Army vehicle's data security plays a crucial role in different fields like Army operations, the Navy, and the Air Force. Thus, an expert system for the identification and privacy preservation of this data in imagery is important. Locating it manually can be difficult, but the advent of cryptographic hash functions, deep learning technology, and image processing has made it effortless. The existing technique for the proposed application is CNN (Convolutional Neural Networks). Still, the limitation is that the model struggles with extracting features from different scales, and dimensions and lacks privacy preservation. This research study provides a novel solution to overcome the limitations by proposing YOLOv8 with additional layers added to YOLOv7. The proposed system can be used to detect and classify 18 types of army vehicles in a designated area, such as war field, air defense, and navy, and can also be used to detect and classify them in real-time for security and safety purposes.