摘要
针对蒙古族传统家具纹样模糊不清、边缘失真等问题,提出了基于加权变换的图像增强方法。首先将蒙古族家具纹样分解成RGB分量,然后利用提升小波变换、平稳小波变换、插值算法和逆提升小波变换获得高分辨率纹样,最后利用加权变换函数对直方图进行修改,对贡献最小的直方图进行滤波,得到高分辨率和对比度增强的纹样。实验结果表明,该方法的评价指标峰值信噪比PSNR和结构相似性SSIM比传统直方图均衡化和双三次插值法的均有提升,有效地增强了蒙古族家具纹样。所提方法对传统蒙古族家具纹样的修复和保护研究具有重要价值。
In order to solve the problems of fuzzy patterns and edge distortion of Mongolian traditional furniture,this paper proposes an image enhancement method based on weighted transformation.Firstly,Mongolian furniture patterns are decomposed into RGB components,and lifting wavelet transform,stationary wavelet transform,interpolation algorithm and inverse lifting wavelet transform are adopted to obtain the high-resolution patterns.Finally,the weighted transformation function is used to modify the histogram,and the histogram with the smallest contribution is filtered to obtain the high-resolution and contrast-enhanced patterns.This method is compared with the traditional histogram equalization and bicubic interpolation method.The experimental results show that the proposed method has better PSNR(Peak Signal to Noise Ratio)and SSIM(Structure Similarity Index)than the tradi-tional histogram equalization and bicubic interpolation methods,and can effectively enhance Mongolian furniture patterns.The proposed method is of great value to the repair and protection of traditional Mongolian furniture patterns.
作者
董霙达
张成涛
多化琼
杜豫怡
DONG Ying-da;ZHANG Cheng-tao;DUO Hua-qiong;DU Yu-yi(College of Material Science and Art Design,Inner Mongolia Agricultural University,Hohhot 010018,China)
出处
《计算机工程与科学》
CSCD
北大核心
2021年第6期1047-1051,共5页
Computer Engineering & Science
基金
国家自然科学基金(31760185)。
关键词
蒙古族家具纹样
小波变换
插值算法
加权变换
图像增强
Mongolian furniture pattern
wavelet transform
interpolation algorithm
weighted transform
image enhancement
