基于邻域关联特性的纸病去噪方法

An Improved Paper Defects Denoising Method Based on Gray Associated with Neighborhood Characteristics

中文关键词:  均值滤波  中值滤波  灰色理论  邻域关联特性

Key Words:mean filtering  median filtering  gray theory  neighborhood characteristics

中文摘要:

      对传统的均值滤波、中值滤波的优缺点进行了分析，引入基于邻域关联特性的图像滤波去噪方法。该方法针对纸病图像噪声源多、样本有限等局限性，根据系统各因素间的内部联系和发展态势的相似程度来度量因素之间的关联程度，从而进行滤波。仿真结果表明，这种算法在高斯噪声和椒盐噪声滤波方面能有效地滤除噪声，同时还能较好地保护图像细节。在纸病检测测试中，该方法可以解决纸病准确去噪的问题。

Abstract:

      This paper proposed an improved method based on gray associated with neighborhood characteristics, and the advantages and disadvantages of the method were compared with the traditional filtering method, such as mean filtering and median filtering methods. According to the characteristics of paper defect images, such as the variety of noise sources and the limit of samples, this method was based on the internal relations between the various factors of the system and the similarity degree of development trend to measure the degree of association between these factors. The simulation results demonstrated that this theory could effectively filtering Gaussian noise and Salt and Pepper noise, at the same time well protected the image details, thus providing a viable image processing tool for diagnosing paper detection.