Garbage Classification using MobileNetV2

This project demonstrates garbage classification using transfer learning with MobileNetV2. The dataset consists of images categorized into several classes: cardboard, glass, metal, paper, plastic, and trash. The goal is to build a model that can classify these images accurately.

Data Preprocessing

The dataset is loaded and preprocessed using OpenCV for image handling and TensorFlow/Keras utilities for data augmentation. Images are resized to 224x224 pixels, which is the input size required for MobileNetV2.

Augmentation

Data augmentation techniques are applied to increase the diversity of the dataset, including rotation, shifting, shearing, zooming, and flipping.

Model Training

Transfer Learning with MobileNetV2

• MobileNetV2 is used as the base model for transfer learning.
• The base model is frozen, and a custom classification head is added on top.
• The model is compiled with Adam optimizer, Sparse Categorical Crossentropy loss function, and Sparse Categorical Accuracy metric.

Training and Evaluation

• The data is split into training (80%) and validation (20%) sets.
• The model is trained with early stopping to prevent overfitting.
• Training progress is monitored with loss and accuracy metrics.

Results

• After training, the model achieves an accuracy of [insert your accuracy here] on the test set.
• Training and validation loss/accuracy curves are plotted for analysis.

Prediction

• The trained model can predict the class of a given image from the test set.
• Example predictions are demonstrated with corresponding images.

Repository Contents

• garbage_classification.ipynb: Jupyter notebook containing the entire code and explanation.
• garbage_classifier.h5: Trained model file.

Usage

To run the notebook:

1. Install required dependencies (opencv-python, numpy, matplotlib, tensorflow, scikit-learn).
2. Open and execute the notebook step-by-step in a Jupyter environment.

License

This project is licensed under the GPL License.