TextileClass-7: Dataset for Fabric Material Classification

Description

This dataset presents a comprehensive collection of real-world fabric images developed to support research in computer vision, image classification, and intelligent textile recognition. The dataset consists of seven commonly used fabric categories: Cotton, Cotton Mixed, Denim, Polyester, Silk, Viscose, and Wool. Images were captured under diverse real-world conditions, including variations in illumination, viewing angles, texture patterns, wrinkles, folds, and background complexity, making the dataset suitable for developing robust deep learning models. All images were manually annotated using the Roboflow annotation platform specifically for image classification. Unlike object detection or instance segmentation datasets, no bounding boxes, polygons, or segmentation masks were created. Instead, each image was assigned a single class label corresponding to the dominant fabric type. The annotation process included careful manual verification to ensure label consistency and minimize annotation errors. The dataset is organized into class-specific folders and is compatible with common deep learning frameworks such as TensorFlow, PyTorch, and Keras. It can be readily used for supervised image classification tasks, transfer learning, benchmark evaluations, and comparative studies of machine learning and deep learning algorithms. Researchers may also utilize this dataset for feature extraction, texture analysis, explainable artificial intelligence (XAI), and vision-based textile inspection systems. This dataset provides a valuable benchmark for the textile industry and the research community by enabling the development of automated fabric recognition systems for smart manufacturing, quality inspection, inventory management, e-commerce product categorization, and intelligent textile applications.

Steps to reproduce

Steps to Reproduce : 1. Collect real-world fabric images representing the seven predefined classes: Cotton, Cotton Mixed, Denim, Polyester, Silk, Viscose, and Wool. 2. Perform image quality screening to remove blurred, duplicate, or low-quality images. 3. Upload the images to the Roboflow platform. 4. Create a Classification project and define the seven fabric classes. 5. Manually assign a single class label to each image. No bounding boxes, polygons, or segmentation masks are used. 6. Review and verify all annotations to ensure labeling consistency and accuracy. 7. Export the annotated dataset in the desired image classification format (e.g., Folder Structure, CSV, TensorFlow, PyTorch, or YOLO Classification). 8. Split the dataset into training, validation, and testing subsets according to the experimental protocol. 9. Train an image classification model (e.g., ResNet, EfficientNet, MobileNet, Vision Transformer, or other CNN architectures) using the provided data. 10. Evaluate the trained model using standard classification metrics such as Accuracy, Precision, Recall, F1-score, and Confusion Matrix. This procedure enables researchers to reproduce the dataset preparation process and benchmark fabric classification models using the same annotation methodology.

Institutions

• Southeast University

  Dhaka Division, Dhaka
• United International University

  Dhaka Division, Dhaka

Categories

Licence