摘要
为提高混合特征畸变织物疵点的检测速度,解决由疵点导致的织物质量下降的问题,提出一种基于自适应离散小波变换方法识别混合特征畸变织物疵点。首先确定所设计小波的优化目标,然后采用二通道方法准确重建正交滤波器的结构实现方式,推导出目标函数,并通过目标函数选择具体的优化方法,建立起优化目标和正交滤波器系数间的函数关系;最后采用构造出的自适应正交小波对3种类型的混合特征畸变疵点进行识别,结果表明:自适应正交小波可在较小尺度下将织物进行小波分解,并获得良好的疵点边缘增强效果,对混合特征畸变织物疵点的识别准确率达到100%,从而验证了该方法的可行性。
In order to increase the detection speed of mixture feature aberrance fabric defect and solve the problem of fabric quality deterioration caused by defects,it proposes a method based on self-adaptive discrete wavelet transform to identify mixture feature aberrance fabric defect.Firstly,the optimizing object of the designing wavelet should be determined.Then,the two-passage method can be adopted to reconstruct the architecture of quadrature mirror filter in accuracy,and object function is derived.The specific optimizing method is chosen according to the object function and the function relationship between optimized object and quadrature mirror filter coefficient is set up.Finally,the constructed self-adaptive orthogonal wavelet can be applied to recognize three kinds of mixture feature aberrance defects.The result shows that self-adaptive orthogonal wavelet can process fabric wavelet decomposition in a smaller scale condition,and obtain good defect sedge enhancement effect.The accuracy rate can reach 100% for the mixture feature aberrance fabric defect,which verifies the feasibility of the method.
出处
《纺织学报》
EI
CAS
CSCD
北大核心
2013年第1期133-137,共5页
Journal of Textile Research
基金
浙江省自然科学基金资助项目(Y1110023)
关键词
自适应正交小波
混合特征畸变疵点
特征提取
边缘增强
模式识别
self-adaptive orthogonal wavelet
mixture feature aberrance defect
feature extraction
edge enhancement
pattern recognition
