纺织学报 ›› 2018, Vol. 39 ›› Issue (05): 125-131.doi: 10.13475/j.fzxb.20170704007

• 管理与信息化 • 上一篇    下一篇

改进频率调谐显著算法在疵点图像分割中的应用

  

  • 收稿日期:2017-07-10 修回日期:2017-12-29 出版日期:2018-05-15 发布日期:2018-05-10

Segmentation of fabric defect images based on improved frequency-tuned salient algorithm

  • Received:2017-07-10 Revised:2017-12-29 Online:2018-05-15 Published:2018-05-10

摘要:

为提高织物疵点分割精度,提出了一种用于织物疵点图像分割的改进频率调谐显著(FT)算法。首先,利用织物疵点和背景区域透光率的不同,将光源和相机分别置于织物两侧来获取图像,提高疵点区域对比度;其次,应用非局部均值滤波器(NLM)替代FT 算法中的高斯滤波器,增强对背景纹理的平滑和降噪能力;研究发现NLM滤波器中滤波参数对疵点分割精度影响较大,提出了基于平均最大类间方差的参数优化方法;然后,将改进FT 算法应用于疵点图像预处理,进一步提高疵点对比度;最后,使用最大类间方差法对疵点显著图进行分割。对2 种不同织物疵点图像的分割实验结果表明,使用改进FT 算法对粗经、竹节、结头、断纬、油污和破洞等常见疵点图像进行预处理,可显著提高疵点分割精度。

关键词: 织物疵点, 非局部均值滤波, 频率调谐显著算法, 图像分割

Abstract:

In order to improve the precision of fabric defects segmentation, an improved frequency-tuned salient (FT) algorithm is proposed for the preprocessing of fabric image. Firstly, the light source and camera are placed on both sides of the fabric to obtain the image, and the contrast ratio of defect area was strengthened by the difference of transmittance between normal area and defect area. Secondly, the non-local mean filter (NLM) was used instead of the Gauss filter in the FT method to enhance the cap ability of texture smoothing and denoising; and it is found that the NLM filter parameter has great influence on the accuracy of image segmentation. A method of parameter optimization using the average of inter-class maximum variance was proposed. Then, the improved FT algorithm was applied to the prepocessing of images to strengthen the contrast ratio of fabric defect area. Finally, OTSU algorithm was used to segment salient image of fabric defect. The experiments of image segmentation were carried out for two different fabric. The experimental result shows that the segmentation precision of fabric defects, including slab yarn, knot, broken warp, oil stain, hole and so on, can significantly increased with the improved FT algorithm.

Key words: fabric defect, non-local mean filter, frequency-tuned salient algorithm, image segmentation

[1] 陈晓春 张彦彬 彭虎 陈翔 褚乃清. 重叠纤类图像的凹点匹配和分割算法[J]. 纺织学报, 2017, 38(11): 143-149.
[2] 张成梁 李蕾 董全成 葛荣雨. 应用区域颜色分割的机采棉杂质检测方法[J]. 纺织学报, 2017, 38(07): 135-141.
[3] 尉苗苗 李岳阳 蒋高明 丛洪莲. 应用最优Gabor滤波器的经编织物疵点检测[J]. 纺织学报, 2016, 37(11): 48-54.
[4] 石美红 张正 郭仙草 陈永当. 基于显著纹理特征的织物疵点检测方法[J]. 纺织学报, 2016, 37(10): 42-049.
[5] 厉征鑫 周建 潘如如 刘建立 高卫东. 应用单演小波分析的织物疵点检测[J]. 纺织学报, 2016, 37(09): 59-64.
[6] 曹丽 胡旭东. 基于多特征融合的织物印花图像分割[J]. 纺织学报, 2016, 37(08): 149-153.
[7] 景军锋 范晓婷 李鹏飞 张蕾 张宏伟. 应用Gaussian回代交替方向图像分解算法的色织物疵点检测[J]. 纺织学报, 2016, 37(06): 136-141.
[8] 周慧 张华熊 胡洁 康锋. 基于平滑滤波和分水岭算法的重组织织物图像分割[J]. 纺织学报, 2015, 36(08): 38-42.
[9] 景军锋 李江南 李鹏飞. 改进型奇异值分解在织物疵点检测上的应用[J]. 纺织学报, 2014, 35(6): 62-0.
[10] 管声启 高照元 吴宁 徐帅华. 基于视觉显著性的平纹织物疵点检测[J]. 纺织学报, 2014, 35(4): 56-0.
[11] 李文羽 程隆棣. 基于机器视觉和图像处理的织物疵点检测研究新进展[J]. 纺织学报, 2014, 35(3): 158-0.
[12] 卢雨正 高卫东. 基于图像分割的拼色纺织品分色算法[J]. 纺织学报, 2012, 33(9): 55-60.
[13] 景军锋, 孟泰, 李鹏飞. 基于模糊C均值聚类的纺织品印花图像分割[J]. 纺织学报, 2012, 33(6): 97-100.
[14] 张扬 蒋高明 姚君洲 童有成. 基于MRF层次模型的贾卡经编针织物图像分割技术[J]. 纺织学报, 2012, 33(12): 102-106.
[15] 李鹏飞;王刚;景军锋;焦珂. 基于JSEG算法的纺织品印花图像分割[J]. 纺织学报, 2010, 31(5): 137-140.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!