Abstract: A deep learning-based MLCC stacked alignment inspection system includes an integrated defect detection unit configured to detect core areas requiring inspection of image data in which a stacked structure is photographed from a semiconductor MLCC chip by using at least one deep learning-based core area detection model, perform segmentation in the detected core areas, determine whether a defect exists according to a standard margin percentage range, and enable defect detection by generating normal and/or defective data based on the determination result, a result analysis unit configured to perform visualization for respective results of the core area detection, segmentation, and defect detection of the integrated defect detection unit, and provide stepwise analysis data for the visualized respective results so as to determine whether to modify corresponding data, and a data storage configured to store the normal and/or defective data, and stepwise analysis data.