Speed-up for N-FINDR algorithm  

作　　者：王立国 张晔 

机构地区：[1]Dept.of Information Engineering,Harbin Institute of Technology [2]Dept.of Information Engineering, Harbin Institute of Technology

出　　处：《Journal of Harbin Institute of Technology(New Series)》2008年第1期141-144,共4页哈尔滨工业大学学报（英文版）

基　　金：Sponsored by the National Natural Science Foundation of China (Grant No 60402025 and 60302019)

摘　　要：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.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.

关 键 词：endmember extraction N-FINDR algorithm PPI algorithm spectral unmixing 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]