局部非负稀疏编码的高光谱目标检测方法研究

Hyperspectral Image Target Detection Approach Based on Local Non-negative Sparse Coding

基于稀疏编码的高光谱图像处理算法能够挖掘高光谱高维数据空间中潜在的数据相关性,能自然地贴近光谱信号的本质特征。本文提出基于局部非负稀疏编码的高光谱目标检测算法。与经典稀疏编码模型相比,非负稀疏编码对编码系数进行非负约束,一方面使得线性编码具有明确的物理解释,另一方面增强了系数的可分性与稳健性。算法首先通过双窗口设计构造局部动态字典,然后利用目标和背景在动态字典上编码的稀疏性差异进行阈值分割最后通过统计判决实现目标检测。仿真数据以及真实数据实验结果证明了算法的有效性。

Sparse representation based hyperspectral image processing methods can excavate potential relationship in high-dimensional hyperspectral data and reveal the essential characteristic of spectral signal. In this paper a novel hyper- spectral image target detection algorithm based on local non-negative sparse coding is presented. Compared with classical sparse representation methods, the linear coding coefficients are enforced non-negative. On one side the linear coding process has tangible physical interpretation. On the other side the coding coefficients are proved more discriminative and ro- bust. The locally dynamic dictionary is first constructed with atoms which are produced by a sliding dual window strategy. Then non-negative coefficients of each pixel are calculated with the dynamic dictionary. The discrimination between targets and background is based on the sparsity of the coefficients. We carried extensive experiments on both simulated and real data to verify the effectiveness of the proposed method.