Crop health care application for disease detection using deep learning

The main aim is to develop this project to provide facility to the user. This application will allow the users to detect the disease of the crop by capturing the image of the crop. Application will take the picture and then apply CNN algorithm and will give disease status in return.Traditional appro

Project Title

Crop health care application for disease detection using deep learning

Project Area of Specialization

Artificial Intelligence

Project Summary

The main aim is to develop this project to provide facility to the user. This application will allow the users to detect the disease of the crop by capturing the image of the crop. Application will take the picture and then apply CNN algorithm and will give disease status in return.Traditional approach is that farmers are required to call a doctor who tells about the crop diseases.He comes and takes a look at the crops and then he tells about the diseases.Other approach is that farmers themselve make assumptions about the disease of crop.We with our app are aiming to minimize this and bring the way by which farmers wont rely on there assumption totally but our app will help them with the disease detection. Along with that it will be able to tell the recommended fertilizers to user. It will also give information to crop so that user can have better understanding of when to sow a crop and when not to.This will be an information  center that will help user to see the crops sowing and harvesting time.Our Application is based on very low cost because there is no special hardware required. It only needs a smart phone and Active internet connection

.

Project Objectives

To develope a system which will allow a user to detect a disease which in turn results in maximizing productivity.

To provide this facility in low cost and less time

Project Implementation Method

Well first of all we will need a dataset.We have selected Corn for that.Though we plan to add more crops in our project.For dataset we will need images and these images will be collected  by visiting the fields.We will need images ranging from 2000-3000.Then after we have calculated the images,next step is data filteration.This has to be an obvious step as there would be many images that are not upto the standard or to the requirements.After filteration we will finalize the dataset .After that ,We will pass the dataset to our algorithm and it will require several hundred epochs or thousand epochs before it gets fully trained and gives the result we require.After that we will put that model on some server.We will make an Android Application that will send the image to that server via an API and then Server will respond with the disease name.

 

Benefits of the Project

• Interactive UI design
• Disease identification at real time 
• Save user time and money
• Recommended fertilizers and insecticides
• Crop Knowledge that might help the farmer
• Ease and availability of above-mentioned features at one place
• Increase the fertility of soil

Technical Details of Final Deliverable

Object recognition is a wide spread research topic in the area of computer vision and image processing. There are many significant potential applications including automatic robotic navigation and manipulation, scene understanding, etc. Many approaches have been proposed in the past few years. Previous works mainly focused on using texture information to discriminate categories. Among them, the conventional neural network is one of the most widely used. This approach has shown to be very effective for object classification in humans and cars.

In the project plant diseases will be detect on the basis of plant’s condition. Their features will be matched with the pictures in the dataset to recognize their disease as well as their fertilizers and other information regarding the plant. Image processing will be used to get the best results instead comparing the pictures only. Our project allows user to capture image from mobile camera and process image to know about disease.

Tools

• PyCharm
• Anaconda
• Jupyter Notebook
• Adobe Photoshop
• MS Word
• MS PowerPoint
• Android Studio
• Camera

Technologies

• Python
• Java

Deliverable:

Machine learning is composed of well-defined phases. Below is the list of phases and their deliverables

Project Report

A complete project report which includes Software Requirement Specification, Software Design Specification and other major tasks performed.

CD Package

A CD package which include documentation and presentation.

Following requirements are necessary for developing this project:

Hardware Requirements

• Personal Computer (PC) or Laptop with a good GPU
• Camera
• Smart Phone
• Wifi Hotspot Device                                                           

Operating System:

• Android 5.0 or higher 
• Windows 8/8.1, 1
• Software
• Pycharm, Anaconda
• Star UML
• PowerPoint
• Word
• Android Studio

PHASE

Research

Data exploration

Modeling

Result Analysis

A/b testing

Productization

Final Deliverable of the Project

Hardware System

Core Industry

Agriculture

Other Industries

Education , IT , Food

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Responsible Consumption and Production

Required Resources

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