In this article, we present a method for combiningimage-based displacements to build time series. This method takesadvantage of the redundancy of these displacements, which comesfrom multiple possible combinations between images. The proposedmethod combines common master series with a differentmaster date to construct a single time series of relative displacements.We were able to test thismethod on displacements computedusing images from a stereo time-lapse device recording imagesof the Argentière glacier during the summer and fall of 2019.Our method was compared to two other displacement aggregationstrategies: a simple common master approach and the classicalinversion method. In order to perform this comparison, displacementdata from four permanent GPS are used. The results showthat our method provides a more accurate time series of relativedisplacements than those obtained with the other methods. Thehigher accuracy of the proposed method makes the detection ofseasonal variations in glacier velocity possible.