Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

To Detection of Leaf and Weeds Diseases using

Image Processing Technology and a CNN Methods
V Parthasaradi¹, L Ramachandhan²,
Assistant Professor and Head, Assistant Professor,
1 2
Department of Electronics and Communication Department of Electronics and Communication
Engineering, Engineering,
E.G.S Pillay Engineering Collage, E.G.S Pillay Engineering Collage,
Nagapattinam Nagapattinam

N Suriya³
PG student,
3
Department of Electronics and Communication Engineering,
E.G.S Pillay Engineering Collage,
Nagapattinam

Abstract:- Agriculture is very important for human life.  Median Filter:

Now a days artificial foods are also prepared without a Median filter is used to remove the noise from the
natural food .so people are suffering a lot .Now a days signal that may come from the images. It is an important
farming is not easy .To make it easier some advances are filter in filter to help remove noise and detecting points on
being made in agriculture one of which is DL approaches edges.
which help greatly in improving agriculture. By using
CNN it helps to classify plants and identify their  How are the Leafs are Affected:
deficiencies.  Root worn causes leaf shriveling.
 Attack of chapatti but turns the leaf red.
Keywords: Convolutional Neural Network, Natural food,  Aphids cause leaf damage.
Farming.

I. INTRODUCTION

Agriculture is also a major cause of climate change.

Due to climate change rain does not fall at the right time
Due to which people suffer a lot and many types of diseases
develop .Improper maintenance of the trees in the forest
many beneficial trees are destroying. DL approaches helps
in classifying vegetation identifying it’s deficiency and
maintaining its properly.

 Existing System
In our proposed system, some methods are used to
diagnose diseases.

 Image Acquisition:
RGB color images are taken using a camera which are
used for leaf analysis.

 Image Segmentation:
Instead of using the whole image, it can be divided into
parts and can be used to easily detect diseases in important
areas.
Fig 1 (a) Cotton Leaf
 Image Preprocessing:
It is used to remove noise and improve image quality.

 Edge Detection:
It is used to detect points on edges in an image.

IJISRT23APR1804 www.ijisrt.com 1719

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 There are three Types of Automation Used:
 Fixed automation
 programmable automation
 flexible automation

 Types of Sensors:
Sound sensors are used to detect light, Temperature
sensor is used to detect fluctuation, Contact sensor is used to
detect any obstacles, Distance sensor used to find the
distance of the object to the robot relative.

 Types of Robots:
 Autonomous Mobile Robots
 Automated Guided Vehicles
 Articulated Robots
 Humanoids

Fig 1 (b) Brinjal Leaf

 Leaf Diseases:
Bacterial light is a diseases that can develop in large
numbers on weeds and stems.

II. METHODOLOGY

Here we use a variety of technologies to detect and

automatically correct plant diseases to improve agriculture.
Recently LSTM has been used to classify the weeds. LSTM
is divided into three sections.

CNN has three parts: Convolutional layer, max polling

layer, fully connected layer.
Fig 2 Autonomous Mobile Robot
 Pooling Layer:
Reduces the size of the film.  Three Main Parts of Robot:
 Sensor is used to gather information
 Convolutional Layer:  Effectors is the important part of the robot
It creates new images.  Control system used to detect the robot’s behavior

 Fully Connected Layers:  Experimentation:

After concatenating all the layers the weights of each In earlier days humans would go there to find infected
neuron or used for linear transformation to the input. weeds and remove it .But now infected weeds can be
 Different regions in images can be identified using automatically detected and removed immediately using
GLCM. robots .CNN is used to accurately classify the weeds. we get
 Gabor filter is used for direct analysis. the idea that we can identify and eliminate affected weeds to
increase agricultural productivity .Infected crops also affect
 Sobel edge detector it helps to make strong changes in
non infected crops and thus reduce productivity .Many
the image.
methods are used to correct these. Poisonous insects can
 This methods are reduces the error rate compared to the
attack humans while removing infected weeds and this can
previous model.
be prevented by using robotics.
Various methods have been invented to remove
III. CONCLUSION
infected weeds one of which is robotics. Robotics can be
used to spray pesticide and dig out infected plants. Affected
Through machine learning and neural network, we
can be removed with a laser weeds. There are several ways
need to increase the production capacity of crops. By using
to remove infected weeds. Laser technology detects and
insecticides to accurately identify the affected crops and
removes infected weeds. computer vision used to determine
cure their diseases. Some crop diseases are not easily
which weeds are affected and un affected. LIDAR sensor
detected and some methods should be introduced to detect
used to weeding robot.
them quickly .while using robotics to remove infected crops

