张贵珍,
李崇禧,
刘婷云,
唐祎玲,
南昌大学信息工程学院 ??南昌 ??330031
基金项目:国家自然科学基金(61662044, 61163023, 51765042, 81501560)，江西省自然科学基金(20171BAB202017)，江西省研究生创新项目(YC2018-S066)
详细信息
作者简介:徐少平：男，1976年生，博士，教授，博士生导师，研究方向为图形图像处理技术、机器视觉、虚拟手术仿真等
张贵珍：女，1993年生，硕士生，研究方向为图像处理与机器学习
李崇禧：男，1994年生，硕士生，研究方向为图像处理与机器学习
刘婷云：女，1996年生，硕士生，研究方向为图像处理与机器学习
唐祎玲：女，1977年生，博士生，研究方向为图像处理与机器学习
通讯作者:唐祎玲 tangyiling@ncu.edu.cn
中图分类号:TP391计量
文章访问数:2161
HTML全文浏览量:557
PDF下载量:69
被引次数:0
出版历程
收稿日期:2018-06-06
修回日期:2018-12-07
网络出版日期:2018-12-13
刊出日期:2019-05-01
A Fast Random-valued Impulse Noise Detection Algorithm Based on Deep Belief Network
Shaoping XU,Guizhen ZHANG,
Chongxi LI,
Tingyun LIU,
Yiling TANG,
School of Information Engineering, Nanchang University, Nanchang 330031, China
Funds:The National Natural Science Foundation of China (61662044, 61163023, 51765042, 81501560), The Project of Jiangxi Province Natural Science Foundation (20171BAB202017), The Jiangxi Provincial Graduate Innovation Special Fund (YC2018-S066)
摘要
摘要:为提高现有随机脉冲噪声(RVIN)检测算法的检测准确率和执行效率，该文试图从构建描述能力更强的特征矢量和训练非线性映射更为准确的预测模型两个方面入手，实现一种基于训练策略的快速RVIN检测算法。一方面，提取多个不同阶的对数绝对差值排序统计值并结合一个能够反映图像边缘特性的统计值作为刻画图块中心像素点是否为噪声的特征矢量。在计算量增加极少的情况下，显著提升了特征矢量的描述能力。另一方面，基于深度置信网络(DBN)训练RVIN预测模型(RVIN检测器)将特征矢量映射为噪声类型标签，实现了比浅层预测模型更为准确的映射。大量实验数据表明：与现有的RVIN检测算法相比，所提算法在检测准确率和执行效率两个方面都更有优势。
关键词:随机脉冲噪声/
噪声检测/
图像局部统计值/
深度置信网络/
计算效率
Abstract:To improve the detection accuracy and execution efficiency of the existing Random-Valued Impulse Noise (RVIN) detectors, a fast training-based RVIN detection algorithm is implemented by constructing a more descriptive feature vector and training a detection model with more accurate nonlinear mapping. On the one hand, multiple Rank-Ordered Logarithmic absolute Deviation (ROLD) statistics are extracted and combined with a statistical value reflecting the edge characteristics in the form of feature vector to describe how RVIN-like the center pixel of a patch is. The description ability of the feature vector is improved significantly while the computational complexity is just increased in small amount. On the other hand, an RVIN prediction model (RVIN detector) is obtained by training a Deep Belief Network (DBN) to map the feature vectors to noise labels, which is more accurate than the shallow prediction model. Extensive experimental results show that, compared with the existing RVIN detectors, the proposed one has better performance in terms of detection accuracy and execution efficiency.
Key words:Random-Valued Impulse Noise (RVIN)/
Noise detection/
Local image statistic/
Deep Belief Network (DBN)/
Computational efficiency
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=be6f3034-372c-4b38-9aac-c3cbb6d262d9
