WHU Buildings Dataset: SLIC Segmentation Results for Parameter k

Asim Shoaib published average performance results for the SLIC image segmentation algorithm on the WHU Buildings Dataset. The data contains results for three and four features incorporated as MC for SLIC parameter k values of 500, 1000, 1500, 2000, and 3000, calculated over 1,550 satellite images. The dataset was last updated on May 15, 2026.

Use Cases

• Benchmarking the SLIC segmentation algorithm based on different parameter (k) values.
• Comparing the impact of incorporating three versus four features as MC on segmentation performance.
• Analyzing segmentation accuracy trends on a large set of building imagery.
• Validating remote sensing image segmentation methods against a known dataset.

Strengths

• Results are based on a substantial set of 1,550 satellite images.
• Performance is evaluated across five distinct SLIC parameter (k) values (500, 1000, 1500, 2000, 3000).
• Dataset has a clear, permissive license (CC-BY-4.0).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small (5.5 KB), indicating limited scope, likely containing only summary statistics.

Provenance

Source

Asim Shoaib via figshare

Collection Method

Likely contains aggregated performance metrics from computational experiments.

Freshness

Last updated 2026-05-15 17:26:41; freshness should be verified.