基于改进量子粒子群的红外图像增强算法

为了提高红外图像增强的效果,采用改进量子粒子群算法。首先构造粒子群多层空间结构,粒子运行空间划分为主层空间、次层空间,粒子信息交流通过主层空间、次层空间分别进行,交流过程受自身信息度因子和交流度因子影响；接着量子旋转门更新通过镜像门操作,梯度方法自适应调整量子门的旋转角度,提高了算法性能；最后将红外图像高频分量和低频分量分离,对其分别进行增强。实验仿真显示本文算法对红外图像增强结果相比其他算法较清晰,优质系数评价指标相比HE、NSCT、MSR、PSO、RSQS算法分别提高了43.60、36.52、25.60、19.24、12.14,对比度指标相比HE、NSCT、MSR、PSO、RSQS算法分别提高了27.99、20.70、15.28、13.97、10.85,性能指标较优。

Infrared image enhancement based on improved quantum particle swarm optimization

In order to improve the effect of infrared image enhancement,an improved quantum particle swarm optimization algorithm is proposed. Firstly,the multi layer space structure of particle swarm optimization was constructed,and the particle running space was divided into the main layer space and the sub layer space. The particle information communication was carried out through the main layer space and the sub layer space respectively,and the communication process was affected by its own information degree factor and communication degree factor. Secondly,the quantum rotation gate was updated through the mirror gate operation,and the rotation angle of the quantum gate was adjusted through the gradient method adaptively,so that improved the algorithm performance. Finally,the high frequency and low frequency components of the infrared image were separated and enhanced respectively. The simulation results show that the improved quantum particle swarm optimization would enhance the infrared image which clearer than other algorithms,the evaluation index of high quality coefficient is improved 43.60%,36.52%,25.60%,19.24% and 12.14% respectively compared with HE,NSCT,MSR,PSO and RSQS,the contrast index is improved 27.99%,20.70%,15.28%,13.97% and 10.85% respectively compared with HE,NSCT,MSR,PSO and RSQS. It is proved that the performance index obtained by this algorithm is better than using other algorithms.