Scribd
Upload
38 views2 pages

Ac 1

This document proposes using deep learning and convolutional neural networks to develop an automatic plant disease detection system using leaf images. The system would first collect and preprocess leaf images, then train a CNN model to classify leaf images as either healthy or unhealthy with a specific disease. The trained model would be tested on new leaf images to evaluate its accuracy in detecting plant diseases. Key aspects of the proposed methodology include image preprocessing, training a CNN model, testing and fine-tuning the model, and evaluating the model's performance using various metrics. The system would utilize Python libraries and be trained on an NVIDIA GPU for efficient processing of leaf image data.

Uploaded by

Prudhvi Kurakula
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
38 views2 pages

Ac 1

This document proposes using deep learning and convolutional neural networks to develop an automatic plant disease detection system using leaf images. The system would first collect and preprocess leaf images, then train a CNN model to classify leaf images as either healthy or unhealthy with a specific disease. The trained model would be tested on new leaf images to evaluate its accuracy in detecting plant diseases. Key aspects of the proposed methodology include image preprocessing, training a CNN model, testing and fine-tuning the model, and evaluating the model's performance using various metrics. The system would utilize Python libraries and be trained on an NVIDIA GPU for efficient processing of leaf image data.

Uploaded by

Prudhvi Kurakula
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Plant Disease Identification using Leaf Images

1 Problem Statement
One of the important sectors of Indian Economy is Agriculture. Employment to almost 50% of
the countries workforce is provided by Indian agriculture sector. India is known to be the world's
largest producer of pulses, rice, wheat, spices and spice products. Farmer's economic growth
depends on the quality of the products that they produce, which relies on the plant's growth and
the yield they get. Therefore, in field of agriculture, detection of disease in plants plays an
instrumental role. Plants are highly prone to diseases that affect the growth of the plant which in
turn affects the ecology of the farmer. In order to detect a plant disease at very initial stage, use
of automatic disease detection technique is advantageous. The symptoms of plant diseases are
conspicuous in different parts of a plant such as leaves, etc. Manual detection of plant disease
using leaf images is a tedious job. Hence, it is required to develop computational methods which
will make the process of disease detection and classification using leaf images automatic.
2 Background Work
Despite of the challenges given in the problem statement plant disease detection is still an active
area of research. Numerous approaches have been proposed over the years. In traditional systems
approach for detection and differentiation of plant diseases can be achieved using Support Vector
Machine algorithms. This technique was implemented for sugar beet diseases and depending on
the type and stage of disease, the classification accuracy was between 65% and 90%. Another
approach based on leaf images and using ANNs as a technique for an automatic detection and
classification of plant diseases was used with K-means as a clustering procedure. ANN consisted
of 10 hidden layers. The number of outputs was 6 which was the number of classes representing
five diseases along with the case of a healthy leaf.
3 Materials and Methodology
Figure 1 demonstrates the framework which will be used in this work.
3.1 Data Collection and Dataset Preparation: Images can be downloaded from the Internet
using the keywords plant and disease names. Subsequently, all the images can be classified into
different groups.
3.2 Methods
3.2.1 Image preprocessing and Labelling: Images in the dataset may be in different
formats, quality and resolution. Hence, the images need to be preprocessed, for instance,
images with smaller resolution and dimension less than 500 px will not considered as
valid images for the dataset. The rest will be resized to 256 × 256 in order to reduce the
time for training.

3.2.2 Training: In this step, training the deep convolutional neural network for making
an image classification model will be done. CaffeNet architecture will be used and
adjusted to support our different categories (classes). Rectified Linear Units (ReLU) will
subsequently be used as substitute for saturating nonlinearities. This activation function
adaptively will learn the parameters of rectifiers and improve accuracy at negligible extra
computational cost.
3.2.3 Testing:In this phase, the test set for prediction of leaf as healthy/Unhealthy with
its disease name will be used to evaluate the performance of the classifier.
(a) Fine-Tuning: Fine-tuning helps to increase the accuracy of prediction by making
small modifications to improve or optimize the outcome. The best suited model for plant
disease detection will be achieved through the process of experimental adjustment of the
parameters.

Testing & Fine


Tuning

Feature Extraction
from a plant

Plant lmage Output


preprocessing
Input Image

Plant Image Dataset

Figure 1: Architecture of Plant Disease Detection System

(b) Evaluation Measures: Measures such as accuracy and mean recall score, mean precision
will be computed to evaluate the performance of the classifier.
4 Experimental Design
4.1 Software and Hardware Requirements : Python based Computer Vision and Deep
Learning libraries will be exploited for the development and experimentation of the project.
Tools such as Anaconda Python, and libraries such as OpenCV, Caffe integrated with CudNN
will be utilized for this process. Training will be conducted on NVIDIA GPUs

You might also like