Automated_banana_farming

tensorflow based model to classify banana as rooten,rippen,not rippen
the flow diagram for our whole code
showing how rover automatically navigates with help of deep reinforcement learning\
show the path taken by rover and the colered coordinates being saved
the final map being produced

What it does

It allows automacy in the farming process and reduce the labour cost,time and increase the production within the same land.

How we built it

The project tackled the challenges of feeding a growing population sustainably using technologies like Pygame, NumPy, and CNN. Deep reinforcement learning and the YOLOv3 algorithm were implemented, while leveraging Postman API.Orcus,Spheron,Quine, FastAPI, and React.js.

Challenges we ran into

During the development of the project, several challenges were encountered. The initial hurdle was learning and becoming familiar with Pygame, a Python library for game development. This involved overcoming the learning curve and understanding the framework, which required time and effort.

Another significant challenge was the limited time available to develop a refined idea within 24 hours. Striking a balance between project complexity and time constraints required careful planning and prioritization.

Implementing deep reinforcement learning was a complex task, as it involved understanding the principles and algorithms of reinforcement learning and effectively integrating them into the project.

Achieving high accuracy in banana classification using a Convolutional Neural Network (CNN) proved demanding. Training the CNN to reach a 98 percent accuracy rate necessitated extensive training, data preprocessing, and fine-tuning of the model.

These challenges highlight the need for perseverance, careful planning, and continuous learning throughout the development process.