Research Proposal

On
Plant Diseases Detection using Machine Learning

Abstract
It is recommended to predict the plant diseases at their early stage in the field of
agriculture is essential to cater the food to the overall population.
The approaches of machine learning and image processing with an accurate
algorithm is identified to detect the leaf diseases in the plants.
Many parts of a plant can be infected by a virus, fungal, bacteria, and other infectious
organisms but here we mainly considered the detection of leaf disease of a plant as a
research topic.
Machine learning (ML) and deep learning (DL) models are used to classify different
types of leaf diseases.
mostly, the samples of plant leaves are resized to 256 × 256 pixels and then
Histogram Equalization is used to improve the quality of plant samples.
The K-means clustering is introduced for partitioning of data space into Voronoi cells.
The boundary of leaf samples is extracted using contour tracing.
Finally, the extracted features are classified using machine learning approaches such
as Support Vector Machine (SVM), Convolution Neural Network (CNN) and K-
Nearest Neighbor (K-NN).
The accuracy of the was tested using SVM (88%), K-NN (97%) and CNN (99.6%) on
tomato disordered samples.

Introduction
Leaf disease is a kind of phenomenon to the natural growth of a plant which is not
only generated hurdles in agribusiness but is also responsible for hampering the
agricultural production of a country.
Several types of bacteria, fungi, viruses , and other natural infectious organisms are
the main causes of leaf disease in their life cycles.
There are many ways to detect and classify different kinds of Plant diseases.
i. The first option is direct observation via naked eyes which is not a prominent
process.
ii.Secondly, one can investigate the leaf stresses either by manual process or applying
any machine learning (ML) algorithms.

Literautre Review
[1] (Vijai Singh, A.K. Misra)(2017)
This paper presents an algorithm for image segmentation technique which is used for
automatic detection and classification of plant leaf diseases. It also covers survey on
different diseases classification techniques that can be used for plant leaf disease
detection. Image segmentation, which is an important aspect for disease detection in
plant leaf disease, is done by using genetic algorithm.

[2]Bhavya Gali
Computer Science and Engineering Department
B V Raju Institute of Technology
Narsapur,
Image Based Plant Disease Detection Using Computer Vision
The purpose of the paper is to inform
farmers about innovative technology that can help to prevent plant diseases. To detect
plant conditions, an accurate algorithm for machine learning and imaging sensors is
identified. Optical techniques, such as RGB imaging, multi- and hyperspectral
sensors, thermography, and chlorophyll fluorescence, have proven effective in
computerized, unbiased, repeatable detection systems for locating and measuring
plant diseases during the initial phases of outbreaks.

[3] S. S. Sannakki and V. S. Rajpurohit, proposed a

“Classification of Pomegranate Diseases Based on Back Propagation Neural
Network” which mainly works on the method of Segment the defected area and color
and texture are used as the features. Here they used neural network classifier for the
classification. The main advantage is it Converts to L*a*b to extract chromaticity
layers of the image and Categorisation is found to be 97.30% accurate. The main
disadvantage is that it is used only for the limited crops. ML and DL methods for
detecting plant diseases, including image processing, feature extraction, CNNs, and
DBNs, and sheds light on the benefits and drawbacks, such as data availability,
imaging quality, and differentiation between healthy and diseased plants. The article
shows that the use of ML and DL techniques significantly increases the precision and
speed of plant disease detection.

[4] Aakanksha Rastogi, Ritika Arora and Shanu Sharma,” Leaf Disease Detection and
Grading using Computer Vision Technology &Fuzzy Logic”. K-means clustering
used to segment the defected area; GLCM is used for the extraction of texture
features, Fuzzy logic is used for disease grading. They used artificial neural network
(ANN) as a classifier which mainly helps to check the severity of the diseased leaf.

[5] P. R. Rothe and R. V. Kshirsagar introduced a” Cotton Leaf Disease Identification

using Pattern Recognition Techniques” which Uses snake segmentation, here Hu’s
moments are used as distinctive attribute. Active contour model used to limit the
vitality inside the infection spot, BPNN classifier tackles the numerous class
problems. The average classification is found to be 85.52%.

