, doi: 10.1109/IBSSC53889.2021.9673493.
[7] N. Gobalakrishnan, K. Pradeep, C. J. Raman, L. J. Ali, and M. P.
Gopinath, “A Systematic Review on Image Processing and
Machine Learning Techniques for Detecting Plant Diseases,”
Proceedings of the 2020 IEEE International Conference on
Communication and Signal Processing, ICCSP 2020, pp. 465–468,
2020, doi: 10.1109/ICCSP48568.2020.9182046.
[8] Z. Liu, R. N. Bashir, S. Iqbal, M. M. A. Shahid, M. Tausif, and Q.
Umer, “Internet of Things (IoT) and Machine Learning Model of
Fig 13: Detection of disease Plant Disease Prediction-Blister Blight for Tea Plant,” IEEE
Access, vol. 10, pp. 44934–44944, 2022, doi:
10.1109/ACCESS.2022.3169147.
The figure 13 demonstrates the detection of disease. This [9] D. Wu, X. Han, G. Wang, Y. Sun, H. Zhang, and H. Fu, “Deep
helps to provide the exact pesticide to the plant in agricultural Learning with Taxonomic Loss for Plant Identification,”
farm land. Computational Intelligence and Neuroscience, vol. 2019, 2019,
doi: 10.1155/2019/2015017.
V. CONCLUSION [10] G. Madhulatha and O. Ramadevi, “Recognition of plant diseases
using convolutional neural network,” Proceedings of the 4th
International Conference on IoT in Social, Mobile, Analytics and
Frequent climatic changes and pollution leads to Cloud, ISMAC 2020, pp. 738–743, 2020, doi: 10.1109/I-
SMAC49090.2020.9243422.
degradation in the quality of crops that are affected by the [11] V. Rajesh Kumar, K. Pradeepan, S. Praveen, M. Rohith, and V.
bacteria and fungus. This leads to the destruction of Vasantha Kumar, “Identification of Plant Diseases Using Image
productivity and adversely affects the human health. To avoid Processing and Image Recognition,” 2021 International
those circumstances, the advanced technology with artificial Conference on System, Computation, Automation and Networking,
ICSCAN 2021, pp. 0–3, 2021, doi:
intelligence are implemented to gain higher yield. These 10.1109/ICSCAN53069.2021.9526493.
innovations in technology play a significant role in the [12] H. Wang, G. Li, Z. Ma, and X. Li, “Image recognition of plant
agricultural production sectors. This includes advanced diseases based on backpropagation networks,” 2012 5th
equipment for various activities. The convolutional neural International Congress on Image and Signal Processing, CISP
2012, no. Icnc, pp. 894–900, 2012, doi:
network with image processing techniques are used to detect 10.1109/CISP.2012.6469998.
the earlier prediction of leaves and fruit diseases in plant. This [13] L. Xu, B. Cao, F. Zhao, S. Ning, P. Xu, and W. Zhang,
is done through various stages through augmented images. It “Physiological and Molecular Plant Pathology Wheat leaf disease
includes k-means clustering algorithm to obtain the exact identification based on deep learning algorithms,” Physiological
and Molecular Plant Pathology, vol. 123, no. September 2022, p.
information as outcome. This helps to detect the kind of 101940, 2023, doi: 10.1016/j.pmpp.2022.101940.
infections and can able to provide the appropriate pesticide to [14] P. Sharma, K. Choudhary, K. Gupta, R. Chawla, D. Gupta, and A.
prevent the disease to spend in the overall farm. Thus the Sharma, “Artificial plant optimization algorithm to detect heart rate
innovative development in the agricultural sectors through & presence of heart disease using machine learning,” Artificial
Intelligence in Medicine, vol. 102, no. November 2019, p. 101752,
artificial intelligence through convolutional neural network 2020, doi: 10.1016/j.artmed.2019.101752.
with image processing is way form in upcoming decades to [15] J. K. Kiruthika, M. Sudha, G. Divya, R. Indhu, and T. Yawanikha,
improve productivity and improve the standard of living. “Identification of Maize Plant Leaf Disease Detection using
ConvNet Model,” 8th International Conference on Smart
Structures and Systems, ICSSS 2022, pp. 29–31, 2022, doi:
10.1109/ICSSS54381.2022.9782225.
[16] M. A. Moid and M. A. Chaurasia, “Transfer Learning-based Plant
Disease Detection and Diagnosis System using Xception,”

