一种定标缺失情况下的高光谱目标识别方法

高光谱遥感图像识别在民用和军事领域有着广泛的应用。在缺乏定标信息、缺乏同步观测大气光学参数情况下,对高光谱图像进行地物识别尚没有系统有效的方法,制约了其在定量遥感方向的应用。对此提出了一种利用粒子群算法优化6S模型参数基础上的高光谱遥感数据校正方法,并将其应用于定标缺失情况下的目标识别中。实验表明:在对遥感图像利用少许先验信息选择参数进行校正后,分类准确率为76.25%。而利用粒子群算法优化参数的6S校正后,分类准确率提高到91.58%,目标识别准确率得到了有效提高。

A Hyperspectral Target Recognition Method without Calibration Information

Hyperspectral remote sensing image has been widely used in civil and military applications. Due to the lack of calibration information and atmospheric optical parameters,no systematic and effective method has been specifically developed for hyperspectral targets recognition, which has restricted its application in quantitative remote sensing. A method using particle swarm optimization to choose the parameters in 6S model is proposed,and is applied to hyperspectral target recognition. Simulations show that:without calibration information and atmospheric optical parameters, this method can be used to inverse the reflectance of hyperspectral images. Compared with the empirical method,the classification accuracy based on 6S model with particle swarm optimization algorithm for parameter optimization has been effectively improved from 76.25%to 91.58%.