基于脉冲耦合神经网络模型的混合噪声滤除

脉冲耦合神经网络模型是一种仿生系统，它模拟了猫猴的视觉神经系统，受到广泛研究并已应用于图像处理的各个领域。针对现有脉冲耦合神经网络模型的缺陷设计了一种阈值线性衰减的输出带权均值型PCNN模型，简称为L&A-PCNN，并应用于混合噪声去除领域。进一步扩充了PCNN模型的应用领域，并提高了处理图像的性能。以改进后的模型为基础，通过数学推理和实验获得模型关键参数的最优设置方法和范围。将L&A-PCNN与中值滤波器结合对图像去噪领域的难点混合噪声进行修复，仿真实验结果证明在去噪性能上使用L&A-PCNN的算法比现有算法有5%-30%的提高。

Mixed-Noise Removal based on PCNN

Pulse coupled neural networks (PCNN) model is a bionic system. It emulates the behavior of visual cortical neurons of cats and has been extensively applied in image processing. In this paper, a modified pulse coupled neural networks model was designed and applied to remove mixed noises. The modified model have linear attenuated threshold and weighted averaged gray level output. Hence, it is named as L&A-PCNN. The modified model expands application area of PCNN model, and improves its image processing performance. The optimal methods of setting key parameters are achieved by mathematical reasoning and experiments. Combine the L&A-PCNN and median filter to recover mixed-noise contaminated images. Experimental results show that the new algorithm improves denoising performance 5% to 30% than current algorithm.