Comparar métodos

	Desmezcla hiperspectral	Factorización de Matrices No Negativas (NMF)	Clasificación de imágenes basada en píxeles
Campo	Teledetección	Aprendizaje automático	Teledetección
Familia	Machine learning	Latent structure	Machine learning
Año de origen	2002	1999	2007
Autor original	Nirmal Keshava & John Mustard	Lee, D. D. & Seung, H. S.	Remote-sensing classification literature
Tipo	Sub-pixel spectral decomposition algorithm	Matrix decomposition with non-negativity constraints	Supervised/unsupervised spectral image classification
Fuente seminal	Keshava, N., & Mustard, J. F. (2002). Spectral unmixing. IEEE Signal Processing Magazine, 19(1), 44–57.	Lee, D. D., & Seung, H. S. (1999). Learning the parts of objects by non-negative matrix factorization. Nature, 401(6755), 788–791.	Lu, D., & Weng, Q. (2007). A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing, 28(5), 823–870.
Alias	Spectral Mixture Analysis, Linear Spectral Unmixing, Blind Source Separation (Hyperspectral), Hiperspektral Ayrıştırma	NMF, NNMF, nonnegative matrix factorization, non-negative matrix approximation	Per-Pixel Classification, Spectral Classification, Pixel-by-Pixel Classification, Piksel Tabanlı Sınıflandırma
Relacionados	2	4	2
Resumen	Hyperspectral unmixing is a signal processing technique that decomposes each pixel of a hyperspectral image into a collection of pure material spectra (endmembers) and their corresponding fractional abundances. Because sensor resolution often causes multiple land-cover types to co-occupy a single pixel, unmixing recovers sub-pixel compositional information that conventional classification cannot. Keshava and Mustard (2002) provided the foundational signal-processing framework that unified prior geological and remote-sensing work under a rigorous linear mixture model.	Non-negative Matrix Factorization (NMF) is a family of algorithms, introduced by Lee and Seung in their landmark 1999 Nature paper, that decomposes a non-negative data matrix V into the product of two lower-rank non-negative matrices W (basis components) and H (encoding coefficients). Unlike PCA or SVD, the non-negativity constraint forces the algorithm to learn strictly additive, parts-based representations, making the factors directly interpretable as building blocks of the original data.	Pixel-based image classification is a fundamental remote-sensing technique that assigns each individual pixel in a satellite or aerial image to a thematic land-cover category based solely on its spectral values across multiple bands. Systematically surveyed and formalized by Lu and Weng (2007), the approach encompasses both supervised methods—where labeled training samples guide the classifier—and unsupervised clustering approaches that discover natural spectral groupings without prior labels.