Recyclable and Household Waste Classification

# Recyclable and Household Waste Classification Dataset

This dataset contains a comprehensive collection of 15,000 images (each 256x256 pixels) depicting various recyclable materials, general waste, and household items across 30 distinct categories. With 500 images per category and 250 images per subcategory, this dataset provides a rich and diverse resource for research and development in the fields of waste classification and recycling. By offering a large number of high-quality images, this dataset aims to support the creation of robust and accurate waste sorting and categorization systems.

## Dataset Structure
The dataset is organized into a hierarchical folder structure to ensure easy navigation and accessibility. The main folder, named "images," contains subfolders representing specific waste categories or items. **These subfolder names serve as the labels for their respective categories,** making it convenient for researchers and developers to identify and utilize the images for their specific needs.

**You have to split up the dataset into test, train, and validation manually.** See [Recyclable and Household Waste Classification Code](https://www.kaggle.com/code/alistairking/recyclable-and-household-waste-classification-code) for an example of how to do that.

Within each category subfolder, there are two distinct folders:
1. `default`: This folder contains standard or studio-like images of the waste item. These images provide a clear and controlled representation of the item, which can be useful for initial training and testing of waste classification models. Each "default" subfolder contains 250 images.
2. `real_world`: This folder contains images of the waste item in real-world scenarios or environments. These images capture the item in various contexts, such as in a trash bin, on the ground, or in a cluttered environment. The real-world images are essential for evaluating the performance and robustness of waste classification models in practical settings. Each "real_world" subfolder also contains 250 images.

All images in the dataset are provided in the PNG format, ensuring high quality and compatibility with a wide range of image processing and machine learning libraries.

## Waste Categories
The dataset covers a wide range of waste categories and items, including:
- **Plastic:** This category includes images of plastic water bottles, soda bottles, detergent bottles, shopping bags, trash bags, food containers, disposable cutlery, straws, and cup lids. These items represent a significant portion of the plastic waste generated in households and are crucial for recycling efforts.
- **Paper and Cardboard:** This category includes images of newspaper, office paper, magazines, cardboard boxes, and cardboard packaging. These items are commonly recycled and play a vital role in reducing deforestation and conserving natural resources.
- **Glass:** This category includes images of beverage bottles, food jars, and cosmetic containers made of glass. Glass is a highly recyclable material, and proper classification and sorting are essential for effective recycling processes.
- **Metal:** This category includes images of aluminum soda cans, aluminum food cans, steel food cans, and aerosol cans. Metal waste is valuable for recycling and can be efficiently processed when correctly identified and separated.
- **Organic Waste:** This category includes images of food waste, such as fruit peels, vegetable scraps, eggshells, coffee grounds, and tea bags. Organic waste can be composted or used for biogas production, reducing the burden on landfills and generating valuable resources.
- **Textiles:** This category includes images of clothing and shoes. Textile waste is a growing concern, and proper classification can aid in recycling efforts and reduce the environmental impact of the fashion industry.

Please refer to the individual subfolders within the dataset for specific examples and instances of each waste category.

## Dataset Usage and Applications
The Recyclable and Household Waste Classification Dataset offers a wide range of possibilities for researchers, engineers, and environmental enthusiasts. Some potential uses and applications of this dataset include:
- Developing and training machine learning models for automatic waste sorting and categorization. The dataset's diverse range of images and real-world scenarios enables the creation of robust and accurate classification models that can be deployed in waste management facilities, recycling centers, and smart waste bins.
- Analyzing the visual characteristics and features of different waste materials. Researchers can use the dataset to study the unique visual properties of various waste items, such as color, shape, texture, and size. This analysis can contribute to the development of more efficient and targeted waste classification algorithms.
- Comparing the performance of waste classification algorithms on default and real-world images. The dataset allows researchers to evaluate and benchmark the accuracy and robustness of their algorithms in both controlled and realistic settings. This comparison can help identify the strengths and weaknesses of different approaches and guide the development of more reliable waste classification systems.
- Investigating the impact of real-world scenarios on waste recognition accuracy. The dataset's inclusion of real-world images enables researchers to study how factors such as lighting conditions, object occlusion, and background clutter affect the performance of waste classification models. This research can lead to the development of more resilient and adaptable algorithms that can handle the challenges encountered in real-world waste management scenarios.

## Dataset License and Citation
The Recyclable and Household Waste Classification Dataset is available for educational, research, and non-commercial purposes. If you use this dataset in your work, please cite it as follows:

Alistair King, www.kaggle.com/datasets/alistairking/recyclable-and-household-waste-classification

## Contact and Support
If you have any questions, suggestions, or inquiries regarding the dataset, please feel free to reach out via the Kaggle discussion section. We welcome feedback and contributions from the community to further enhance and expand this valuable resource.

Let's work together towards a cleaner and more sustainable future by advancing the field of waste classification and recycling through research and innovation 🤗.

