Launched in year 1997, Tropical Rainfall measuring Mission (TRMM) Precipitation Radar (PR) is known to be the first space borne observation platform for mapping precipitation over the tropics. Rainfall measured from TRMM PR is important in order to study global intensity and distribution of precipitation. Ground validation is a critical component in the TRMM system. However, the ground sensing systems have quite different characteristics from TRMM in terms of resolution, scale, viewing aspect, and uncertainties in the sensing environments. In this paper a novel hybrid Neural Network model is presented to train ground radars for rainfall estimation using rain gauge data and subsequently using the trained ground radar rainfall estimation to train TRMM PR based Neural networks. This hybrid neural network model will derive the relation between rain gauges and ground radar measurements, and transfer this relation to adaptive rainfall estimation for TRMM-PR in order to estimate rainfall and generate global rainfall maps. One year of data over Melbourne Florida from gauges, ground radar and TRMM/PR is used to demonstrate this hybrid approach. The performance of the rainfall product estimated from TRMM PR is compared against TRMM standard products. A direct gauge comparison study is done to demonstrate the improvement brought in by this hybrid neural networks approach.