基于小波变换的织物疵点检测

本文基于小波变换原理进行织物疵点检测。使用DH-SV 1410FM数字相机采集平纹坯布图样,应用中值滤波和Laplacian算子对图像做去噪和锐化处理。采用Mallat算法对预处理后的图像进行小波分解,分别提取经纬子图的能量、方差、熵、逆差距和极差等特征值,并对提取的特征值进行归一化处理。实验结果表明,疵点使织物图像在经纬方向上产生了不规则的纹理走向,不同类疵点图像的特征值归一化曲线在疵点处有明显的变化。该研究对判断织物是否有疵点的存在、判断疵点的大致走向和确定疵点的位置具有一定的参考价值。     

基于织物自适应正交小波的疵点检测

应用织物自适应正交小波对织物疵点的检测和识别进行了分析,首先介绍了织物图像的小波分解算法和紧支撑正交小波,在此基础上提出了织物自适应小波的构造,由自适应小波对织物图像分解,然后对分解后的纬向和经向子图像提取特征,由特征什检测和识别疵点。实验证明了该方法对素色织物的常见疵点具有快速、准确的检测效果。     

基于计算机视觉的织物疵点自动检测

用小波变换结合BP神经网络对织物疵点进行检测和分类.采用平纹坯布作为试样,对油渍、缺经、缺纬疵点进行识别.摄相机摄取256×256图像,将拾取图像进行小波分解,然后把小波分解后的图像灰度值作为特征参数输入到BP神经网络进行检测识别.实验结果表明,用这种方法识别织物疵点,识别率可达到98%.     

基于小波变换和最大熵的织物疵点检测

首先选取最优小波基对织物疵点图像进行分解,并对分解后的水平及垂直高频系数进行灰度值归一化,然后分别求出水平和垂直细节图像的最大熵及平均熵,通过比较判断出疵点类型,最后对平均熵大的细节图像进行最大熵原理的分割,得到最终的疵点检测结果。仿真实验表明该方法对常见纬向和径向类织物疵点的检测是有效的。     

用于疵点检测的织物自适应正交小波的实现

应用随机算法并给以一定的约束条件，可以有效地由满足正交归一条件的滤波器中，优化获得与织物纹理相匹配的滤波器。应用这种织物自适应滤波器对织物灰度图像进行小波分解，就可使分解后的子图像能很好地分别包含织物的纬向和经向纹理信息，从而可有效地应用于织物疵点的自动检测。     

基于灰度共生矩阵和稳健马氏距离的织物横档类疵点检测

在分析横档类疵点纹理特点的基础上,提取织物纹理图像的灰度共生矩阵单特征值--对比度.利用最小中值平方估计的快速算法,获得正常织物纹理训练样本的稳健马氏距离,并应用契比晓夫不等式确定在一定置信度条件下判断待检织物为疵点的马氏距离的阈值.对8种不同纹理结构、织物密度和纱线线密度的织物进行了横档类疵点的检测,在90%置信度下,可检出90%以上的横档类疵点,误检率为3.28%,检测效果较好.     

基于多尺度下LBP和PSO的织物瑕疵检测方法研究

针对丝织品疵点纹理与正常情况下织物纹理特征的不同,提出一种多尺度下LBP和PSO结合的织物疵点检测算法.首先对织物图像进行非线性的金字塔多尺度分解,再对各尺度图像提取对光照和旋转具有不变性的LBP特征,在此基础上,利用PSO的快速收敛特点计算各分块的LBP直方图的差异确定出疵点位置.实验结果表明,提出的方法具有较好的鲁棒性.     

基于最优Gabor滤波器的织物疵点检测

织物的疵点检测在纺织工业生产的质量控制中起着至关重要的作用.针对织物疵点,论文提出一种基于最优Gabor滤波器的织物疵点检测算法.首先,针对4尺度4方向的16个Gabor滤波器,借助的方向模板对疵点图像滤波后的信息量确定2个最优方向;其次,综合信噪比和损失评价函数在2个方向的8个滤波器中各自选取1个最优滤波器;然后,对滤波后的疵点图像做融合和二值化,实现了疵点检测.从实验结果看,该算法能够有效地检测出织物疵点.     

基于熵和变异度的织物疵点图像分割方法

针对现有织物疵点图像分割方法对光照不均匀敏感的问题,提出了一种基于局部熵和变异度的织物疵点图像分割方法。首先对织物图像进行局部熵和变异度计算,提取疵点的类边缘和区域信息;然后基于人工神经网络脉冲耦合（PCNN ）的区域生长法分割织物疵点图像。通过对T ILDA数据库中的疵点图像和基于线阵CCD在线检测的织物疵点图像进行测试,并与已有的相关方法进行对比实验和评价。结果表明,该方法不仅能有效地抑制光照不均匀和复杂背景干扰的影响,而且分割质量有了明显改进。     

一种基于纹理方向性分析的织物疵点检测方法

根据织物图像纹理自身特点,从图像纹理的方向性入手,提出了基于纹理方向性分析的织物疵点检测方法.通过对不同方向性织物疵点图像检测实验,证明本文提出方法对具有方向性的斜纹类织物疵点检测具有较好的有效性和可靠性.     

