论文部分内容阅读
将独立成分分析(ICA)算法用于高光谱图像解混时,算法对丰度的独立性要求与实际地物分布相矛盾;同时,采用梯度算法对解混目标函数进行优化时,易收敛到局部极值点。针对上述问题,提出在非负ICA(NICA)模型的目标函数中引入丰度和为一约束(ASC),确保解混出的丰度与实际地物分布一致;同时,采用布谷鸟搜索(CS)算法,利用其优异的全局搜索性能对提出的目标函数进行优化求解。为减少参数维数并缩小CS算法的搜索范围,利用矩阵QR分解理论,将对解混矩阵的搜索转化为对一系列Gives矩阵的识别。仿真数据和真实高光谱图像数据实验结果表明,提出的算法能有效地克服上述问题,在噪声为30dB、像元纯度为0.8时,解混指标光谱角距离(SAD)和均方根误差(RMSE)达到了0.03以下,达到良好解混效果。
When the Independent Component Analysis (ICA) algorithm is used to decompose hyperspectral images, the algorithm’s independence of abundance contradicts the actual object distribution. Meanwhile, when using the gradient algorithm to optimize the unmixing objective function, it is easy to converge to Local extreme points. In view of the above problems, it is proposed to introduce the abundance and a constraint (ASC) into the objective function of the non-negative ICA (NICA) model to ensure that the unmixed abundance is consistent with the distribution of the actual features. At the same time, ) Algorithm, using its excellent global search performance to optimize the proposed objective function. In order to reduce the parameter dimension and narrow the search range of CS algorithm, the matrix QR decomposition theory is used to transform the search of the unmixing matrix into the identification of a series of Gives matrices. Simulation results and real hyperspectral image data show that the proposed algorithm can effectively overcome the above problems. When the noise is 30dB and pixel purity is 0.8, the SAD and RMSE ) Reached below 0.03, to achieve a good solution effect.