Thanks to their high strength to mass ratio, composite materials are now widespread in the aerospace industry. Nevertheless, this type of material is subject to various internal damages and it is mandatory to monitor in real time their structural integrity. Structural Health Monitoring (SHM) is a process based on embedded sensors whose aim is to detect, locate, classify and quantify potential damages appearing in a structure in order to avoid structures catastrophic failures and to estimate their residual life. The most widely used technique to perform SHM of aeronautical structures made up of composite materials is based on the use of ultrasonic Lamb waves. However, even if robust and precise SHM algorithms exist for damage detection and localization, there is still a huge need for reliable algorithms for damage quantification. In this paper, a damage quantification strategy based on a post-processing step of the results of damage imaging method is presented. Such a method allows for damage size assessment of a delaminated area by post-processing the images produced by damage localization algorithms. Localization methods take raw signals from sensor as input and return a map of index representing the likelihood of presence of a damage over the surface of the structure under study. From this spatial probability map, region of high localization index can be identified around the estimated damage location and the area this region can be computed. A data-driven model representing the mathematical relationship between the computed area and the actual size of the damage is then inferred. The proposed method is successfully validated on numerical simulation data carried out on CFRP plate samples equipped with a stiffener and of a piezoelectric sensor-actuator network with several configurations of damage size.