Understanding the river dynamics is very important to be able to make use of all river functions and to protect ourselves against floods. Hydrodynamic models are used to predict river behaviour and one of the input parameters is the river geometry. In-situ measuring of river geometry can be very expensive and time consuming. Remote sensing offers a more efficient method to monitor rivers, because it has the ability to continuously monitor the Earth surface at multiple scales. Rivers in monsoon dominated regions show a strong seasonal variation in discharge and have a high variability in morphodynamics. Therefore it is very important to have a high spatial-temporal resolution of river geometry data as input for hydrodynamic models to keep up with the changes in the river. Both optical and microwave images can be used in providing information about the river geometry. The microwave images need a lot of processing to deal with noise, but they are not limited by clouds, whereas optical sensors are troubled by clouds but produce less noise. The benefit of combining these methods is because the microwave images could fill the gap of cloud covered optical images during the monsoon. The main result from this thesis is that more investigation is needed on how to deal with the noise produced by the microwave images before the methods can be combined. Nevertheless it is shown that using the Canny Edge detector and Otsu thresholding improves the results. With this research we are a small step further in global river monitoring. This is important in particular for parts of the world where rivers are not continuously monitored because they are situated in hard to reach terrain and because the country will not invest in local gauge stations. Having more knowledge about river behaviour will help to get a better understanding of water losses along the river course, habitat change and flood risks.