With the development of society, the increase in the number of construction projects leads to more and more pollution of the environment by construction and demolition waste, and how to effectively recycle and utilize this waste has become an important issue. Currently, the inefficient management of construction site waste leads to numerous challenges, creating an urgent need for advanced technologies capable of rapidly identifying waste locations. This study presents an advanced method for rapid searching of construction waste. We utilize UAV and computer vision technology to quickly identify and mark the location and type of construction waste, and provide technical support to improve the efficiency of on-site waste searching. It is proved by experimental results that the proposed method can realize high-precision construction waste recognition. The method and data proposed in this study can provide the basis for fully automated recycling of construction waste.