Fast Endmember Extraction for Massive Hyperspectral Sensor Data on GPUs

Hyperspectral imaging sensor becomes increasingly important in multisensor collaborative observation. The spectral mixture problem seriously influences the efficiency of hyperspectral data exploitation, and endmember extraction is one of the key issues. Due to the high computational cost of algorithm and massive quantity of the hyperspectral sensor data, high-performance computing is extremely demanded for those scenarios requiring real-time response. A method of parallel optimization for the well-known N-FINDR algorithm on graphics processing units (NFINDR-GPU) is proposed to realize fast endmember extraction for massive hyperspectral sensor data in this paper. The implements of the proposed method are described and evaluated using compute unified device architecture (CUDA) based on NVIDA Quadra 600 and Telsa C2050. Experimental results show the effectiveness of NFINDR-GPU. The parallel algorithm is stable for different image sizes, and the average speedup is over thirty times on Telsa C2050, which satisfies the real-time processing requirements.

[1]  Antonio J. Plaza,et al.  Real-Time Implementation of the Pixel Purity Index Algorithm for Endmember Identification on GPUs , 2014, IEEE Geoscience and Remote Sensing Letters.

[2]  Antonio J. Plaza,et al.  Recent Developments in High Performance Computing for Remote Sensing: A Review , 2011, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[3]  Chein-I Chang,et al.  Estimation of number of spectrally distinct signal sources in hyperspectral imagery , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Antonio J. Plaza,et al.  Parallel implementation of endmember extraction algorithms using NVidia graphical processing units , 2009, 2009 IEEE International Geoscience and Remote Sensing Symposium.

[5]  Antonio J. Plaza,et al.  Real-Time Implementation of the Vertex Component Analysis Algorithm on GPUs , 2013, IEEE Geoscience and Remote Sensing Letters.

[6]  S. Pascucci,et al.  Hyperspectral Sensor Data Capability for Retrieving Complex Urban Land Cover in Comparison with Multispectral Data: Venice City Case Study (Italy) , 2008, Sensors.

[7]  P. J. Williams,et al.  Data fusion on a distributed heterogeneous sensor network , 2006, SPIE Defense + Commercial Sensing.

[8]  Qian Du,et al.  High Performance Computing for Hyperspectral Remote Sensing , 2011, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[9]  Wenfei Luo,et al.  Parallel implementation of N-FINDR algorithm for hyperspectral imagery on hybrid multiple-core CPU and GPU parallel platform , 2011, International Symposium on Multispectral Image Processing and Pattern Recognition.

[10]  Antonio J. Plaza,et al.  Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[11]  Sanguthevar Rajasekaran,et al.  Fast GPU algorithms for endmember extraction from hyperspectral images , 2012, 2012 IEEE Symposium on Computers and Communications (ISCC).

[12]  Antonio J. Plaza,et al.  Parallel implementation of the N-FINDR endmember extraction algorithm on commodity graphics processing units , 2010, 2010 IEEE International Geoscience and Remote Sensing Symposium.

[13]  Antonio J. Plaza,et al.  Real-time implementation of remotely sensed hyperspectral image unmixing on GPUs , 2012, Journal of Real-Time Image Processing.

[14]  Mario Winter,et al.  N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data , 1999, Optics & Photonics.