摘要
鉴于Gamma分布的SAR图像相干斑经对数变换后可近似为高斯分布,提出一种基于粒子群优化的BP神经网络复原去噪算法。首先用高斯噪声对无噪图像进行模糊处理,然后将结果和原图像组成训练对,用于训练优化后的神经网络,最后利用训练好的神经网络对SAR图像进行复原,从而达到去除相干斑的目的。实验表明,该算法能有效解决传统去噪算法在图像失真、边缘模糊方面的问题,收敛速度快,迭代次数少,归一化均方误差(NMSE)和峰值噪比(PSNR)效果更好。
After processing by logarithmic transformation,the Gamma distribution speckle of SAR images are analogous to Gasussian distribution.In view of this,a BP neural network restoration denoising method based on particle swarm optimi-zation is proposed.Firstly,noiseless images are process by Gasussian noise.then,the result image and the noiseless images are made training pair,which is used in training the optimizational BP neural network.Lastly,using the BP neural network to restore SAR Images for the purpose of removing speckle.The experiment shows,compared with traditional denoising algo-rithm,the method can effectively solve the problem of image distortion and edge burring,have fast convergence rate and less iterations,is better in normalized mean square error （NMSE）and peak signal-to-noise ratio （PSNR）.
出处
《计算技术与自动化》
2014年第3期92-96,共5页
Computing Technology and Automation
基金
国防十二五预研基金项目(40405070102)
