群体智能算法的遥感图像处理研究

针对传统图像增强方法缺乏适应性的缺点, 提出了一种用最优化过程进行图像增强的方法。首先对量子粒子群优化(quantum-behaved particle swarm optimization, QPSO)算法进行改进, 提出了一种实变参数量子粒子群优化(time varying parameters QPSO, QPSO-tp)算法。标准测试函数的实验结果表明, 改进后的算法在全局搜索能力和收敛精度上要优于原QPSO算法, 具有调节参数少、随机性更强等优点。然后将遥感灰度图像的非线性变换增强过程用最优化问题进行处理, 用QPSO-tp算法进行参数寻优。实验结果表明, 图像的增强效果得到了较大提高。

Research on swarm intelligence algorithm for remote sensing image processing

This paper proposed an image enhancement method with optimization problem as the lack of adaptability of traditional methods. Firstly, it presented a QPSO-tp algorithm. And standard test function experimental results show that the improved algorithm is superior to original QPSO algorithm on global search ability and convergence accuracy, and it has few adjustable parameters and stronger randomness. Secondly, it processed the nonlinear-transform enhancement of remote sensing gray-scale image with the QPSO-tp algorithm for optimization parameters. The experimental results show that the image enhancement effect is improved significantly.