Detection of Internal Defects Based on Empirical Mode Decomposition and Wavelets Denoising

The internal defects of industrial components such as magnetic tiles seriously affect their performance. With the development of intelligent manufacturing technology, industrial manufacturing enterprises need an automatic method to efficiently and accurately detect the internal defects of magnetic tiles. In this paper, a signal pre-processing algorithm based on Empirical Mode Decomposition (EMD) and wavelets denoising is proposed for echo signals for defect detection. Then the variance curve and the adaptive processing method are used to locate the defects accurately. The experimental results show that the algorithm proposed in this paper can been successfully used in defect specimen with different transducer frequency, different defect size and different defect depth. Compared with the original B-scan image, and the internal defects of the specimen could be detected more prominently in enhanced B-scan image, and the accuracy of the defect depth could reach 98.76%, which is better than existing state of the art. Thus, the proposed method has been proved to be effective for optimizing ultrasonic B-mode scanning and accurately locating defects inside magnetic tiles.