Enhanced Classification of Gravitational Wave Images using CNN Model

The discovery of gravitational waves resulting from the merger of binary black holes revolutionized physics and astronomy. Amid signal complexity and noise capture, data performance is critical. This work presents a convolutional neural network (CNN) algorithm using real-time data from the "Gravity Spy" dataset at https://www.kaggle.com/datasets/tentotheminus9/gravity-spy-gravitational-waves. A combination of CNN model and Fourier transform accuracy of gravitational wave signal expression. It is worth noting that frequency domain signal analysis is better than time domain signal analysis in capturing signal characteristics. Experimental results demonstrate the effectiveness of CNN in improving signal recognition by achieving an accuracy of 94.63%. This research helps advance the exploration of the void by proposing a deeper study, thus promoting a deeper understanding of astrophysical phenomena.