基于传统字典目标检测算法存在的不足,通过对原始字典进行增殖来提高字典中训练样本的数目,提出了基于子空间异常增殖字典的高光谱图像目标检测算法(SOPDSR).实验仿真证明了此算法可以提高目标检测的精度。
For the lack of the traditional dictionary, the target detection algorithm based on the subspace outlier proliferation dictionary (SOPDSR) is proposed. The subspace outlier proliferation dictionary is built by adding the number of training samples in the traditional dictionary. The simulation experiments are carried out to prove that the subspace outlier proliferation dictionary target detection algorithm does work and the new algorithm can improve the accuracv in target detection