An Improved Pixel Purity Index Endmember Extraction Algorithm for Hyperspectral Images
Conference: ICMLCA 2021 - 2nd International Conference on Machine Learning and Computer Application
12/17/2021 - 12/19/2021 at Shenyang, China
Proceedings: ICMLCA 2021
Pages: 4Language: englishTyp: PDF
Personal VDE Members are entitled to a 10% discount on this title
Authors:
Yang, Huadong (School of Information Science and Engineering, Shenyang Ligong University, Shenyang, China)
Xu, Nan (BIM and Computing Technology Research Center, Shenyang Jianzhu University, Shenyang, China)
Abstract:
In this work, we derive a new version of pixel purity index endmember extraction algorithm. It resolves several major issues exist in originally PPI. First, the proposed algorithm obtains projection vectors (called skewers) directly from the data points instead of randomly generated unit vectors. Second, it reduces the data points using unconstrained spectral unmixing instead of dimension reduction. Finally, the reduction data are orthogonally projected onto the projection vectors, and endmember can be determined in terms of pixel purity index of each pixel. Both simulation and real hyperspectral data experimental results indicate that the proposed algorithm could overcome the inconsistency and irreproducibility caused by random projection vectors, and could reduce projection computation in much degree.