期刊文献+

基于视觉显著性的织物起球客观等级评价 被引量:2

Fabric Pilling Objective Evaluation Based on Visual Saliency
下载PDF
导出
摘要 传统的主观织物起球等级评价容易受检验人员经验、心理等因素影响,难以满足准确评价起球等级的需要。在分析视觉显著性机制的基础上,提出一种基于视觉显著性的织物起球等级客观评价新方法。首先利用小波多层静态分解以及各子图之间选择性的中央-周边操作,提高织物起球的近似、水平及垂直细节信息的显著度;在此基础上,通过差分子图融合形成织物起球显著图。然后,根据织物起球特征所确定的阈值分割出织物起球兴趣区。最后,从兴趣区提取织物起球特征,并通过BP(back propagation)人工神经网络进行起球疵点客观等级评价。试验结果表明,该方法能够有效地进行织物起球疵点客观等级评价,并且具有较强的抗噪声能力。 The subjective evaluation of fabric pilling grade is easily affected by the traditional inspection based on personnel experience, psychology and other factors, which is difficult to meet the accurateness of the pilling grade evaluation. With the analysis of visual saliency mechanism, a new objective evaluation method of fabric pilling grade was put forward. Firstly, the approximate, horizontal detail and vertical detail information saliency about fabric pilling were improved by using wavelet static multi- decomposition and central-surround operation of the selectivity among sub-images. Based on this, the fabric pilling saliency map was formed through the difference sub-images fusion. Then, the fabric pilling of interest region was determined by threshold according to the fabric pilling characteristic. Finally, extract pilling features from the interest region, and fabric pilling was evaluated objectively by the BP (back propagation) artificial neural network. The test results show that this method can make an objective evaluation for pilling grades effectively, which have a strong anti-noise capacity.
出处 《东华大学学报(自然科学版)》 CAS CSCD 北大核心 2018年第1期59-66,共8页 Journal of Donghua University(Natural Science)
基金 中国纺织工业联合会科技指导性资助项目(2016065) 河南省功能性纺织材料重点实验室资助项目
关键词 织物起球 视觉显著性 阈值分割 客观评价 fabric pilling visual saliency threshold segmentation objective evaluation
  • 相关文献

参考文献5

二级参考文献54

  • 1钟小勇,姚桂国,梁金祥,左保齐.局部阈值分割用于织物疵点检测[J].江苏纺织,2009(9):54-55. 被引量:4
  • 2袁端磊,路立平,宋寅卯.织物疵点自动检测技术的研究进展[J].郑州轻工业学院学报(自然科学版),2005,20(3):69-73. 被引量:10
  • 3Wang Xuewei Liu Songtao Zhou Xiaodong.New algorithm for infrared small target image enhancement based on wavelet transform and human visual properties[J].Journal of Systems Engineering and Electronics,2006,17(2):268-273. 被引量:1
  • 4HU M C. The Lnspection of Fabric Defects by Using Wavelet Transform[J]. Journal of Textile Institute,2000,91(3): 420-433.
  • 5HU M C, TSAI I S. Fabric Inspection Based on Best Wavelet Packet Bases[J]. Textile Res J,2000,70(8) : 662 - 670.
  • 6YOSHIO Shimizu. Expert System to Inspect Fabric Defects by Pattern Recognition[J]. Seni Gakkaishi,1990,46:460 - 469.
  • 7RIBOLZI S. Real-Time Fault Detection on Textiles Using OptoElectronic Processing[J]. Textile Res J,1993,63(2): 61 - 71.
  • 8TSAI I S, HUM C. Automatic Inspection of Fabric Defects Using an Artificial Neural Network Technique [ J ]. Textile Res J,1996,67(6): 401 - 405.
  • 9JASPER W J, POTLAPALLI H. Image Analysis of Mispicks in Woven Fabric[J]. Textile Res J,1995,65(11): 683 - 692.
  • 10NGAN H Y T, PANG G K H. Regularity analysis for patterned texture inspection [ J]. IEEE Transactions on Automation Science and Engineering, 2009, 6 ( 1 ) : 131 - 144.

共引文献26

同被引文献16

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部