论文部分内容阅读
为有效滤除图像中严重脉冲噪声干扰,提出了一种基于改进型脉冲耦合神经网络(PCNN)噪声检测的两级脉冲噪声滤除算法。该算法首先利用PCNN同步脉冲发放特性区分定位噪声点和信号点位置,其次根据噪声点局部邻域信息对噪声进行第1级自适应滤波,然后再利用具有保护边缘细节特点的多方向信息中值滤波器(MF进行第2级辅助滤波。实验结果表明,该算法在噪声检测中无需设定检测阈值,噪声检测精度较高;在去噪过程中不但有效滤除噪声干扰,而且能很好地保护图像边缘细节等信息,具有较好的主观视觉效果和客观评价指标,比传统MF及其它相关算法有更优的滤波性能,去噪能力强、信噪比高和适应性好,特别是对受严重噪声污染的图像,显示了更大的优越性。
In order to effectively filter out the serious impulsive noise in the image, a two-stage impulse noise filtering algorithm based on improved pulse coupled neural network (PCNN) noise detection is proposed. Firstly, PCNN synchronization pulse characteristics are used to distinguish the location of noise and signal points. Secondly, the noise is first-level adaptively filtered according to local neighborhood information of noise point, and then the median of multi-directional information with protection edge details Filter (MF), the second level of auxiliary filtering.The experimental results show that the algorithm need not set the detection threshold in the noise detection, the noise detection accuracy is high; in the denoising process not only effectively filter out noise interference, but also well Protect the image edge details and other information, has better subjective visual effects and objective evaluation index than traditional MF and other related algorithms have better filtering performance, denoising ability, high signal to noise ratio and good adaptability, especially for Images contaminated with severe noises show even greater advantages.