Fioresnap: Android Application for Flowers Classification

There are more than 250,000 species of flowering plants. In our daily life, we can see different types of flowering plants on the roadside, on paths, and on fields. Normally, botanists can recognize flowers according to their species. Many people do not know about flowers and their types. People hav

Project Title

Fioresnap: Android Application for Flowers Classification

Project Area of Specialization

Artificial Intelligence

Project Summary

There are more than 250,000 species of flowering plants. In our daily life, we can see different types of flowering plants on the roadside, on paths, and on fields. Normally, botanists can recognize flowers according to their species. Many people do not know about flowers and their types. People have to consult the web pages and books for information about flowers and their characteristics by using different keywords. But some keywords are only for scientists and are not productive for all people. 
It is very difficult and important to recognize naturally occurring objects and their types. Flower classification is useful in various fields, For Example Botany research, Agriculture, Farming, Gardening, etc.
So, we need a system /model that can help all people in the classification of flowers and their types. We can do this by image processing technique. By image processing, we can differentiate flower color, structure, shape, etc. We are going to develop a system that will perform digital image processing.
Following are some classification techniques that can help us in image classification: Feature extraction, pattern recognition, and classification. Our system will be helpful to find that image processing techniques will use to extract different features from flowers and classify them by using supervised classification algorithms.
We are going to develop an android application that will allow the user to classify flowers quickly and simply.
 

Project Objectives

• Flower Recognition System (Based on Image Processing), that will be helping to the people in the agriculture/botany domain.
• Proposing a suitable feature extraction and classification AI model for flower specie recognition
• Develop an android application that is able to take image as input and the process the image through integrated AI module

Project Implementation Method

For classifying images, a particular type of deep neural network called a convolutional neural network has proved to be particularly powerful. However, modern convolutional neural networks have millions of parameters. Training them from scratch requires a lot of labeled training data and a lot of computing power (hundreds of GPU hours or more). We only have a few thousand labeled photos and want to spend much less time, so we need to be cleverer.
We will use a technique called transfer learning where we take a pre-trained network (trained on about a million general images), use it to extract features, and train a new layer on top for our own task of classifying images of flowers.
 

Benefits of the Project

The project will be helpful to the botanical researchers, agriculture domain, and gardening enthusiasts for simpler and fast identification of flowering plants. 

Technical Details of Final Deliverable

• The Machine Learning model
• Pickle representation of Machine Learning Model
• Application setup and Troubleshooting Document
• System requirements, and for android apps, minimum hosting requirements as well

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People, Sustainable Cities and Communities, Climate Action

Required Resources

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