The identification of fraudulent stamps is an important problem in various fields, such as finance, law enforcement, and customs. In this paper, we propose a reverse engineering analysis of stamps manufactured by a CNC engraving machine, which can provide important information for identifying the production conditions of stamps. We first analyzed the images of a large number of CNC-engraved stamps to extract key features such as stroke width and spacing, which are closely related to the manufacturing process. We then developed a database of known stamp production conditions, which enabled us to identify the production parameters of unknown stamps through a process of feature matching. Our results showed that even when two stamps are produced by the same engraving machine and appear similar in most aspects, small differences in production conditions can be identified and quantified through our feature-based analysis. This makes it possible to distinguish between two stamps with greater reliability, which can be useful in forensic analysis and legal proceedings. Overall, our research demonstrates that reverse engineering analysis of CNC-engraved stamps can be an effective tool for identifying production conditions and improving the reliability of stamp authentication.