MACHINE LEARNING ENABLED

PLANT DISEASE
DETECTION
-A
REVIEW

Dr. Rakesh Sharma Amit Banga

Associate Professor, Department of ECE Research Scholar, Department of ECE
BMU, Rohtak BMU, Rohtak
INDEX

INTRIDUCTION
MOTIVATION
REVIEW OF LITERATURE
DATASETS
CONCLUSION
INTRODUCTION
 Recent advancements in machine learning (ML) and deep learning (DL) have
revolutionized plant disease detection, offering automated solutions to
enhance agricultural productivity and food security.
 This paper synthesizes current literature on ML and DL techniques applied to
image-based plant disease classification. Key studies reviewed highlight the
effectiveness of DL models, particularly convolutional neural networks (CNNs),
in accurately identifying and classifying plant diseases from leaf images.
 The integration of IoT devices for real-time data acquisition and datasets such
as Plant Village has significantly improved disease detection capabilities.
Comparative analyses of ML and DL architectures underscore the superior
performance of DL in handling complex image features and achieving high
accuracy rates.
 Future research directions include optimizing model interpretability, scalability,
and addressing data quality challenges to further enhance disease
management systems in agriculture.
 Agriculture stands as a cornerstone of
human civilization, providing sustenance,
economic stability, and cultural identity
across the globe. Central to its success is
the ability to ensure crop health and
productivity. However, the threat posed  As agriculture faces the dual challenge of
by plant diseases looms large, feeding an ever-growing population while
significantly impacting crop yields and mitigating environmental impacts, the need for
food security worldwide. Plant diseases, efficient disease management strategies
caused by various pathogens including becomes increasingly urgent. Automated
fungi, bacteria, viruses, and pests, result disease detection using machine learning and
in substantial economic losses each year. deep learning techniques presents a promising
These losses not only affect farmers' avenue to bolster crop health monitoring and
livelihoods but also contribute to management practices. By enabling early and
fluctuations in food availability and prices, accurate detection of plant diseases, these
ultimately impacting global food security. technologies offer potential solutions to
mitigate losses, optimize resource allocation,
and sustainably enhance agricultural
productivity.
MOTIVATION
 The imperative for automated disease detection in agriculture using
machine learning and deep learning methodologies arises from the
increasing complexity and scale of modern agricultural systems. These
technologies enable rapid and accurate analysis of large-scale agricultural
data, detecting subtle patterns indicative of disease onset before
symptoms become visually apparent. By facilitating early detection,
farmers can implement timely interventions, optimizing resource usage
and minimizing environmental impacts.

 Automated disease detection systems also provide real-time

monitoring across diverse geographical regions, supporting proactive
disease management strategies and enhancing overall agricultural
resilience. This integration not only revolutionizes disease detection
but also bolsters global food security by safeguarding crop health
more effectively than ever before.
REVIEW OF LITERATURE

AUTHOR YEAR CONTRIBUTIONS

Comprehensive review of ML and DL

techniques for image-based plant disease
Sajitha et al. 2024
classification tailored for industrial farming
systems.
REVIEW OF LITERATURE
AUTHOR YEAR CONTRIBUTIONS

Application of DL techniques for plant disease

Adekunle et al. 2024 detection, showing high accuracy of CNN-
based models.
REVIEW OF LITERATURE

AUTHOR YEAR CONTRIBUTIONS

Comprehensive review of early plant disease

Prasad and
2024 detection using IoT, ML, and DL, emphasizing
Thyagaraju
integration for enhanced accuracy.
DATASETS
Methods/ Dataset
Study Focus Key Findings
Architectures Used
Demilie Comparative study of Various ML, DL methods Various Provides insights into
(2024) disease detection datasets strengths and limitations of
techniques different detection
techniques for specific
applications.
Reis and Hybrid DL model for Depthwise separable Potato leaf Hybrid approach improves
Turk (2024) potato leaf disease convolutions, disease accuracy and efficiency in
detection transformers dataset disease detection
compared to traditional DL
models.
Kulkarni ML techniques for SVM, RF Rice leaf Evaluates SVM, RF for
and Shastri rice leaf disease disease classifying rice leaf
(2024) detection dataset diseases, emphasizing
algorithm performance in
agriculture.
Bouacida et DL for cross-crop CNNs Cross-crop Proposes a generalized
al. (2024) disease detection disease model for detecting plant
dataset diseases across different
crop species.
Dubey and Adaptive feature DL with adaptive feature Paddy leaf Enhances classification
Choubey selection with DL for selection disease accuracy by selecting
(2024) paddy leaf diseases dataset relevant features from leaf
images.
DATASETS
Datase Crop Disease Numbe Source Preprocessi
t Name Type Types r of s ng Applied
Images

