Based on satellite images of two different wavelengths acquired by ALOS-2 and Sentinel-1, the authors used DInSAR two-track method and stacking technology to analyze and delineate the active landslide that is undergoing surface deformation in the central part of Maoxian County, Sichuan Province. First, in view of the long time baseline of the ALOS-2 satellite and the small amount of image data, the two-track DInSAR method was used to detect landslide deformation. Secondly, on the basis of the short-time baseline multi-scene Sentinel-1 data, the stacking technology was used to detect landslide deformation. Finally, comprehensive analysis of the deformation rate results of the two data sets was conducted, and the potential landslide area in the middle of Maoxian County of Sichuan was delineated. The results show that the suspected active landslides in Baibu Village of Maoxian County and other places have obvious surface deformation, and the maximum absolute amount of radar line-of-sight deformation rate reaches 200 mm/a; combined with optical image characteristics and existing historical survey data, 8 active landslides were delineated, and the exploration result of two kinds of data shows that the detection results of the data can correspond to each other in the spatial distribution of 7 active landslides. Field surveys were conducted on 6 active landslides, and signs of ground deformation were found. The work in this paper shows that using a small number of long-band ALOS-2 images and the DInSAR two-track method can detect more obvious landslide surface deformation in mountainous areas with certain vegetation coverage; when C-band Sentinel-1 images are applied, accumulation of continuous multi-scene data is required; in addition, by applying time series analysis method, the detection effect is better than the two-track method using ALOS-2 images.