Abstract : Among different real time imaging techniques of macro molecular interactions, the emerging SPR (Surface Plasmon Resonance) approach is one of the most promising: no molecular labeling is necessary to reveal interactions, and many protein/protein couples can be studied in the same experiment. Such a real time monitoring of various biomolecular interactions raises several challenges in terms of image segmentation, extraction and interpretation of the mean hybridization signal of each spot, and computation of the affinity constants for each protein/protein couples. This paper describes an automated approach for SPR image analysis resolving most of the previous issues. First, a differential signal is computed in order to only conserve the interactions according to the time. Second, a spatio-temporal filtering is used to remove the noise present in the SPR raw data. Then, the data is segmented using the k-means method, in order to identify regions of the spots according to their temporal behavior. Non-uniform hybridization within a spot can thus be detected and the pixels excluded from analysis. In the same manner, the presence of image artifacts (chip scratch, deposit leakage) can be detected and not taken into account for analysis. The mean signal over each spot is then extracted and its temporal behavior provides the kinetic parameters characterizing the biological interaction (association and dissociation constants). The preliminary results obtained on test biomolecular interactions confirmed our expectations concerning our detection capability even at low concentration of analytes. The affinity constants obtained has been compared to those obtained by ELISA. The results are in good agreement, validating therefore the SPR approach for protein/protein interactions studies.