摘要
为了提高布匹瑕疵图像的滤波效果,针对检测过程中图像易受到多种混合噪声污染的特点,提出一种基于多尺度迭代思想以改进AMSS算子。将偏微分方程中的AMSS方程以有限差分的形式离散化,依据量化尺度参数t和尺度步长Δt,采用小尺度步长对污染图像进行多次迭代滤波。针对改进AMSS算子的滤波结果,介绍了一种采用水平线算法进行效果分析的新方法。实验结果表明,和其他算法相比,该滤波算法在滤除受未知混合噪声污染图像干扰的同时,能够更好地保留边缘信息,从而达到更理想的平滑效果。
Focused on fabric flaw detection, an improved AMSS operator based on multi-scale and iterative method was presented. The AMSS equation was discreted with finite difference, and according to quantitative scale parameter t and scale step length At, the noise in the object image was iteratively filtered by small scale step length, then aimed at the filter result of AMSS operator, a new evaluation method on level line was introduced. The experimental result illustrates that this improved operator can efficiently keep edge information, and the performance of tilting noise is more effective than other methods for image mixed with multi-oise.
出处
《传感技术学报》
EI
CAS
CSCD
北大核心
2006年第3期709-712,753,共5页
Chinese Journal of Sensors and Actuators
基金
浙江省重点科研资助项目(2006C21037)
