Common surface temperature measurement techniques, when applied to monitoring the temperature of surfaces with complex morphology, suffer from reduced spatial resolution, which compromises the measurement accuracy of the system. To improve the spatial resolution of temperature measurement technology and maintain high temperature sensitivity, we designed a microscopic morphology thermometric LiDAR (MMTL) system based on the Scheimpflug principle, which realizes the real-time restoration of the 3D morphology and temperature of the surface of micro-structured objects. The 3D spatial resolution of the system is better than 3 mu m. The theoretical resolution of the self-designed reflective spectrometer can reach 0.9 nm, which improves the sensitivity and accuracy of the upconversion hybrid nanomaterials thermometry based on the intensity ratio. In the wide temperature range of 373.15-508.15 K, the highest relative temperature sensitivity can reach 2.07%/K, the optimal temperature resolution is 0.0131 K, and the error is less than 1 K. Finally, the temperature change trend of the mold surface under different heating voltages is accurately restored. The MMTL system can provide accurate temperature distribution data and hotspot location identification for scenarios such as optimizing thermal management design and real-time risk monitoring, and it has application potential in industrial manufacturing and for electronic products.