Statistical Test Results for Urban Road Crack Detection Method

Jun Qiao's statistical test results from a study on automated urban road crack detection. The dataset, last updated in April 2026, compares the performance of a novel U-net and ResNeXt fusion method against other detection techniques. It likely contains metrics evaluating detection accuracy, memory efficiency, and classification performance for different crack types.

Use Cases

• Benchmarking computer vision model performance based on the described detection metrics
• Analyzing memory efficiency for edge deployment based on the reported 2.1GB video memory peak
• Comparing classification accuracy for horizontal and vertical cracks versus other crack types as mentioned in the description

Strengths

• Dataset is openly licensed under CC-BY-4.0
• Specific performance metric reported: peak video memory efficiency controlled within 2.1GB
• Results compare a proposed method against numerous other detection methods

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment
• The 5.5 KB file size suggests a very limited scope, likely containing summary statistics rather than raw data

Provenance

Source

figshare, author Jun Qiao

Collection Method

Likely generated from experimental results of a machine learning model for road crack detection.

Time Range

Update date is 2026-04-21; temporal coverage of the underlying study is unknown.

Freshness

Last updated 2026-04 21 17:34:38

Geography

Focus is on urban roads; specific geographic location is not stated.