Dataset Viewer
Auto-converted to Parquet Duplicate
file_name
stringclasses
3 values
quality
stringclasses
3 values
car_number
stringclasses
3 values
object_type
stringclasses
2 values
bounding_box
stringclasses
3 values
detection_confidence
stringclasses
1 value
weather_condition
stringclasses
2 values
lighting_condition
stringclasses
2 values
damage_severity
stringclasses
3 values
maintenance_status
stringclasses
2 values
0a8c3a8e2a92dc14d0abf9c0aa3afda9.jpg
1280*769
Unrecognizable
Compartment
[100, 150, 500, 300]
High
Clear
Night
No apparent damage
Unknown
8825facf0626af264d44cfa43f1d089e.jpg
1154*1538
Unknown
Carriage
Coordinates Unknown
High
Sunny
Daytime
Minor
Unknown
edd05548c45921c0b7b01cfaa0a83e4a.jpg
1280*720
2213
Carriage
[100, 200, 300, 400]
High
Sunny
Daytime
None
Normal

Train Carriage Operational Safety Dataset

The current transportation industry faces significant challenges in train operational safety, particularly in carriage monitoring and accident prevention. Most existing monitoring systems rely on manual inspections, which are inefficient and difficult to comprehensively cover, making it hard to eliminate safety hazards. This dataset aims to support the research and application of automated object detection technology by providing high-quality images for carriage safety monitoring, enhancing monitoring efficiency and reducing the accident rate. The dataset includes images of train carriages in operation, captured in different environments and time periods with high resolution to ensure image quality. Professional photography equipment was used during data collection, capturing in various lighting and weather conditions, and rigorous quality control measures, including multiple rounds of annotation and expert review, were taken to ensure the consistency and accuracy of data annotation. The data is stored in JPEG format, with a clear structure that is easy to use for subsequent machine learning tasks.

Technical Specifications

Field Type Description
file_name string File name
quality string Resolution
car_number string A unique identifier for each train car.
object_type string The type of object identified in the image, such as car, obstacle, etc.
bounding_box string Coordinates of the rectangular bounding box around the object.
detection_confidence float Confidence score when a target is detected.
weather_condition string Weather conditions at the time of image capture, such as sunny, rainy, etc.
lighting_condition string Lighting level at the time of image capture, such as daytime, nighttime.
damage_severity string The severity of the damage detected on the car, such as minor, severe.
maintenance_status string Current maintenance status information of the car.

Compliance Statement

Authorization Type CC-BY-NC-SA 4.0 (Attribution–NonCommercial–ShareAlike)
Commercial Use Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization No PII, no real company names, simulated scenarios follow industry standards
Compliance System Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Source & Contact

If you need more dataset details, please visit Mobiusi. or contact us via contact@mobiusi.com

Downloads last month
8