Nondestructive Testing of Steel Wire Rope Based on Gagnetic Signal and Infrared Information

This paper designs a two-dimensional magnetic signal detection device under weak magnetic excitation, which solves the problem of large volume and single signal acquisition of traditional one-dimensional detection devices. To reduce the original noise, a noise reduction algorithm combining wavelet transform and improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is proposed. A super-resolution fusion algorithm is proposed to fuse two-dimensional magnetic signals to achieve image enhancement. Finally, the convolutional neural network is used to extract the features of the two types of images, and then the features are fused, and the support vector machine (SVM) is used to classify. Under the condition of zero broken wire error, compared with the subjectively extracted color features and texture features of the two types of images as the SVM input, this algorithm's recognition rate is increased by 37.26%.