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1413

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Smart Electronics and Communication (ICOSEC-2023)
IEEE Xplore Part Number : CFP23V90-ART ; ISBN : 979-8-3503-0088-8

Proceedings of the 5th International Conference on I-SMAC (IoT

in Social, Mobile, Analytics and Cloud), I-SMAC 2021, pp. 451–
455, 2021, doi: 10.1109/I-SMAC52330.2021.9640694.
[17] C. Hu, X. Liu, Z. Pan, and P. Li, “Automatic detection of single
ripe tomato on plant combining faster R-CNN and intuitionistic
fuzzy set,” IEEE Access, vol. 7, pp. 154683–154696, 2019, doi:
10.1109/ACCESS.2019.2949343.
[18] A. Pandey and K. Jain, “A robust deep attention dense
convolutional neural network for plant leaf disease identification
and classification from smart phone captured real world images,”
Ecological Informatics, vol. 70, no. June, p. 101725, 2022, doi:
10.1016/j.ecoinf.2022.101725.
[19] I. Yahyaoui, M. Chaabene, and F. Tadeo, “A fuzzy based energy
management for a photovoltaic pumping plant for tomatoes
irrigation,” IREC 2014 - 5th International Renewable Energy
Congress, 2014, doi: 10.1109/IREC.2014.6827015.
[20] S. J. Wei, D. F. Al Riza, and H. Nugroho, “Comparative study on
the performance of deep learning implementation in the edge
computing: Case study on the plant leaf disease identification,”
Journal of Agriculture and Food Research, vol. 10, no. September,
p. 100389, 2022, doi: 10.1016/j.jafr.2022.100389.
[21] C. Jackulin and S. Murugavalli, “A comprehensive review on
detection of plant disease using machine learning and deep learning
approaches,” Measurement: Sensors, vol. 24, no. August, p.
100441, 2022, doi: 10.1016/j.measen.2022.100441.
[22] P. S. Thakur, P. Khanna, T. Sheorey, and A. Ojha, “Trends in
vision-based machine learning techniques for plant disease
identification: A systematic review,” Expert Systems with
Applications, vol. 208, no. April, p. 118117, 2022, doi:
10.1016/j.eswa.2022.118117.
[23] V. Janani and N. M. S. Mangai, “Analysis and Classification of
Rhizome rot Disease for Turmeric Plant using Artificial
Intelligence,” International Conference on Sustainable Computing
and Data Communication Systems, ICSCDS 2022 - Proceedings,
pp. 411–414, 2022, doi: 10.1109/ICSCDS53736.2022.9760884.
[24] R. Sujatha, J. M. Chatterjee, N. Z. Jhanjhi, and S. N. Brohi,
“Performance of deep learning vs machine learning in plant leaf
disease detection,” Microprocessors and Microsystems, vol. 80,
no. December 2020, p. 103615, 2021, doi:
10.1016/j.micpro.2020.103615.
[25] S. Muthuselvan, K. Somasundaram and S. Rajaprakash, "Analysis
of software requirement analysis in software development process
using intelligent agent with intuitionistic fuzzy analytical
hierarchical process", International Journal of Scientific and
Technology Research, Volume 9, Issue 2, Pages 1844 -
1852February 2020
[26] N. Nisar, A. Awasthi, M. Chhabra, and A. I. Abidi, “Image based
recognition of plant leaf diseases: A review,” Proceedings of the
4th International Conference on IoT in Social, Mobile, Analytics
and Cloud, ISMAC 2020, pp. 373–378, 2020, doi: 10.1109/I-
SMAC49090.2020.9243434.

979-8-3503-0088-8/23/$31.00 ©2023 IEEE 1414

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on December 18,2023 at 09:56:50 UTC from IEEE Xplore. Restrictions apply.