Abstract: ‍ ‍In recent years， the unmanned aerial vehicle （UAV） industry and its applications have witnessed rapid development due to their irreplaceable roles in various sectors. However， incidents such as “black flights” of UAVs and those involving the carriage of dangerous goods occur frequently. Therefore， the detection and recognition of UAVs have become imperative. Traditional UAV detection and identification methods， relying mainly on rule-based and classical computer vision approaches， have become increasingly inadequate. and unable to meet the demands of contemporary security needs. It is within this context that the rapid advancement of deep learning technology has emerged as a game-changer， offering a potent and precise solution for UAV detection and recognition. Deep learning models， characterized by their remarkable ability to autonomously learn intricate features and extract high-level representations from vast and complex datasets， have demonstrated exceptional efficacy in the realm of UAV detection and identification. These models have not only substantially elevated levels of accuracy but have also showcased adaptability to the vast spectrum of complex environmental conditions and UAV types. The incorporation of deep learning techniques into UAV security systems has ushered in a new era of precision and efficiency. This comprehensive review paper seeks to provide a detailed exposition of the latest advancements in UAV detection and recognition technology grounded in deep learning principles. The paper explores a spectrum of modalities， including UAV visual detection and identification， UAV audio detection and identification， UAV radar detection and identification， and UAV radio frequency detection and identification. Each of these modalities represents a unique dimension in enhancing the overall effectiveness of UAV security systems. In conclusion， this paper concludes by conducting a meticulous analysis of the current issues pervading the field of UAV detection and recognition. Additionally， it offers a forward-looking perspective on the future research directions that hold the potential to further fortify the security and efficiency of UAV applications in an ever-changing technological landscape.