Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
51 views2 pages

Smart Sorting Project Documentation

This project presents a smart solution for identifying rotten fruits using deep learning and transfer learning, achieving over 93% accuracy. A MobileNetV2-based model is developed and deployed via a Flask web application for real-time predictions, reducing human involvement in quality checking. The approach is scalable and can be adapted for industry use with real-time camera systems.

Uploaded by

rinkuinjx
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
51 views2 pages

Smart Sorting Project Documentation

This project presents a smart solution for identifying rotten fruits using deep learning and transfer learning, achieving over 93% accuracy. A MobileNetV2-based model is developed and deployed via a Flask web application for real-time predictions, reducing human involvement in quality checking. The approach is scalable and can be adapted for industry use with real-time camera systems.

Uploaded by

rinkuinjx
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Smart Sorting: Transfer Learning for

Identifying Rotten Fruits and Vegetables

Figure: Web App Output - Prediction of Rotten Fruit

Abstract
In the agriculture and food industry, identifying spoiled produce is essential to prevent
contamination and reduce waste. Manual sorting of fruits is time-consuming, inconsistent,
and prone to human error. This project proposes a smart solution using deep learning and
transfer learning to automatically classify fruits as fresh or rotten. A MobileNetV2-based
model is trained and deployed using a Flask web application to provide real-time
predictions.

Objectives
- Develop a model that classifies fruits as fresh or rotten.
- Reduce human involvement in quality checking.
- Provide a real-time prediction interface using a web app.
- Use Transfer Learning for high performance with fewer resources.

Technologies Used
• Python
• TensorFlow/Keras
• MobileNetV2
• Flask
• HTML/CSS
• Google Colab
• Pillow

Dataset
The dataset used contains labeled images of various fruits categorized into two classes:
fresh and rotten. It includes common fruits like apples, bananas, and oranges. The dataset is
split into training, validation, and test sets (70-20-10).

Methodology
1. Data Preprocessing: Images were resized to 224x224 and normalized.
2. Data Augmentation: Applied flips, rotations, and zooms.
3. Model Building: MobileNetV2 with custom dense layers.
4. Model Training: Trained with validation monitoring.
5. Deployment: Flask web app with image upload and prediction.

Results
The model achieved over 93% accuracy on test data. The web app provides fast, real-time
predictions and can classify common fruits accurately.

Conclusion
This project demonstrates the effective use of transfer learning for automating fruit quality
detection. The app reduces human labor and improves consistency in fruit sorting systems.
It is scalable and can be extended to real-time camera-based systems for industry use.

You might also like