论文部分内容阅读
高斯径向基核函数是基于光谱向量间欧氏距离的度量,对于因光照强度变化而引起的地物光谱变异敏感,当同类地物光谱发生变异时,基于高斯径向基核的高光谱影像地物检测算法的性能下降。为了解决该问题,基于光谱曲线形状相似性描述提出了光谱角度余弦核测度这一非正定核函数,并应用于一种非正定OCSVM方法的高光谱影像地物检测。最后利用两幅高光谱影像进行了实验分析,实验结果证明了本文算法的有效性。
Gaussian Radial Basis Function is a measure based on the Euclidean distance between spectral vectors, which is sensitive to the spectral variations caused by the change of illumination intensity. When the spectra of the same kind of objects change, hyperspectral images based on Gaussian radial base nuclei The performance of the object detection algorithm is degraded. In order to solve this problem, the non-positive definite kernel function of spectral angle cosine kernel is proposed based on the description of shape similarity of spectral curve, and applied to the detection of hyperspectral image features by non-positive OCSVM method. Finally, two hyperspectral images are used for experimental analysis, and the experimental results show the effectiveness of the proposed algorithm.