Multispectral Texture Benchmark

0579556 - ÚTIA 2024 RIV NL eng C - Conference Paper (international conference)
Kříž, P. - Haindl, Michal
Multispectral Texture Benchmark.
Procedia Computer Science : Volume 225, 27th International Conference on Knowledge Based and Intelligent Information and Engineering Sytems, KES 2023. Amsterdam: Elsevier, 2023, s. 3143-3152. ISSN 1877-0509.
[International Conference on Knowledge-Based and Intelligent Information & Engineering Systems 2023 (KES 2023) /27./. Athens (GR), 06.09.2023-08.09.2023]
Institutional support: RVO:67985556
Keywords : textural features * benchmark * representation * multispectral features
OECD category: Automation and control systems
Method of publishing: Open access
http://library.utia.cas.cz/separaty/2023/RO/haindl-0579556.pdf https://www.sciencedirect.com/science/article/pii/S1877050923014667?via%3Dihub

Dozens of textural features have been published, but their realistic validation for efficient recognition applications still needs to be discovered. Textural features are derived using various approaches. We present a benchmark that can be used to evaluate these features and categorize them based on their information efficiency. We propose how the features can be benchmarked and explain different ways of measuring their properties and performance. Most textural feature-extracting algorithms are only based on information extraction from monospectral images (gray-level). Apart from native multispectral algorithms, we generalize some of these originally monospectral features for hyperspectral textures in our illustrating examples.
Permanent Link: https://hdl.handle.net/11104/0349560