Structured light-based 3D reconstruction serves as a fundamental component in the perception of the three-dimensional real world by intelligent systems， exhibiting significant potential across various industrial domains such as industrial inspection， intelligent manufacturing， biomedicine， robotic vision guidance， virtual reality， and embodied intelligence. This study first reviews prevalent fringe projection techniques， emphasizing both the established mainstream DLP fringe projection and the novel MEMS mirror-based fringe projection technologies. A comprehensive comparison is conducted regarding operational principles， performance metrics， and respective advantages and limitations， highlighting the promising developmental trajectory of MEMS mirror-based structured light technology. Subsequently， calibration challenges arising from the non-invertible camera characteristics inherent to MEMS structured light are addressed by summarizing several physically interpretable calibration models tailored for MEMS systems. Distinct from DLP structured light， MEMS structured light employs laser beam scanning and introduces unique error sources， which are systematically analyzed to inform future strategies for measurement accuracy enhancement. Finally， the applications of structured light are discussed， and future development trends of MEMS structured light technology are projected.