Ever since the inception of modern technologies such as IoT and AI/ML, the technologies have been proved to be very helpful in development of industries. One such use case has been carefully studied, and Smart Fruit Segregation Machine solves one such problem. Sorting of fruits is a big task within the agricultural sector, and this task has been realized in an innovate way with the help of IoT, ML and Automation. In this project the software for the Smart Fruit Segregation Machine is presented.
The features of the machine are presented as below-
- Provides a fully integrated machine capable of segregating fruits on the basis of their marketed grades using Computer Vision, Spectroscopy and Machine Learning.
- Provides a Graphic User Interface making it easy for the user to use the machine.
- Provides the user ability to retrain the model with images of fruits fed within the machine.
- Provides a cost-friendly mechanical/mechatronic architecture flexible and customizable for different fruits.
- Achieves a 80% accuracy in segregating Red Delicious Apples from Himachal, into it's three grades.
Smart Fruit Segregation Machine uses these particular tech stacks to work properly-
- Python - Programming Language widely accepted both in Automation and Machine Learning.
- Jupyter Notebook - Used Google Colab with this, online Python Notebook for ML model prototyping.
- Scikit-Learn - Used as the primary Library for Prototyping ML Models.
- H20.ai Driverless AI - Used for AutoML.
- Raspberry Pi - Used as the CPU of the machine with Serial Communication for controlling the Micro-Controller
- Arduino - Used as the Primary Micro-Controller of the system for handling actuations and sensing.
- PyQT5 - Used to create a user-friendly GUI for controlling the Machine.
- Microsoft Azure CustomVision.Ai - Used as the primary cloud service for Image Processing in the system