摘要
图像信号在获取、传送过程中往往因环境、成像系统等因素的影响,导致图像出现噪声污染严重影响图像有效分割、图像边缘检测以及图像目标提取等一系列图像处理工作。为了提高图像质量,滤除图像噪声,本文提出了一种改进型脉冲耦合神经网络图像噪声滤波算法。该算法在传统的脉冲耦合神经网络(PCNN)基础上对其进行改进,在传统的PCNN相似神经元能够同步点火对图像噪声进行自适应检测的基础上,对图像噪声采用自适应滤波法以及对前面的滤波结果采用多方向信息中值滤波的方法再进行处理。实验说明本文提出的方法能够提高噪声检测精度,在不损失图像边缘等重要信息的情况下有效对噪声进行滤除,与传统除噪声算法相比具有更好的性能以及适应性。
In obtaining and transmitting the image signal, due to environment or influence of the imaging system, causes noise pollution seriously affects a series of images like image segmentation, image edge detection and image extraction and processing. In order to improve the image quality and remove the image noise, a modified pulse coupled neural network image noise filtering algorithm is proposed in this paper. The algorithm is improved based on the tradi- tional pulse-coupled neural networks（PCNN） ,on the basis the traditional PCNN similar neurons firing synchronously and adaptive detection for image noise, using adaptive filtering method and that the previous filtering results with multi -directional information median filtering method and then processed. Experiment shows this method can improve the noise detection accuracy, effectively to filter out the noise without any loss of image edges and other important informa- tion, compared with traditional noises algorithm, have better performance and adaptability.
出处
《激光杂志》
北大核心
2016年第1期142-144,共3页
Laser Journal
基金
河南省科学技术厅科技成果鉴定项目(2015102)
关键词
图像信号
脉冲耦合神经网络
噪声污染
自适应滤波
image signal
pulse coupled neural network
noise pollution
adaptive filtering
