摘要
针对服装视错图案客观性评价欠缺的问题,提出了一种利用灰度共生矩阵、小波分析和BP神经网络来客观评价视错图案感官舒适度的方法。提取10款视错裙图像灰度共生矩阵的对比度、相关性、熵值、均匀度和能量值5个特征参数,以及Haar小波分解五层时水平、垂直和对角线方向上的维度特征参数。最后选取这两类纹理特征值中与主观评价结果显著相关的特征参数,输入到BP神经网络中进行迭代训练和预测评估。结果显示:该算法的识别正确率高达100%,预测效果较好,可以利用灰度共生矩阵、小波分析和BP神经网络相结合的方法,进行视错图案感官舒适度的客观评价。
Aiming at the lack of objective evaluation of the apparel with optical illusion pattern,a method to objectively evaluate the sensory comfort of optical illusion patterns was proposed by using Gray-level Co-occurrence Matrix(GLCM),wavelet analysis and Back Propagation(BP)neural network.Five characteristic parameters of the GLCM(including contrast,correlation,entropy,uniformity and energy value)were extracted from 10 skirts images with optical illusion.Besides,the dimensional characteristic parameters were extracted in horizontal,vertical and diagonal directions when Haar wavelet decomposed five layers.Finally,the characteristic parameters that were significantly related to the subjective evaluation results were selected from the two texture feature values.Then,they were input into the BP neural network for iteration training and forecast evaluation.The results show that the proposed algorithm has a correct recognition rate of 100%,with good forecast effect.Thus,gray-level co-occurrence matrix,wavelet analysis and BP neural network can be combined for objective evaluation of the sensory comfort of optical illusion patterns.
作者
王佳宁
刘成霞
WANG Jianing;LIU Chengxia(School of Fashion Design&Engineering,Zhejiang Sci-Tech University,Hangzhou 310018,China;Zhejiang Province Engineering Laboratory of Clothing Digital Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)
出处
《丝绸》
CAS
CSCD
北大核心
2020年第6期1-6,共6页
Journal of Silk
关键词
视错图案
感官舒适度
灰度共生矩阵
小波分析
BP神经网络
optical illusion pattern
sensory comfort
Gray-level Co-occurrence Matrix
wavelet analysis
BP neural network
