SH17 Dataset for PPE Detection
SH17 Dataset for PPE Detection
Datasets
SH17 Dataset for PPE Detection
File
SH17 Dataset for PPE Detection
Use Case
SH17 Dataset for PPE Detection
Description
Explore the SH17 Dataset for PPE Detection, featuring 8,099 images of diverse industrial activities. Curated from Pexels, this dataset includes accurate tags and minimizes regional and racial biases, ideal for developing robust PPE detection models.
Description:
We present the Safe Human (SH17) dataset, comprising 17 different objects. This dataset was curated by scraping images from the Pexels website, which provides clear usage rights for all its images. The images depict a variety of human activities across diverse industrial operations.
Download Dataset
To gather relevant images, we utilized multiple search queries such as “manufacturing worker,” “industrial worker,” “human worker,” and “labor.” The tags associated with Pexels images were found to be reasonably accurate. After removing duplicate samples, we compiled a dataset of 8,099 images. The SH17 dataset showcases significant diversity, representing manufacturing environments from around the world, thereby minimizing potential regional or racial biases.
Dataset Structure
The dataset is organized into multiple components for flexibility and ease of use:
- images/ – Contains all dataset images
- labels/ – YOLO format annotations
- voc_labels/ – Pascal VOC format annotations
- train_files.txt – Training image list
- val_files.txt – Validation image list
This structure allows seamless integration into various machine learning frameworks.
Key Features
- 17 Object Classes: Includes human body parts, safety gear, and tools
- High-Resolution Images: Ranging from 1920×1002 up to 8192×5462
- Dense Annotations: Average of 9.38 objects per image
- Small Object Detection: Thousands of annotations for small objects like ears and earmuffs
- Global Representation: Images from diverse industrial environments
Classes Included
The dataset covers a wide range of categories:
Person, Head, Face, Glasses, Face-mask-medical, Face-guard, Ear, Earmuffs, Hands, Gloves, Foot, Shoes, Safety-vest, Tools, Helmet, Medical-suit, Safety-suit
Potential Applications
This dataset is highly suitable for:
- Industrial Safety Monitoring
- Personal Protective Equipment (PPE) Detection
- Human Activity Recognition
- Smart Surveillance Systems
- AI-driven Workplace Automation
Ethical and Responsible Use
This dataset is intended strictly for research, educational, and machine learning purposes. Since the images are sourced from Pexels:
- Usage must comply with Pexels licensing terms
- Images should not be used in harmful, misleading, or unethical contexts
- Redistribution or resale of unmodified images is not permitted
Users are encouraged to follow ethical AI practices and consider the societal impact of their applications.
Conclusion
The Safe Human (SH17) dataset provides a powerful and diverse foundation for developing AI models focused on industrial safety and human activity recognition. With its high-quality annotations, global diversity, and real-world relevance, it enables accurate, scalable, and responsible AI solutions.
This dataset is sourced from Kaggle.
Contact Us
Quality Data Creation
Guaranteed TAT
ISO 9001:2015, ISO/IEC 27001:2013 Certified
HIPAA Compliance
GDPR Compliance
Compliance and Security
Let's Discuss your Data collection Requirement With Us
To get a detailed estimation of requirements please reach us.