Contribution of non-negative matrix factorization to the classification of remote sensing images

Remote sensing has become an unavoidable tool for better managing our environment, generally by realizing maps of land cover using classification techniques. The classification process requires some pre-processing, especially for data size reduction. The most usual technique is Principal Component Analysis. Another approach consists in regarding each pixel of the multispectral image as a mixture of pure elements contained in the observed area. Using Blind Source Separation (BSS) methods, one can hope to unmix each pixel and to perform the recognition of the classes constituting the observed scene. Our contribution consists in using Non-negative Matrix Factorization (NMF) combined with sparse coding as a solution to BSS, in order to generate new images (which are at least partly separated images) using HRV SPOT images from Oran area (Algeria). These images are then used as inputs of a supervised classifier integrating textural information. The results of classifications of these "separated" images show a clear improvement (correct pixel classification rate improved by more than 20%) compared to classification of initial (i.e. non separated) images. These results show the contribution of NMF as an attractive pre-processing for classification of multispectral remote sensing imagery.

[1]  Ronald C. Henry,et al.  Current factor analysis receptor models are ill-posed , 1987 .

[2]  Danielle Nuzillard,et al.  BSS, Classification and Pixel Demixing , 2004, ICA.

[3]  Christian Jutten,et al.  Space or time adaptive signal processing by neural network models , 1987 .

[4]  J. Eggert,et al.  Sparse coding and NMF , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[5]  Yannick Deville,et al.  A time-frequency blind signal separation method applicable to underdetermined mixtures of dependent sources , 2005, Signal Process..

[6]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[7]  Hao Zhou,et al.  A New Scheme for Blind Decomposition of Mixed Pixels Based on Non-negative Matrix Factorization , 2005, ISNN.

[8]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[9]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[10]  Abdelmounaime Safia,et al.  La transformation en ondelettes pour l'extraction de la texture‐couleur. Application à la classification combinée des images (HRV) de SPOT , 2006 .

[11]  Jean-Francois Cardoso,et al.  Blind signal separation: statistical principles , 1998, Proc. IEEE.

[12]  Patrik O. Hoyer,et al.  Non-negative Matrix Factorization with Sparseness Constraints , 2004, J. Mach. Learn. Res..

[13]  Victoria Stodden,et al.  When Does Non-Negative Matrix Factorization Give a Correct Decomposition into Parts? , 2003, NIPS.

[14]  Erkki Oja,et al.  Independent Component Analysis , 2001 .

[15]  Christian Jutten,et al.  Detection de grandeurs primitives dans un message composite par une architecture de calcul neuromime , 1985 .

[16]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[17]  Yannick Deville,et al.  Temporal and time-frequency correlation-based blind source separation methods. Part I: Determined and underdetermined linear instantaneous mixtures , 2007, Signal Process..

[18]  D. Pham,et al.  Séparation de sources par l'indépendance et la parcimonie , 2003 .