The evaluation of tourism efficiency is crucial for regional economic development. However, conventional methods rely on low-dimensional statistical data that fail to capture the industry’s complex and high-resolution dynamics. Therefore, a new tourism efficiency evaluation system was developed in this study, utilizing multisource sensor data, including satellite remote sensing (Landsat 8 and Sentinel-2), mobile phone signal trajectories, and points of interest data. We applied a three-stage data envelopment analysis and the Malmquist index to 11 cities in Jiangxi Province (2013–2019). The results showed that while the average comprehensive technical efficiency declined from 0.776 to 0.718 (representing 71–78% of the optimal level), the total factor productivity (TFP) increased at an average annual rate of 22.2% (TFP index = 1.222). This increase was driven by technological progress, with an average technological change index of 1.242. A significant contribution of this study to sensor technology is the establishment of a standardized data fusion method achieving a spatial coverage completeness of more than 95%, enabling high-resolution (1 × 1 km2 grid) spatiotemporal monitoring. These findings prove that integrating multisource sensor networks offers superior analytical depth for identifying bottlenecks in scale efficiency and optimizing regional resource allocation.