Early
PlantVi PlantVill
Blight,
llage age Normalizatio
Tomato Late 10,000
Datase reposito n, cropping
Blight,
t ry
Septoria

Wheat Stripe Researc

Rust Rust, h Resizing,
Wheat 5,000
Datase Stem instituti rotation
t Rust on

Agricult
Grapev Downy Color
ural
ine Grapevin Mildew, adjustment,
7,500 researc
Datase e Powdery augmentatio
h
t Mildew n
project
CONCLUSION
 The “application of machine learning (ML) and deep learning (DL) in plant disease
detection represents a transformative advancement in agricultural technology.
 This literature review has synthesized recent research that underscores the efficacy of ML
and DL techniques in automating and improving the accuracy of disease identification
across various crop species. Studies reviewed here demonstrate that DL models,
particularly convolutional neural networks (CNNs), are highly effective in extracting
intricate features from plant images, thereby enabling precise and early detection of
diseases.
 The integration of IoT devices for real-time data acquisition further enhances the
scalability and efficiency of disease management systems in agriculture. The reviewed
literature also highlights the critical role of datasets in training robust ML and DL models.
 Diverse datasets such as specific crop disease collections and broader datasets like Plant
Village have facilitated the development and benchmarking of these models, leading to
significant improvements in disease classification accuracy and scalability. Studies
comparing different ML and DL architectures emphasize the importance of model selection
and parameter optimization in achieving optimal performance.
 The reviewed literature underscores the transformative potential of ML and DL in
revolutionizing plant disease management. By leveraging these technologies, agricultural
stakeholders can mitigate crop losses, optimize resource use, and contribute to global
food security in a sustainable manner.”
REFERENCES
 Sajitha, P., Andrushia, A. D., Anand, N., & Naser, M. Z. (2024). A Review on Machine Learning and Deep
Learning Image-based Plant Disease Classification for Industrial Farming Systems. Journal of Industrial
Information Integration, 100572.
 Khalid, M. M., & Karan, O. (2024). Deep learning for plant disease detection. International Journal of
Mathematics, Statistics, and Computer Science, 2, 75-84.
 Demilie, W. B. (2024). Plant disease detection and classification techniques: a comparative study of the
performances. Journal of Big Data, 11(1), 5.
 Adekunle, T. S., Lawrence, M. O., Alabi, O. O., Afolorunso, A. A., Ebong, G. N., & Oladipupo, M. A. (2024).
Deep learning technique for plant disease detection. Computer Science and Information
Technologies, 5(1), 55-62.
 Reis, H. C., & Turk, V. (2024). Potato leaf disease detection with a novel deep learning model based on
depthwise separable convolution and transformer networks. Engineering Applications of Artificial
Intelligence, 133, 108307.
 Prasad, S. R., & Thyagaraju, G. S. (2024). Leaf analysis based early plant disease detection using Internet
of Things, Machine Learning and Deep Learning: A comprehensive review. Journal of Integrated Science
and Technology, 12(2), 734-734.
 Kulkarni, P., & Shastri, S. (2024). Rice Leaf Diseases Detection Using Machine Learning. Journal of
Scientific Research and Technology, 17-22.
 Bouacida, I., Farou, B., Djakhdjakha, L., Seridi, H., & Kurulay, M. (2024). Innovative deep learning
approach for cross-crop plant disease detection: A generalized method for identifying unhealthy
leaves. Information Processing in Agriculture.
 Chin, P. W., Ng, K. W., & Palanichamy, N. (2024). Plant Disease Detection and Classification Using Deep
Learning Methods: A Comparison Study. Journal of Informatics and Web Engineering, 3(1), 155-168.
 Dubey, R. K., & Choubey, D. K. (2024). An efficient adaptive feature selection with deep learning model-