Hyper-spectral picture analysis

In our project, we investigated the impact of the PCA (principal component analysis) process on the MF (matched filter) algorithm. Our project involved examining the influence of the PCA procedure on the MF algorithm. We aimed to comprehend the significance of each stage and determine how we can forecast the algorithm's performance prior to its implementation. Furthermore, we observed the functionality of PCA with various filters. Our primary innovation was utilizing the $PL_0$ MF for the PCA space. We focused on the data's stationarity and attempted to enhance the efficiency of the PCA space by incorporating the local properties of each pixel.