Rocks are commonly found on the pavement of roads servicing mountainous areas. Rocks detached from a mountainous slope may reach the pavement, trigger a new rockfall through the detachment of other rocks, or even stop somewhere on the slope (i.e., due to vegetation). It is evident that early identification of a rockfall is vital for minimizing impacts on roads, infrastructure and human lives. Several techniques and methods have been proposed for this purpose ranging from monitoring of geophysical signals induced from the falling rock to identification of surface change using conventional and advanced geodetic measurements. The architecture of GEOIM, a rockfall monitoring system for near real-time detection of rockfall incidents, is hereby presented. The proposed system relies upon the integration of geophysical sensors and image analysis techniques and aims to be cost-effective and continuously operating while minimizing false alarm incidents. Suitability of the individual components was verified by small-scale experiments in controlled test environments. The instrumentation for the establishment of efficient and reliable communication links for sensor management and data archival and analysis is also presented.