针对目前体积最大化端元提取技术存在提取精度低、计算复杂性高等问题,提出了结合遗传算法和体积最大化的端元提取方法(Extraction Method Based on Genetic Algorithm and Volume Maximization,EE-GAVM),以改进现有的端元提取技术.EE-GAVM方法通过寻找单纯形体积最大的像元向量集合来定位到端元集合,将端元提取问题表述为单一目标,节约计算资源的同时实现了精度的提高.实验结果表明:所提方法在性能和精度方面优于其他对比算法.
Endmember Extraction from Hyperspectral Images Based on Genetic Algorithm and Volume Maximization
Aiming at the problems of low extraction accuracy and high computational complexity in the current volume maximization endmember extraction technology,the Extraction Method Based on Genetic Algorithm and Volume Maximization (EE-GAVM) is proposed to improve the existing endmember extraction methods. The EE-GAVM method locates the endmember set by searching for the pixel vector set with the largest volume of the simplex,and expresses the endmember extraction problem as a single target,which saves computing resources and improves accuracy. The experimental results show that the proposed method outperforms other comparative algorithms in terms of performance and accuracy.
万方数据
维普
