论文部分内容阅读
超声检测技术在工业和科研领域得到广泛的应用。在恶劣的条件下,超声信号受到混响的干扰。传统的匹配滤波方法在这种情况下检测性能不理想。该文提出一种基于特征的检测方法,对接收信号进行分类再进行检测。该方法基于模式识别来区分是否存在回波。先利用Wigner-Ville分布和双谱提取信号的特征,然后进行主成分分析降低特征的维数,降维的特征向量送入有监督学习的分类器。实验表明,与传统的匹配滤波方法相比,该方法在-5dB的信混比时具有较好的检测性能。
Ultrasound detection technology in the field of industry and scientific research has been widely used. Under harsh conditions, the ultrasonic signal is disturbed by the reverberation. The traditional matching filtering method in this case detection performance is not satisfactory. This paper presents a feature-based detection method that classifies the received signal and then detects it. The method distinguishes echo based on pattern recognition. The Wigner-Ville distribution and bispectrum are firstly used to extract the features of the signal, then the principal component analysis (PCA) is used to reduce the dimensionality of the feature, and the dimensionality reduction feature vector is sent to the supervised learning classifier. Experimental results show that the proposed method has better detection performance than the traditional matched filter at -5dB SNR.