Problem Statement
Recognizing plant diseases promptly as they manifest on a plant’s leaves.
The suggested system is computer software that identifies and categorizes plant
diseases automatically. For the RGB
image input right away, the software builds a structure for color transformation.
The disease can be located in the classifier, along with an option for therapy. The
farmer receives this information via a GSM modem.

These researchers also have the challenging task to deal with different weather
conditions, the direction of capturing an image, the distance between electronic
devices and infected leaves and different occlusion-based images, and many more.
One can obtain several images for different configurations by using various electronic
devices namely mobile phones, digital cameras, etc.

Solutions:
By intergrated Machine learning algorithm with Various techniques of image filtering
are generally used such as box filter , Gaussian filter , Gradient, and Laplacian filter to
avoid noise-related problems in IP.
After filtering Images we can apply CNN for infected leafs clustring.
Objective

1) To detect unhealthy regions of plants accurately.

2) To supply pesticides as a treatment for the appropriate

diseases and pests.

3) Sorting plant leaf diseases based on textural characteristics

Methodology
A Data Preprocessing
A feature vector is produced by combining the data from DWT, GLCM, and PCA.
This feature vector is used as a parameter set by the classifier to recognise and
categorise the images

B. Model Training
The methods utilized for classifying the data include support vector machines (SVM),
k-nearest neighbor algorithms (K-NN), and convolutional neural networks (CNN)

C. Segmentation Algorithm
Segmentation promotes excellent disease detection and classification by making it
possible for the location and recognition of disease patterns with great precision.
It aids in reducing unnecessary noise and artifacts, enhancing the system’s ultimate
precision and efficacy in detecting leaf illness

D. Classification Algorithm
Computer vision applications like detecting leaf damage commonly use CNNs or
convolutional neural networks. They are adept in automatically learning nested
characteristics from images and have acquired state-of-the-art performance in an array
of image classification applications.

E. Validation
we evaluate the efficiency and generalizability of the trained model on the testing set.
Create metrics like precision, recall, accuracy, F1-score, and confusion matrix to
assess how well the algorithm distinguished between healthy and diseased leaves

Conclusion
The techniques including RGB conversion to gray, HE, K-means clustering, contour
tracing is employed in preprocessing stage. The multiple descriptors Discrete Wavelet
Transform, principal Component Analysis and GLCM are used to extract the
informative features of the leaf samples.

The machine learning approaches such as SVM, K-NN and CNN are used to
distinguish diseased or non-diseased leaf. The analysis of the was well suited for
CNN machine learning classification technique with a desired accuracy compared to
other state of the art method. In future, the model can be improved using fusion
techniques for extraction of significant features and examined for other leaf samples
of datasets.

References

[1] Vijay Singh, A.K. Mishra, A. (2016). Detection of plant leaf diseases using image segmentation and

soft computing techniques. Information Processing in Agriculture, 4(1), 41–49.

https://doi.org/10.1016/j.inpa.2016.10.005

[2] Jaina et al. (2024). Image Based Plant Disease Detection Using Computer Vision. Research Gate,
12, 1–6. https://doi.org/10.1109/icaect60202.2024.10468857

[3] Sannakki, S.S., & Rajpurohit, V.S. (2015). Classification of Pomegranate Diseases Based on Back
Propagation Neural Network.

[4] Rastogi, A., Arora, R., & Sharma, S. (2015, February). Leaf disease detection and grading using
computer vision technology & fuzzy logic. In 2015 2nd international conference on signal processing
and integrated networks (SPIN) (pp. 500-505). IEEE.
[5]Rothe, P. R., & Kshirsagar, R. V. (2015, January). Cotton leaf disease identification using pattern
recognition techniques. In 2015 International conference on pervasive computing (ICPC) (pp. 1-6).
IEEE.