A new algorithm for decomposition of mixed pixels based on orthogonal bases of data space is proposed in this paper. It is a simplex-based method which extracts endmembers sequentially using computations of largest si...A new algorithm for decomposition of mixed pixels based on orthogonal bases of data space is proposed in this paper. It is a simplex-based method which extracts endmembers sequentially using computations of largest simplex volumes. At each searching step of this extraction algorithm, searching for the simplex with the largest volume is equivalent to searching for a new orthogonal basis which has the largest norm. The new endmember corresponds to the new basis with the largest norm. This algorithm runs very fast and can also avoid the dilemma in traditional simplex-based endmember extraction algorithms, such as N-FINDR, that it generally produces different sets of final endmembers if different initial conditions are used. Moreover, with this set of orthogonal bases, the proposed algorithm can also determine the proper number of endmembers and finish the unmixing of the original images which the traditional simplex-based algorithms cannot do by themselves. Experimental results of both artificial simulated images and practical remote sensing images demonstrate the algorithm proposed in this paper is a fast and accurate algorithm for the decomposition of mixed pixels.展开更多
N-FINDR is a very popular algorithm of endmember (EM) extraction for its automated property and high efficiency. Unfortunately, innumerable volume calculation, initial random selection of EMs and blind searching for E...N-FINDR is a very popular algorithm of endmember (EM) extraction for its automated property and high efficiency. Unfortunately, innumerable volume calculation, initial random selection of EMs and blind searching for EMs lead to low speed of the algorithm and limit the applications of the algorithm. So in this paper two measures are proposed to speed up the algorithm. One of the measures is substituting distance calculation for volume calculation. Thus the avoidance of volume calculation greatly decreases the computational cost. The other measure is resorting dataset in terms of pixel purity likelihood based on pixel purity index (PPI) concept. Then, initial EMs can be selected well-founded and a fast searching for EMs is achieved. Numerical experiments show that the two measures speed up the original algorithm hundreds of times as the number of EMs is more than ten.展开更多
基金Supported in part by the National Natural Science Foundation of China (Grant No. 60672116)the National High-Tech Research & Development Program of China (Grant No. 2009AA12Z115)the Shanghai Leading Academic Discipline Project (Grant No. B112)
文摘A new algorithm for decomposition of mixed pixels based on orthogonal bases of data space is proposed in this paper. It is a simplex-based method which extracts endmembers sequentially using computations of largest simplex volumes. At each searching step of this extraction algorithm, searching for the simplex with the largest volume is equivalent to searching for a new orthogonal basis which has the largest norm. The new endmember corresponds to the new basis with the largest norm. This algorithm runs very fast and can also avoid the dilemma in traditional simplex-based endmember extraction algorithms, such as N-FINDR, that it generally produces different sets of final endmembers if different initial conditions are used. Moreover, with this set of orthogonal bases, the proposed algorithm can also determine the proper number of endmembers and finish the unmixing of the original images which the traditional simplex-based algorithms cannot do by themselves. Experimental results of both artificial simulated images and practical remote sensing images demonstrate the algorithm proposed in this paper is a fast and accurate algorithm for the decomposition of mixed pixels.
基金Sponsored by the National Natural Science Foundation of China (Grant No 60402025 and 60302019)
文摘N-FINDR is a very popular algorithm of endmember (EM) extraction for its automated property and high efficiency. Unfortunately, innumerable volume calculation, initial random selection of EMs and blind searching for EMs lead to low speed of the algorithm and limit the applications of the algorithm. So in this paper two measures are proposed to speed up the algorithm. One of the measures is substituting distance calculation for volume calculation. Thus the avoidance of volume calculation greatly decreases the computational cost. The other measure is resorting dataset in terms of pixel purity likelihood based on pixel purity index (PPI) concept. Then, initial EMs can be selected well-founded and a fast searching for EMs is achieved. Numerical experiments show that the two measures speed up the original algorithm hundreds of times as the number of EMs is more than ten.
