We propose a hybrid technique named Deep Learning enhanced Optical and Pixel-based video Frame Detection that can detect forgery in digital video sequences. This algorithm uses optical flow and pixel-wise information along with DL-based approaches for frame insertions, deletions, or replication detection. It detects forgery detection using preprocessing, optical flow analysis, adaptive pixel-wise consistency checks, deep learning-based analysis, and decision fusion with explainable AI. Experimental results demonstrate that the proposed algorithm yields remarkably superior performance compared to existing approaches in accuracy, precision, recall, and F1-score, proving its robustness in detecting diverse video forgery methods. Besides, the algorithm also gives a visual explanation of tampering detections via Grad-CAM, which makes the decision transparent and explainable. Based on the high accuracy, future work will be done on reducing both the polynomial time complexity and the error rates to process large-scale video datasets.