Hyperspectral Imaging, Hyperspectral Classification, Statistical Test
There are numerous statistics solutions dedicated to hyperspectral classification (considered from the "target detection" point of view) based on hypothesis testing. Examples include, but are not limited to, the following ones:
- C.-I Chang, "Statistical Detection Theory Approach to Hyperspectral Image Classification", 2018 -
Article Statistical Detection Theory Approach to Hyperspectral Image...
- C.-I Chang et al., "Target Detection Approaches to Hyperspectral Image Classification", Chapt. 20, 2022 -
Chapter Target Detection Approaches to Hyperspectral Image Classification
- He et al., "Instance-Based Classification Through Hypothesis Testing", 2021 - https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9333560
Hi
There are several reasons why statistical tests might not be applied:
1. Sample Size and Variability.
2. Marginal Improvements.
3. Computational Complexity.
4. Focus on Other Metrics.
5. Methodological Preference.
6. Lack of Standardization.
generally considered good practice to apply statistical tests in such scenarios to rigorously validate the results
- Badges
- Science topic