论文部分内容阅读
智能分类算法是遥感影像分类研究的热点,遗传算法作为一种智能全局优化技术在遥感影像分类中具有良好应用前景.针对现有多光谱遥感影像分类方法的不足,提出了基于自适应遗传算法的超平面分类方法(hyper plane-adaptive genetic algorithm,HP-AGA)并应用于遥感影像分类,该方法利用神经网络中的神经元激活函数Sigmoid函数,对遗传算法中交叉率、变异率进行非线性自适应性调整,不再需要反复训练遗传参数,同时利用快速全局寻优特点,确定分类超平面的各个位置参数,从而获取最佳分类超平面集进行分类.多光谱遥感影像分类方法的应用实验表明,基于自适应遗传算法的超平面遥感分类方法能更快、更稳定地收敛到全局最优解,具有更好的效率及鲁棒性,并能取得优于简单遗传超平面分类算法及传统分类方法的分类精度.
Intelligent classification algorithm is a hot research field of remote sensing image classification, and genetic algorithm as a kind of intelligent global optimization technology has a good application prospect in remote sensing image classification.Aiming at the shortcomings of the existing multi-spectral remote sensing image classification methods, this paper proposes an adaptive genetic algorithm This paper uses the neuron activation function Sigmoid function in neural network to study the nonlinearity of the crossover rate and mutation rate in genetic algorithm Adaptability adjustment, it is no longer necessary to repeatedly train genetic parameters, and at the same time, the fast global optimization feature is used to determine the location parameters of the classification hyperplane, so as to obtain the best classification hyperplane set for classification. The application of multi-spectral remote sensing image classification method shows that , The hypersurface remote sensing classification method based on adaptive genetic algorithm can converge to the global optimal solution faster and more stably with better efficiency and robustness and can achieve better performance than the simple genetic hyperplane classification algorithm and the traditional classification Method classification accuracy.