A high-pass cut-off frequency in filtering is critical to processing strong-motion records. The various processing procedures available nowadays are based on their own needs and are not universal. Regardless of the methods, a visual inspection of the filtered acceleration integration to displacement is required to determine if the selected filter passband is appropriate. A better method is to use a traversal search combined with visual inspection to determine the cut-off frequency, which is the traditional method. However, this method is inefficient and unsuitable for processing massive strong-motion records. In this study, convolutional neural networks (CNNs) were used to replace visual inspection to achieve the automatic judgment of the reasonableness of the filtered displacement time series. This paper chose the pre-trained deep neural network (DNN) models VGG19, ResNet50, InceptionV3, and InceptionResNetV2 for transfer learning, which are only trained in the fully connected layer or in all network layers. The effect of adding probability constraints on the results when predicting categories was analyzed as well. The results obtained through the VGG19 model, in which all network layers are trained and probability constraints are added to the prediction, have the lowest errors compared to the other models. The coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are 0.82, 0.038, 0.026, and 2.99%, respectively.

In structural response array observation, the vibration response of a structure during an earthquake or from the natural environment is recorded and stored using high-sensitivity strong motion seismographs, and the dynamic characteristics of the structure are analyzed and determined using random signal data processing technology. Due to the use of field test data for analysis, this may be the most accurate and effective way to obtain the actual characteristics of the structure, which can be further used to verify the accuracy of theoretical analysis, experimental results, and numerical simulations. Therefore, this technique plays an important role in earthquake prevention and disaster reduction, with the application of strong motion observation data. In this paper, field vibration tests were performed on the highest reinforced masonry structure in China. With the test environmental vibration data, the natural vibration frequency values and mode shapes of the structure were identified using the peak picking method. A numerical modal analysis was then performed to verify the accuracy of the field test results. In addition, the structural response records obtained during an earthquakes in Songyuan were also used to identify the natural vibration frequency of the structure and the changes in the natural vibration frequency before, during, and after the earthquake. The results showed that the structure was not damaged during the earthquake and remained in an elastic state.