Western Altun in Xinjiang is an important area, where lithium (Li)-bearing pegmatites have been found in recent years. However, the complex terrain and harsh environment of western Altun exacerbates in prospecting for Li-bearing pegmatites. Therefore, remote-sensing techniques can be an effective means for prospecting Li-bearing pegmatites. In this study, the fault information and lithologyical information in the region were obtained using the median-resolution remote-sensing image Landsat-8, the radar image Sentinel-1 and hyperspectral data GF-5. Using Landsat-8 data, the hydroxyl alteration information closely related to pegmatite in the region was extracted by principal component analysis, pseudo-anomaly processing and other methods. The high spatial resolution remote-sensing data WorldView-2 and WorldView-3 short-wave infrared images were used and analyzed by principal component analysis (PCA), the band ratio method and multi-class machine learning (ML), combined with conventional thresholds specified the algorithms used to automatically extract Li-bearing pegmatite information. Finally, the Li-bearing pegmatite exploration area was determined, based on a comprehensive analysis of the faults, hydroxyl alteration lithology and Li-bearing pegmatite information. Field investigations have verified that the distribution of pegmatites in the central part of the study area is consistent with that of Li-bearing pegmatites extracted in this study. This study provides a new technique for prospecting Li-bearing pegmatites, which shows that remote-sensing technology possesses great potential for identifying lithium-bearing pegmatites, especially in areas that are not readily accessible.