In this study, we aimed to enhance seismic volumes from the Opunake field using Artificial Intelligence (AI) by building and comparing two different Convolutional Neural Network (CNN) models. The seismic volumes utilised in this investigation were impacted by several imaging problems, including noise and resolution deterioration, impeding the subsurface structure�s interpretation. The traditional CNN model was trained using a dataset of low-quality seismic volumes and their corresponding high-quality versions. The generative adversial network (GAN) model, on the other hand, was trained using a dataset of high-quality seismic volumes. Both models were then used to enhance the seismic test volumes. The performance of the models was evaluated using three commonly used image quality metrics: PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), and SNR (Signal-to-Noise Ratio). The evaluation results showed that both models could enhance the seismic volumes to a certain extent. However, the GAN model performed better in terms of all three metrics. The GAN model achieved a higher PSNR, SSIM, and SNR value than the conventional CNN model, indicating that the enhanced seismic volumes were of higher quality and contained less noise. In conclusion, the study demonstrated the potential of AI for enhancing seismic volumes affected by imaging issues. The GAN model was a better choice for enhancing seismic volumes from the Opunake field as it showed superior performance in terms of PSNR, SSIM, and SNR. This study highlights the potential of AI as a powerful tool for addressing imaging challenges in the field of exploration geophysics. © 2024 Daniel Asante Otchere.