Smart-Waste-Management-Dataset

This dataset is designed for developing and testing image classification and object detection models focused on waste categorization. It includes images that need to be classified into different waste types based on the bounding box annotations provided in CSV files. While the dataset does not include pre-labeled images, even in the training set, it provides bounding boxes for precise localization of waste types.

Dataset Highlights

• Classification: Images are categorized into multiple waste types.
• Bounding Box Annotations: CSV files provide coordinates for bounding boxes, enabling object detection and classification tasks.
• Unlabeled Images: Original images in the training set are unlabeled, offering opportunities for semi-supervised and unsupervised learning experiments.

Applications

This dataset supports a range of applications, including:

• Waste Management: Develop AI models to automate waste sorting and recycling.
• Object Detection: Train and benchmark models for detecting and classifying waste in images.
• Environmental Research: Use AI to address challenges in waste categorization and disposal.

Key Features

• Ideal for deep learning models such as YOLO, Faster R-CNN, and RetinaNet.
• Suitable for research in environmental AI and sustainable development.
• Provides bounding box annotations for precise object detection and classification.