IJISRT23APR1804 www.ijisrt.com 1720

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
some errors occur and new methods should be introduced to [14]. Szegedy, C.; Liu, W.; Jia, Y.; Sermanet, P.; Reed, S.;
correct the errors. we need to combine robotics with some Anguelov, D.; Erhan, D.; Vanhoucke, V.;
enlightened ways to remove infected crops even at night Rabinovich, A. Going deeper with convolutions. In
using robotics. Proceedings of the IEEE Conference on Computer
Vision and Pattern Recognition, Boston, MA, USA,
REFERENCES 7–12 June 2015; pp. 1–9.
[15]. Abade, A.; Ferreira, P.A.; de Barros Vidal, F. Plant
[1]. A. Camargo and J.S. Smith, “An image-processing diseases recognition on images using convolutional
based algorithm to automatically identify plant neural networks: A systematic review. Comput.
disease visual symptoms,” Biosystems Engineering, Electron. Agric. 2021, 185, 106125. [CrossRef]
vol.102, pp.9–21, January 2009.
[2]. J.S. Cope, D. Corney, J.Y. Clark, P. Remagnino, and
P. Wilkin, “Plant species identification using digital
morphometrics: A review,” Expert Systems with
Applications, vol.39, pp.7562–7573, June 2012.
[3]. Dr. K. Thangadurai, K. Padmavathi, “Computer
Vision image Enhancement For Plant Leaves Disease
Detection”, 2014 World Congress on Computing and
Communication Technologies.
[4]. Monica Jhuria, Ashwani Kumar, and Rushikesh
Borse, “Image Processing For Smart Farming:
Detection Of Disease And Fruit Grading”,
Proceedings of the 2013 IEEE Second International
Conference on Image Information Processing (ICIIP
2013).
[5]. K. Oki, S. Mitsuishi, T. Ito, M. Mizoguchi, “An
agricultural monitoring system based on the use of
remotely sensed imagery and field server web camera
data,” Giscience & Remote Sensing, vol. 46, pp. 305-
314, 2009.
[6]. G. F. Zhang, X. D. Liu, Y. Q. Zhu, and B. P. Zhai,
“An open WebGIS-based monitoring system on crop
pests and diseases,” Journal of Nanjing Agricultural
University, vol. 32, pp. 165-169, 2009.
[7]. Kaur, S.; Pandey, S.; Goel, S. Plants disease
identification and classification through leaf images:
A survey. Archives of Computational Methods in
Engineering 2019, 26, 507–530. Boca Raton, FL,
USA, 2018.
[8]. Gebbers, R.; Adamchuk, V.I. Precision agriculture
and food security. Science 2010, 327, 828–831.
[CrossRef] [PubMed]
[9]. Carvalho, F.P. Agriculture, pesticides, food security
and food safety. Environ. Sci. Policy 2006, 9, 685–
692. [CrossRef]
[10]. Mohanty, S.P.; Hughes, D.P.; Salathé, M. Using deep
learning for image-based plant disease detection.
Front. Plant Sci. 2016, 7, 1419. [CrossRef] [PubMed]
[11]. Miller, S.A.; Beed, F.D.; Harmon, C.L. Plant disease
diagnostic capabilities and networks. Annu. Rev.
Phytopathol. 2009, 47, 15–38. [CrossRef]
[12]. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning.
Nature 2015, 521, 436–444. [CrossRef]
[13]. Najafabadi, M.M.; Villanustre, F.; Khoshgoftaar,
T.M.; Seliya, N.; Wald, R.; Muharemagic, E. Deep
learning applications and challenges in big data
analytics. J. Big Data 2015, 2, 1–21. [CrossRef]

IJISRT23APR1804 www.ijisrt.com 1721