## 可回收及家庭废弃物分类数据集

本数据集汇聚了15,000张（每张256x256像素）全面展示各类可回收材料、一般废弃物和家庭用品的图片，涵盖30个不同的类别。每个类别包含500张图片，每个子类别包含250张图片，该数据集为废弃物分类和回收领域的研发提供了丰富多样的资源。通过提供大量高质量图片，本数据集旨在支持构建稳健且精确的废弃物分拣和分类系统。

## 数据集结构

数据集采用层级文件夹结构组织，以确保便捷的导航和访问。主文件夹“images”包含代表特定废弃物类别或物品的子文件夹。**这些子文件夹名称作为其对应类别的标签，方便研究人员和开发者识别和利用图片以满足特定需求。**

**您需要手动将数据集分割为测试、训练和验证集。**请参阅[可回收及家庭废弃物分类代码](https://www.kaggle.com/code/alistairking/recyclable-and-household-waste-classification-code)以了解如何进行分割。

在每个类别子文件夹中，存在两个不同的文件夹：
1. `default`：该文件夹包含标准或类似摄影棚的废弃物物品图片。这些图片提供了物品的清晰且可控的表示，对于废弃物分类模型的初始训练和测试非常有用。每个“default”子文件夹包含250张图片。
2. `real_world`：该文件夹包含废弃物物品在现实场景或环境中的图片。这些图片捕捉了物品在各种情境下的状态，如垃圾桶内、地面上或杂乱的环境中。现实世界图片对于评估废弃物分类模型在实际环境中的性能和稳健性至关重要。每个“real_world”子文件夹也包含250张图片。

数据集中的所有图片均以PNG格式提供，确保了高质量和与广泛的图像处理和机器学习库的兼容性。

## 废弃物类别

本数据集覆盖了广泛的废弃物类别和物品，包括：
- **塑料**：本类别包括塑料水瓶、汽水瓶、洗涤剂瓶、购物袋、垃圾袋、食品容器、一次性餐具、吸管和杯盖的图片。这些物品代表了家庭产生的塑料废弃物的重要组成部分，对于回收工作至关重要。
- **纸张和纸板**：本类别包括报纸、办公纸、杂志、纸箱和纸板包装的图片。这些物品通常被回收，在减少森林砍伐和保护自然资源方面发挥着至关重要的作用。
- **玻璃**：本类别包括饮料瓶、食品罐头和化妆品容器等玻璃制品的图片。玻璃是一种高度可回收的材料，正确的分类和分拣对于有效的回收过程至关重要。
- **金属**：本类别包括铝制汽水罐、铝制食品罐、钢制食品罐和喷雾罐的图片。金属废弃物对于回收具有重要价值，并且在正确识别和分离后可以高效地处理。
- **有机废弃物**：本类别包括水果皮、蔬菜碎屑、蛋壳、咖啡渣和茶包等食品废弃物的图片。有机废弃物可以用于堆肥或生物气生产，减少对垃圾填埋场的负担并产生有价值的资源。
- **纺织品**：本类别包括衣物和鞋子的图片。纺织品废弃物是一个日益关注的问题，正确的分类有助于回收工作并减少时尚行业对环境的影响。

请参阅数据集中的各个子文件夹以获取每个废弃物类别的具体示例和实例。

## 数据集使用和应用

可回收及家庭废弃物分类数据集为研究人员、工程师和环保爱好者提供了广泛的可能性。本数据集的一些潜在用途和应用包括：
- 开发和训练用于自动废弃物分拣和分类的机器学习模型。数据集丰富的图片和现实场景使创建稳健且精确的分类模型成为可能，这些模型可以部署在废弃物管理设施、回收中心和智能垃圾桶中。
- 分析不同废弃物材料的视觉特征和属性。研究人员可以使用该数据集研究各种废弃物物品的独特视觉特性，如颜色、形状、纹理和大小。这种分析有助于开发更高效和有针对性的废弃物分类算法。
- 比较废弃物分类算法在默认图片和现实世界图片上的性能。该数据集允许研究人员评估和比较其在受控和现实环境中的准确性和稳健性。这种比较有助于识别不同方法的优点和不足，并指导更可靠的废弃物分类系统的开发。
- 研究现实世界场景对废弃物识别准确性的影响。数据集中包含的现实世界图片使研究人员能够研究照明条件、物体遮挡和背景杂乱等因素如何影响废弃物分类模型的性能。这种研究有助于开发更坚韧和适应性强的算法，以应对现实世界废弃物管理场景中遇到的挑战。

## 数据集许可和引用

可回收及家庭废弃物分类数据集可供教育、研究和非商业用途使用。如果您在使用本数据集时将其用于您的作品，请按照以下方式引用：
Alistair King, www.kaggle.com/datasets/alistairking/recyclable-and-household-waste-classification

## 联系和支持

如果您有任何关于数据集的问题、建议或查询，请随时通过Kaggle讨论区联系。我们欢迎社区提供的反馈和贡献，以进一步丰富和扩展这一宝贵资源。

让我们共同努力，通过研究和创新推动废弃物分类和回收领域的发展，共创更清洁、更可持续的未来 🤗。