Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5
CNN based Leaf Disease Identification and
Remedy Recommendation System
Suma V
R Amog Shetty, Rishab F Tated, Sunku Rohan, Triveni S Pujar
Department of Information Science and Engineering, Dayananda Sagar College of
Engineering, Kumaraswamy Layout, Bangalore-560078, India
Abstract—Agriculture is one field which has a high impact
on life and economic status of human beings. Improper
management leads to loss in agricultural products. Farmers
lack the knowledge of disease and hence they produce less
production. Kisan call centers are available but do not offer
service 24*7 and sometimes communication too fail. Farmers
are unable to explain disease properly on call need to analysis
the image of affected area of disease. Though, images and
videos of crops provide better view and agro scientists can
provide a better solution to resolve the issues related to
healthy crop yet it not been informed to farmers.
It is required to note that if the productivity of the crop is
not healthy, it has high risk of providing good and healthy
nutrition. Due to the improvement and development in
technology where devices are smart enough to recognize and
detect plant diseases. Recognizing illness can prompt faster
treatment in order to lessen the negative impacts on harvest.
This paper therefore focus upon plant disease detection
using image processing approach This work utilizes an open
dataset of 5000 pictures of unhealthy and solid plants, where
convolution system and semi supervised techniques are used
to characterize crop species and detect the sickness status of 4
distinct classes.
Key words: CNN, leaf disease, Classification, deep
learning, remedies.
I. I NTRODUCTION
Health of human beings depends on the type of food they
consume. If the food is unhealthy, it certainly leads to
poor nutrition and emergence of several types of health
issues. Thus, having good crop productivity depends on
healthy plants. [2]Any type of disease in plants yields
unhealthy crops. Hence, detection of plant disease forms
basic and most important step in yielding good crops.
However, manual mode of such detection is not accurate
and is time consuming. Hence, it is now possible to
conduct such a detection using advanced technological
support. Deep Learning technology can accurately detect
presence of pests and disease in the farms. Upon this
Machine learning algorithm can even predict accurately
the chance of any disease and pest attacks in future.[1] A
normal human monitoring cannot accurately predict the
978-1-7281-0167-5/19/$31.00 ©2019 IEEE
amount and intense of pests and disease attacked in farm for
spraying correct and enough fertilizers/pesticides to eliminate
the host. Therefore, artificial Perceptron indicates an accurate
value and provide corrective measure of appropriate amount of
pesticides/ fertilizers to be sprayed at specified target areas.
The aim of this paper is to help the farmers to protect his
farm from any kind of pests and disease attacks and eliminate
them without disturbing the decorum of the soil and untouched
parts of other plants.[4] Mostly in India, farmers use manual
monitoring and some apps which have huge database
limitations and are only bound to detection part. Since,
Prevention is better than cure, this paper aims at detection of
attack of pests/diseases in future thereby making farmer to
prevent such attacks.
Technology has laid its influence in developing farms and
agro-based industries. Today, it is possible to grow crops in
deserts by using technology. Technology has dived into
depths in agriculture sector. Automation technology is the
present most demanded tool in agriculture. Many companies
have come up with latest solutions in Machine Learning,
Artificial Intelligence transforming agriculture into a Digital
Agriculture etc. Many tests have proved that deploying
technology in farms, will increase crop yield and farmer’s
revenue thereby. This paper discusses and tests Deep
Learning technology implementation in agriculture.
Diagnosis is always a concern for farmers in India. At the
same time due to fear of attack of pests/diseases, farmer
uniformly sprays pesticides/fertilizers in whole farm which
may lead to damage of soil as well as plant. The aim of this
paper is to make the farmer to spray a limited and enough
pesticide/fertilizer at a specified target area where either
pest/disease is present or maybe an occurrence of attack in
future. This helps the farmers mainly to prevent any such
attacks on his farm as well as eliminate them if present any
