Deep Learning Based Mobile Application for Plant Disease Detection

Just like with COVID-19, testing is important in the detection of plant diseases as well. Since the past days and in the present too, farmers usually detect crop diseases with their naked eye which makes them take tough decisions on which fertilizers to use. It requires detaile

Project Title

Deep Learning Based Mobile Application for Plant Disease Detection

Project Area of Specialization

Computer Science

Project Summary

Just like with COVID-19, testing is important in the detection of plant diseases as well. Since the past days and in the present too, farmers usually detect crop diseases with their naked eye which makes them take tough decisions on which fertilizers to use. It requires detailed knowledge of the types of diseases and a lot of experience needed to make sure the actual disease detection.

So, How to prevent this from happening?

To prevent this situation we need better and perfect guidance on which fertilizers to use, to make the correct identification of diseases, and the ability to distinguish between two or more similar types of diseases in visuals.
This is where Neural Networks and Deep learning come in handy.

How does it look like The Deep Neural Network?

Simple Neural Nets are good at learning the weights with one hidden layer which is in between the input and output layer. But, it’s not good at complex feature learning so rather than a Simple Neural Network we use Deep Neural Networks.

Deep neural networks have recently been successfully applied in many diverse domains as examples of end to end learning. Neural networks provide a mapping between an input such as an image of a diseased plant to output such as a crop disease pair. 

The nodes in a neural network are mathematical functions that take numerical inputs from the incoming edges and provide a numerical output as an outgoing edge. Deep neural networks are simply mapping the input layer to the output layer over a series of stacked layers of nodes. The challenge is to create a deep network in such a way that both the structure of the network as well as the functions (nodes) and edge weights correctly map the input to the output. Deep neural networks are trained by tuning the network parameters in such a way that the mapping improves during the training process. This process is computationally challenging and has in recent times been improved dramatically by a number of both conceptual and engineering breakthroughs.

Why we need image classifiers?

In order to develop accurate image classifiers for the purposes of plant disease diagnosis, we needed a large, verified dataset of images of diseased and healthy plants. A dataset is to be created and thousands of images are trained which result makes it easier in detecting plant diseases.

What makes it a game-changer?

It has not only its ability to diagnose plant pests and diseases but the fact that it works offline too.

Project Objectives

Our aim is to enable farmers to work more productively and profitably, to secure their livelihood, and to keep their risk as low as possible, to provide ease to the farmer by reducing his laboring power. Our primary focus is on the new generation farmer. Farmers interested in learning modern tools and want to discover, in disease identification, man has to be much more experienced. But our application has kicked off this problem.

Project Implementation Method

1. Gathering Data (Images)

We will gather the data sets as many as possible with Images affected by diseases and also which are healthy. we will require bulk data.

2. Building CNN.

Build CNN using some popularly used open-source Libraries for the development of AI, Machine Learning, and also Deep Learning.

3. Training

It is good to train models as it requires massive computation power our normal machines laptops and the computer won’t sustain so, we need to have a good GPU config system to train in your local machine

4. Fitting into Model

• Training process
• Epochs and batch size

5. Accuracy and Loss

• Tensorflow graph
• Accuracy plot
• Smoothing

6. Final Prediction

• Disease detection
• Remedies

Benefits of the Project

Our solution of plant disease detection is based on CNN, which is now the most robust technique for image classification.

The main advantages include:

Heal Your Crop: Detect pests and diseases on crops and get recommended treatments

Disease Alerts: Be the first to know when a disease is about to strike in your district

Get Your Questions Answered By Experts: Whenever you have questions concerning agriculture, reach out to the PDD App! Benefit from agricultural experts’ know-how or help fellow farmers with your experience.

Diagnose and Treat Crop Issues: Whether your crops are suffering from a pest, disease, or nutrient deficiency, just by clicking a picture of it with the PDD app you will get a diagnosis and suggested treatments within seconds.

However, our further research is related to the precise recognition of particular diseases. After extensive training on diverse datasets, our machine learning model will be capable of distinguishing a large number of different diseases.

Technical Details of Final Deliverable

PDD turns your Android phone into a mobile crop doctor with which you can accurately detect pests and diseases on crops within seconds. PDD serves as a complete solution for crop production and management.

It is good to train models as it requires massive computation power our normal machines laptops and the computer won’t sustain so, we need to have a good GPU config system to train in your local machine

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Agriculture

Core Technology

Artificial Intelligence(AI)

Other Technologies

Cloud Infrastructure, Others, Big Data

Sustainable Development Goals

Good Health and Well-Being for People, Climate Action

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com