基于自适应正交小波的机织物密度自动检测的研究

为了将计算机视觉应用于织物经纬密自动检测，提出了建立机织物自适应小波分解滤波器并研究了织物密度自动检测的方法。应用织物自适应滤波器对织物灰度图像进行小波分解，就可使分解后的子图像能很好地分别包含织物的纬向和经向纹理信息，从而可将织物子图像二值化后，根据其黑白交替变化规律，计算出织物的经纬纱密度。实验结果证明自动检测和人工测量相比，误差在3％以内。     

RGB and LBP-texture deep nonlinearly fusion features for fabric retrieval

Fabric retrieval is very challenging since problems like viewpoint variations, illumination changes, blots, and poor image qualities are usually encountered in fabric images. In this work, a novel deep feature nonlinear fusion network(DFNFN) is proposed to nonlinearly fuse features learned from RGB and texture images for improving fabric retrieval. Texture images are obtained by using local binary pattern texture(LBP-Texture) features to describe RGB fabric images. The DFNFN firstly applies two feature learning branches to deal with RGB images and the corresponding LBP-Texture images simultaneously. Each branch contains the same convolutional neural network(CNN) architecture but independently learning parameters. Then, a nonlinear fusion module(NFM) is designed to concatenate the features produced by the two branches and nonlinearly fuse the concatenated features via a convolutional layer followed with a rectified linear unit(ReLU). The NFM is flexible since it can be embedded in different depths of the DFNFN to find the best fusion position. Consequently, DFNFN can optimally fuse features learned from RGB and LBP-Texture images to boost the retrieval accuracy. Extensive experiments on the Fabric 1.0 dataset show that the proposed method is superior to many state-of-the-art methods.     

基于光照投影的起球织物图像的采集、预处理和毛球分割

采用一种新的图像分析方法进行织物起球等级的分析。先获取一系列起球织物切面光照投影图，利用LOG边缘检测算子和线性插值提取每帧图像中织物的轮廓线高度数据，并拼接成一幅反映织物三维轮廓高度的距离图像，最后对该距离图像进行预处理和基于自适应阈值法的毛球分割。     

基于光度立体图像三维重建的起皱织物表面形态

采用不同光照条件下拍摄的多幅图像,利用光度立体技术对织物表面进行3D重建.首先根据物体表面的照度方程,引入广义逆的概念求解物体表面方向进而确定表面梯度,再运用线积分计算表面高度,结合变分和有限差分思想对所得拟合表面进行进一步的迭代和修正,获得最佳重建表面.将该算法运用到AATCC织物平整度模板图像的三维重建,可获取三维深度信息,并结合4个特征值表征织物起皱程度.     

Fabric Defect Detection Using Adaptive Wavelet Transform

A method of woven fabric defect detection using the wavelet transform adaptive to the fabric has been developed. With reference to the orthogonality constrains of Daubechies wavelet, by taking the minimization of the energy or the gray level of the pixels hi the output sub-images as the additional conditions and using the random algorithm method, two sets of wavelet filters adapted to the fabric texture were formed. The original images of normal fabric texture and the fabric texture with defects were decomposed into horizontal and vertical sub-images by using these filters and the feature indices of these sub-images were also extracted. By comparing the feature indices of the normal texture with that of the defective texture, the fabric defects can be successfully detected and located.     

Innovative Principle and Method for Digital Jacquard Fabric Designing

Digital jacquard fabric has its design concept and method directly borrowed from computer images and color modes, which enabled creation of jacquard fabric design that is far beyond what freehand patterns can express. In this paper, the principles of digital jacquard fabric design were classified into two parts: colorless mode and colorful mode, and an innovative layered combination design method has been suggested contenting with this new design concept, by which digital jacquard fabric can be designed from colorless single-layer structure to colorful compound structure. As a result, designed colorless and colorful jacquard fabrics are capable of expressing picturesque and photo-realistic effects with a mega level color number on face of fabric. It is envisaged that the results of the study are of tremendous benefit to the creation of new jacquard fabric ith an inimitable digital effect and this creation pose no problem in mass production.     

Fabric Defect Detection Using GMRF Model

It has been testified that the Gauss Markov random field model is most suitable for the characterization of fabric texture among a variety of available models because of its approximately constant character and the normality of the gray-level distribution found with typical fabric images. However, the general Gauss-Markov random field(GMRF) method for fabric defect detection is not always ideal in practice since in some cases, the estimated model parameters make the Markov error covariance not positively definite, which may render the method to fail thoroughly. In this paper, the use of the GMRF model for defect detection of fabric is discussed and an approach to this problem is proposed. Some detailed texture may be overlooked in this way, but good detection results can still be expected as far as fabric defect detection is concerned.     

Assessing Fabrics Wrinkling by Image Analysis

In this paper, the features of fabric wrinkle structure are analyzed, by image processing technique. The wrinkle images are processed and character parameters are extracted. The wrinkle side surface area ratio of image gray level intensity, the standard deviation of total pixels' gray level intensity and the standard deviation of wrinkle block area are adopted to evaluate fabric wrinkling. The results show that these three character parameters can be served for ranking the wrinkling. It provides a feasible objective method for fabrics wrinkling degree assessing.     

快速富里叶变换在织物结构分析中的应用

在计算机图像处理和横式识别领域，快速富氏变换（ＦＦＴ）技术起着非常重要的作用．织物（此文只论及机织物）的结构是由相同单元规则地排列而成的，呈现很强的周期性．这使得ＦＦＴ技术在分析织物结构特征中特别有效．本文讨论如何应用ＦＦＴ去识别织物组织、纱线密度、纬纱倾斜度等其它结构参数．     

图像处理用于织物疵点自动检测的研究进展

织物疵点自动检测、代替传统的人工检测方法是纺织品质量控制和管理的重要环节，同时又是纺织自动化的重要标志之一，基于图像处理的织物疵点检测，主要是通过图像分析，从图像中发现与正常纹理不同的部分，这方面的研究一直在持续，新的方法不断出现，但较为成熟的检测系统较少，开发适合我国纺织业的织物疵点自动检测系统，具有深远意义。