BTC-Net V2: Processed Hyperspectral Images from AVIRIS Sensor

BTC-Net V2 is a dataset of processed Hyperspectral Image (HSI) data from the AVIRIS sensor, structured for compression research. The dataset was released by author 'the-bad-one' and was last updated on 2026-06-04. It is hosted on Hugging Face and is intended to support the BTC-Net V2 model for spatial-priority hierarchical reconstruction.

Use Cases

• Training neural networks for hyperspectral image compression based on spatial-priority hierarchical reconstruction.
• Benchmarking compression algorithms using processed data from the AVIRIS sensor.
• Developing and validating new methods for spectral-spatial feature extraction in remote sensing.

Strengths

• Data is sourced from the AVIRIS sensor, a standard in airborne hyperspectral imaging.
• Dataset is officially released and associated with a published model architecture (BTC-Net V2).
• Last update timestamp is precisely recorded as 2026-06-04 11:35:43.

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.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.

Provenance

Source

AVIRIS sensor, processed by the dataset author.

Collection Method

Likely processed and structured for machine learning model training.

Freshness

Last updated 2026-06-04 11:35:43; freshness should be verified.