基于组织协同进化分类算法的遥感图像目标识别

Remote Sensing Target Recognition Based on Organizational Coevolutionary Classification Algorithm

针对遥感图像目标识别问题,提出了一种基于组织协同进化分类算法的识别方法。它没有复杂的运算,训练和识别速度都很快。对实测遥感图像的实验表明,本文方法性能稳定,优于文献[3]和[4]中基于支撑矢量机的方法,识别率均达到了95％以上,且训练时间非常短,不到1秒钟。

An organizational coevolutionary classification algorithm is proposed for remote sensing target recognition. This method has not complex computation, so its computational cost is very low. Experimental results show our method outperforms the methods proposed in literature [3] and [4], which is based on SVM, and its predictive accuracy is higher than 95%. furthermore, the training the of our method is less than 1 second.