During the manufacturing and application of fiber-reinforced composites (FRP), issues such as impact damage and fatigue accumulation cause irreversible subtle damage to the internal structure. Acoustic emission (AE) technology, with its high precision and real-time property, has become an important means to monitor the damage evolution and failure mechanisms of FRP. The applications of acoustic emission technology in the damage characterization of FRP in recent years was reviewed. By conducting research on AE technical means such as parameter analysis, waveform analysis, pattern analysis, and deep-learning analysis, the results showed that parameter analysis and waveform analysis could complement each other in terms of signal characteristics during the detection process, achieving a qualitative description of damage behaviors such as the deformation and fracture of composite structures. Methods such as deep-learning analysis provided important theoretical support for the health monitoring and life prediction of fiber-reinforced composites. Overall, acoustic emission technology can monitor and evaluate the composite structures in operation in real-time. It has great development potential for maintaining the health of FRP materials and preventing sudden failures. In the future, it can be further combined with artificial intelligence technology to improve the accuracy and efficiency of damage identification.