Unmanned aerial vehicles (UAVs) have become increasingly widespread in a variety of industries due to their versatility and efficiency in applications such as agriculture, surveillance, logistics, and construction. However, their rapid adoption has introduced challenges related to detection and classification, especially in the context of privacy, public safety, and national security. Conventional UAV detection methods, such as radar, thermal imaging, and acoustic systems, face limitations in accurately distinguishing between UAVs and other airborne objects. Additionally, these systems often fail to differentiate between UAVs constructed from different materials, such as carbon fiber-reinforced polymers (CFRP) and glass fiber-reinforced polymers (GFRP), which significantly affect the UAV's radar and thermal profiles. This paper presents a promising approach for UAV detection based on the material composition of their structures using hyperspectral imaging (HSI) and K-Means (K-M) clustering. Using the proposed approach, we found that CFRP can be detected at 700 nm. While GFRP can be detected at 530 nm. By applying the K-M clustering algorithm to the spectral data, we successfully classify these materials without prior knowledge of object types. The proposed method shows high effectiveness in accurately distinguishing between UAVs based on their material composition, offering improvements over traditional detection methods that rely on shape, size, or heat signatures. This research contributes a new dimension to UAV detection by focusing on material-specific classification, providing significant potential for applications in security and surveillance, where understanding the structural composition of a UAV is critical for effective identification and mitigation strategies.