基于稀疏分解的超声无损检测信号处理  

Ultrasonic nondestructive signals processing based on sparse decomposition

作　　者：陈保立[1] 陈宇[2] 张跃飞[1] 张志杰[1] 

机构地区：[1]中北大学信息与通信工程学院,山西太原030051 [2]中北大学机电工程学院,山西太原030051

出　　处：《电子技术应用》2012年第8期140-142,共3页Application of Electronic Technique

摘　　要：为了在复杂背景噪声情况下,对缺陷的大小和位置实现精确的识别,提出了一种基于Ga-bor原子库稀疏分解的信号处理方法 ,利用匹配追踪算法将信号在超完备Gabor原子库中进行稀疏表示。采用相干比阈值作为迭代终止条件,根据信号噪声水平自适应调整迭代次数。针对算法计算量大的缺点,引入遗传算法,大大提高了计算的效率。实验表明,该方法可以有效减小噪声的影响,具有计算效率高、稳定性好的特点。In order to improve the effect of recognition when background noise is strong and complex,a new method is proposed for ultrasonic signals processing. Based on sparse decomposition with Gabor dictionary, signal is equated with linear combination of several Gabor atoms in this method. Take the coherent ratio threshold as iteration termination condition, the iteration times vary with the noise level adaptively. Adaptive genetic algorithm was introduced to matching pursuit,which make computation speed improved greatly. Experiments show that this method is efficient and stable, and can effectively reduce the influence of the noise.

关 键 词：Gabor原子库 稀疏分解 超声信号处理 遗传算法 

分 类 号：TN911.72[电子电信—通信与信息系统]