by spraying in limited amount and not polluting soil and
other parts of plants. Major advantage of this is to increase
395
 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5
farmer’s annual monetary revenue and minimizing crop
loss caused by pests/disease attacks. The work thus
carried out in this part of research is to put forth a
solution which consists of 3 major steps: Data
acquisition, Pre-processing and Classification for plant
disease detection. It further helps in resolving those
issues and produce healthy crop growth.
II. LITERATURE SURVEY
Authors of [11] give a brief knowledge of Support Vector
Machine. They focus on proposing an automated system for
diagnosis of common paddy diseases by making use of k-
means clustering and feature extraction. While the authors
of [12] focuses on plant diseases that are affected by
climatic causes in Thailand. Accordingly, they state that
system should have an application which can operate for
specific disease diagnosis using rule based model of data
mining technique.
Authors of [13] focuses on proposing a model that provides
automatic method find out leaf diseases by inspecting if an
image which is subjected to examination in the system has
been affected by any disease or not. They are generated by
using different cluster sizes using image segmentation and
thereby obtaining an optimized results. However, the work
carried out by the authors [14] emphasize to determine the
nitrogen deficiency and further anticipate the right quantity
of fertilizers required for the type of area using feature
extraction, text extraction.
However, authors of [14] feel that digital image recognition
of plant diseases is one of the thrust areas and hence came
out with a model which comprises of back propagation
networks and probabilistic neural networks. It is further
depending on color features, shape features and text
features extracted from disease image.
Also, the work of authors [15] focuses on occurrence of risk
factor in apple. Beta regression model is used to predict the
subsystem and the severity of the apple disease so that it
helps the farmers to take the decision of pesticide spray and
reduce the diseases.
Thus, huge amount of research is always going on in the
domain of agriculture in order to yield better and
satisfactory results.
978-1-7281-0167-5/19/$31.00 ©2019 IEEE
III. SYSTEM DESIGN AND
ARCHITECTURE
MATERIALS AND METHODS
In order to develop a model for plant disease recognition, the
approach used is deep CNN.
1. DATASET
For the purpose of image-based identification which includes,
training phase to evaluation phase where the performance of
classification algorithms are evaluated, it is necessary to have huge
data sets. Hence, the source of data is collected from PlantVillage
website. The images thus collated are labeled with four different
categories-bacterial spot , yellow leaf curl virus , late blight and
healthy(in order to differentiate healthy leaves from affected ones).
Subsequently, there is a need to enhance the dataset by adding the
images that are augmented. This paper further train the network to
learn features that differentiates one class from other.
Correspondingly, a database comprising of more than 5000 images
are used to train and around 1000 images are further used to
validate the same.
2. PROCESS AND LABEL OF IMAGES
Several samples of images are collected from plant village which
are spread across in several formats having varying levels of
resolutions and hence the variations in quality. Thus, to acquire a
reasonable feature extraction, the final images are used as input
data for classifier which are then pre-processed to achieve
consistency.
It is further ensured that at the time of data collection, those of the
images whose resolution is smaller and which has a dimension less
than 500px is not taken into consideration as valid images for the
dataset. As such, images having higher resolution form the
potential candidates for this investigation purpose. Consequently,
images are ascertained to contain all the required information for
feature learning. Accordingly, images used for the dataset
were image resized to 50 X 50. This ensures that there is a
reduction of the time required for training and automatically
computes it using written script in Python, using the OpenCV
framework. Pre-processing images involves outsourcing
background noise, intensity normalization of individual
image particles, removing reflections and masking portions of
images.
This work is broadly classified into 2 algorithms:
Artificial Neural Networks algorithm:
