Digital images are becoming a tangible information source as imaging technology advances rapidly. Due to this, it is often a complicated task to maintain the authenticity of the digital images as there are several powerful and sophisticated tools that can tamper these images efficiently, leaving no trace to the end-users. Copy-move forgery detection (CMFD), which detects the tampered area, effectively recognizes the modified image region. In this study, we proposed a method based on the state-of-the-art deep learning architecture of MobileNetV2. It involves the classification of two classes, original and forged, resulting from MobileNetV2 after making necessary adjustments. In this regard, the study employs MICCF2000 and MICC-F220, which contain both original and tampered images. The performance of the study is evaluated through recall, precision, accuracy, F1-score, and False Positive Rate (FPR), which shows that the technique applied in this study is applicable for future use as it has more accuracy and less computational time.