SAFE-Net-Dataset

Dataset Description for SAFE-Net-Dataset1. IntroductionThis dataset was created as part of the research titled " SAFE-Net: Multi-Head Attention Enhanced Framework for Defect Detection in Anti-Corrosion Coatings on Steel Structures". It focuses on detecting defects commonly found in steel structure coatings, including corrosion, cracks, flaking, and blistering. The goal of this dataset is to facilitate research in computer vision for defect detection, specifically for high-resolution image segmentation and object detection tasks.2. Dataset OverviewName: SAFE-Net-DatasetVersion: 1.0Number of Images: 8,193 images.Defects Categories:Class 0: CorrosionClass 1: CrackClass 2: FlakingClass 3: BlisteringData Formats:Images: RGB images in .jpg format.Annotations: Provided in .txt format, where each .txt file corresponds to an image and contains information about the bounding boxes for all objects in that image.3. Methodology3.1. Data CollectionThe images were collected from industrial steel structures under real-world conditions. Images were captured using high-resolution cameras to ensure fine detail resolution of small defects like microcracks.3.2. Data AnnotationAnnotation tools used include Labelimg for bounding boxes. Then, the XML file that has been obtained is to be converted into TXT format.4. CitationIf you use this dataset, please cite the following paper:[Paper Title: SAFE-Net: Multi-Head Attention Enhanced Framework for Defect Detection in Anti-Corrosion Coatings on Steel Structures]Authors: [Yue Yu, Shouchao Jiang, Yijun Wang, Shaojun Zhu*]5. LicenseThis dataset is released under the GPL license. Users are free to use, modify, and distribute the dataset, provided that appropriate credit is given to the original authors.For questions or issues, please contact [e-mail].