ANN is used to detect the plant swelling (moisture content),
disease and pest along with soil analysis. The dataset of the plant
leaf, various diseases, pests and soil images are trained in python
tool and classified into various clusters which classifies various
396
 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5
labels. Convolution Neural Networks is designed for
accurate analysis. Unsupervised Learning classification is
used since the input image is unknown and new to the
algorithm. Most of the real time applications need
unsupervised learning data since the input is always
unknown to the algorithm.
Machine Learning algorithm:
Based on the feature extraction parameters, the algorithm
predicts whether the crop is going to get any pest and disease
attacks in future. Machine learning algorithm uses
CART(classification and regression tree) to predict the
condition of the plant in future based on the given trained
data. Starting from the basic image reading process, dataset
of nearly 900-1000 each set of healthy plant images,
different pest infected images and diseases infected images
are fed. Therefore, each set of categories will have 900-1000
images. The feature extraction is GLCM texture extraction
and edge detection for plant disease identification. These
features describe the actual condition of the plant based on
the pixel values. Figure 1 indicates the flow of activities for
the aforementioned steps.
Figure 1: Flow Chart depicting overview of the
process
The parameters of completely disease and pest infected
plants, partially infected disease and pest plants and healthy
plant images are taken as image dataset. These values are
trained and clustered based on their respective categories.
Right from healthy to partially to completely disease and pest
infected plants are classified into separate labels. The values
are trained in accordance with Unsupervised Learning
algorithm.
The input image taken called as test image, after the
extraction of features of it, each feature gets compared with
the trained data features. Therefore, a centroid is formed
where the nearest corresponding feature cluster is selected.
The label under nearest corresponding label gets displayed
978-1-7281-0167-5/19/$31.00 ©2019 IEEE
like “Disease X and Pest Y”. This phase is Disease/Pest Detecting
part. The Neural Network here is the creation of Convolution
Neural Network (CNN) used for Non-Linear Regression models.
An artificial neuron gets created having several hidden layers in it.
IV. IMPLEMENTATION
1. NEURAL NETWORK TRAINING
This paper put forth a model which is used to train CNN in order
to identify leaf disease. Tenser flow, an open source library is
used to carry out numerical computations in neural networks
along with data flow graphs. Nodes represent mathematical
operations while graph edges represent multidimensional data
arrays. Convolution neural network in machine learning is a type
of feed forward artificial neural network in which neurons are
associated in a pattern that is stimulated by the organization of
animal visual cortex. Receptive field which happens to be
restricted region of space is the location where individual cortical
neurons respond to stimuli. Different neurons in the respective
fields partially overlap so as to tile the visual field. Individual
neurons react to the stimuli within its field. Convolution
operation approximates this scenario mathematically. These are
stimulated by biological process and variations of multilayer
perceptron that are intended to use pre-processing in minimal
amounts. Consequently, applications in image and video
recognition, natural language processing and recommender
systems are large in number. Convolution networks may include
global or local pooling layers which combines output of neuron
clusters.
The first convolution layer filters the input image with 32 kernels
of size 3x3. After max pooling is applied, the out- put is given as
an input for the second convolution layer with 64 kernels of size
4x4. The last convolution layer has 128 kernels of size 1x1
followed by a fully connected layer of 512 neurons. The output
of this layer is given to softmax function which produces a
probability distribution of the four output classes. Figure 2
depicts sample images from a database of types of leaves while
Figure 2: Sample images from the database a) healthy leaf
image taken under a constant background b) healthy leaf im-
age taken under uncontrolled environment [c-e] leaf images
397
 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5

