Study on Objective Evaluation of Seam Pucker Based on Wavelet Probabilistic Neural Network

摘要：

A new method to objectively evaluate seam pucker is brought out in this paper. Firstly, AATCC 88B seam pucker standard pictures are taken by digital camera. After wavelet transform of images, the six parameters that are standard deviation of horizontal, vertical and diagonal detail coefficients on 5th dimension, horizontal detail coefficients and histogram and image entropy are extracted, on 4th are extracted. Then, objective evaluation model of seam pucker based on probabilistic neural network is constructed and its prediction accuracy is more than 90％ by test This prediction model can be used to evaluate seam pucker grades of unknown samples, so that to overcome ambiguity and uncertainty of subjective evaluation.

关键词： seam pucker wavelet transform probabilistic neural network objective evaluation

作者: Li Yanmei Qiu Xiaokun Jiang Zhenzhen

作者单位: Fashion College Shanghai University of Engineering Science, Shanghai China Design College Fudan Shanghai Institute of Vision Art Shanghai China Fashion College Shanghai University of Engineering Science,Shanghai China

会议类型: 国际会议

会议名称: 2011 Seventh International Conference on Natural Computation(第七届自然计算国际会议 ICNC 2011)

会议地点: 上海

会议语种:英文

页码: 270-273

在线出版日期: 2011-07-26（万方平台首次上网日期，不代表论文的发表时间）