This paper addresses the estimation of different context parameters of a primary user network based on signal strength measurements obtained by a sensor network and without any prior knowledge of the antenna types used by primary transmitter or the propagation model. The obtained context information can be stored in a database like a Radio Environment Map (REM) and be used to decide the appropriate configuration of a secondary cognitive radio network. Proposed methodology is based on applying image processing over the set of measurements to identify the existing transmitters in the scenario and their parameters such as position, antenna pattern and propagation model. ; Peer Reviewed ; Postprint (published version)