from a plant affected by: c) septorial leaf blight d) V. RESULTS

frogeye leaf spot e) downy mildew

The dataset is divided into 70% for the training, 10% for
validation and 20% for testing. Different models with different
architectures and learning rate are tested. The parameters of the
network like the kernel size, filter size, learning parameter were
selected by trial and error. Table 1 depicts classification results
from different models using different architectures

Figure 2: Sample segmented images resized to

64x64 pixels
a) healthy leaf image taken under a constant background
b) healthy leaf image taken in uncontrolled environment [c-
e] leaf images from a plant affected by: c) septorial leaf
blight d) frogeye leaf spot e) downy mildew

Figure 6 indicates graph of training accuracy versus

actual accuracy

Figure 6: Training verses validation accuracy of the base

Figure 4: illustrates working of CNN layers From the result, the classification accuracy from the color
images is better than the gray scale and the segmented images.
This shows the color feature is important to extract important
features for classification. The model that provides good
classification accuracy contains three convolutional layers each
followed by max pooling layer. The graphs of the training
accuracy versus validation ac- curacy of the model is shown. It
can be seen from the graphs that the model is overfitting.
Overfitting happens when the model fits too well to the training
set. It then becomes difficult for the model to generalize to new
examples that were not in the training set.

Experiments were conducted to see the effect of each technique

Figure 5: Visualization of feature maps in the first activation
layer a) sample image b) feature maps of the first activation
layer
on the performance of the model. Since the dataset is too small
when compared to the total number of trainable parameters of
the model, the first experiment carried out is to increase the
training data by rotating, flipping, rescaling of the images. The
data augmentation is performed only on the training data.

978-1-7281-0167-5/19/$31.00 ©2019 IEEE 398

 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5
Figure 7 indicates activation process of the feature maps
Figure 7 Training vs validation accuracy of the models.
Table 2 illustrates the results of testing and validation
accuracy
Table 2. Effect of dropout and regularization
From the Table 2, it is clear that the model provides the
required efficiency
VII. CONCLUSION
Convolution neural network is used to detect and classify
plant diseases. The Network is trained using the images
taken in the natural environment and achieved 99.32%
classification ability. This shows the ability of CNN to
extract important features in the natural environment which
is required for plant disease classification.
Image classification, Image Categories, Feature Extraction,
and Training Data is carried out. The whole development of
algorithm is done in Python tool. Using several toolboxes
like Statistics and Machine Learning toolbox, Neural
Network Toolbox and Image Processing Toolbox the outputs
as of now are the training data in form of image categories,
image classification using K-Means clustering and moisture
content along with predicting of withstanding. The algorithm
978-1-7281-0167-5/19/$31.00 ©2019 IEEE
is implemented with training data and classification of given
image dataset. The test input image is compared with the trained
data for detection and prediction analysis. From the results, it is
clear that model provides reliable results
VIII. REFERENCES
1. Machine Learning: What it is and why it matters, 09
2016, [online
2. R.E. Schapire, "The boosting approach to machine
learning: An overview" in Nonlinear estimation and
classification, New York:Springer, pp. 149-171, 2003.
3. A. Ben-Hur, D. Horn, H. T. Siegelmann, V. Vapnik,
"Support vector clustering", Journal of machine
learning research, vol. 2, pp. 125-137, Dec 2001.
4. J. R. Otukei, T. Blaschke, "Land cover change
assessment using decision trees support vector
machines and maximum likelihood classification
algorithms", International Journal of Applied Earth
Observation and Geoinformation, vol. 12, pp. S27-
S31, 2010.
5. Mukhopadhyay S.C. (2012) Smart Sensing
Technology for Agriculture and Environmental
Monitoring. Vol. 146, Springer Berlin Heidelberg.
6. German, L., Ramisch, J.J. & Verma R. (2010) Beyond
the Biophysical, Knowledge, Culture, and Power in
Agriculture and Natural Resource Management,
Springer Publ.
7. Jun Wu, Anastasiya Olesnikova, Chi-Hwa Song, Won
Don Lee (2009). The Development and Application of
Decision Tree for Agriculture Data. IITSI :16-20.
8. Leemans, V., Destain, M.F.,2004.A real-time grading
method of apples based on features extracted from
defects. J. Food Eng. 61, 83-89.
9. Quinlan, J.R.(1985b). Decision trees and multi-valued
attributes. In J.E. Hayes & D. Michie (Eds.), Machine
intelligence 11. Oxford University Press (in press).
10. Zelu Zia (2009). An Expert System Based on Spatial
Data Mining used Decision Tree for Agriculture Land
Grading. Second International Conference on
Intelligent Computation Technology and Automation.
Oct10-11, China.
399