Fruit Bugs

This final year project which is carried out through two semesters and needs a lot of research and data set. Before that, research on different application has been done to predict the sweetness, ripeness level of the watermelon and also identify the affected ones by using machine learning technique

Project Title

Fruit Bugs

Project Area of Specialization

Artificial Intelligence

Project Summary

This final year project which is carried out through two semesters and needs a lot of research and data set. Before that, research on different application has been done to predict the sweetness, ripeness level of the watermelon and also identify the affected ones by using machine learning techniques through a mobile application named as fruit bugs, used for android and ios users. The data is collected through Google forms in the form of images i.e. outer, inner, top and bottom also including color(inner, outer) etc and is used to trained the model at the server side, while to test the watermelon client has to send the requested image to the server where the server will compare the requested images with the trained model by using 

python notebook. And, Server will return the data to the client in the json format and display the predicted result on the client’s mobile. People mostly gets wrong prediction while buying watermelon by just tapping on the bottom to check its ripeness level and also check the affected ones by looking for a yellow spot on its surface which could result in a wrong prediction, and an unsatisfied customer service.

The application will help every individual for buying a good and fresh watermelon. China produces a wide range of watermelon and use it in a large quantity also including turkey where they are mostly used.

Project Objectives

1. To predict the sweetness level of watermelon.

2. To predict the ripeness level of watermelon.

3. To identify the affected ones.

4. To develop a mobile application for android and IOS.

5. To utilize the power of machine learning solve real world problems.

Project Implementation Method

1. Buying watermelons to create dataset.
2. Target of 2000 dataset of watermelons are collecting.

3. Google forms for labeling the datasets. And, store dataset in google drive.

4. Using Google Drive Api to download the Dataset and labelized them.

5. Resizing the images of dataset according to our Deep learning model needs.

6. Building a Deep learning model (convolution neural network).

7. Python language and editor for organizing datasets.

8. Python notebook for debugging.

9. Wampp used for requesting the server, and the result will be return on the user's mobile. Also, Flask Api and gunicorn server to deploy the application to the linux based servers.

10. Flutter SDK is used for designing and developing application for both android an well as IOS. Android studio as tool for developing application.

11. Picture is taken from the mobile which is processed by the server using convolution neural network. return the expected result in json format.

Benefits of the Project

1. Every individual who wants to eat the watermelon.

2. Dealer can check the watermelon to increase its market value.

3. People who use watermelon as prevention from cardiovascular disease and prostate cancer.

4. The existing system is unable to predict the sweetness level of watermelon.

Technical Details of Final Deliverable

• Android and iOS App

• CNN (Convolutional neural network) Model

  • Keras API

  • AlexNet Architecture

  • Trained Model for Watermelon

• Connectivity Between app and CNN Model through server

• Web Application which is for debuging

• Deployed Android and iOS App. Also, Web Application.

Final Deliverable of the Project

Software System

Type of Industry

Agriculture

Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Decent Work and Economic Growth, Industry, Innovation and Infrastructure

Required Resources

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