Epilepsy, a chronic neurological disorder that impacts countless individuals globally, is characterized by recurring seizures. Accurate and timely classification of different seizure types is crucial for effective medical intervention and patient care. By employing a deep stacking approach, this paper aims to enhance the accuracy and precision of seizure classification, which can significantly improve treatment strategies and outcomes for individuals with epilepsy. The paper aims to enhance the effectiveness of EEG signals, the primary diagnostic tool for epilepsy. In addition to analyzing EEG data, this paper incorporates images derived from these signals, providing a multi-modal perspective that captures the diverse aspects of the underlying condition.