International audience Characterization of randomness and spatial variability of material properties, loading or deterioration processes of structures, from inspection, is a very challenging task that concentrates the improvement of Non Destructive Testing (NDT) tools on more efficient and higher structural coverage. In case of characterization of random loading or material properties, this challenge is arduous because of the limited number of measures and the quasi-infinite potential positions of local failures. This paper presents a two stages procedure for the stochastic characterization of random fields from NDT measurements. According to the stationarity property, the optimal quantity of NDT measurements and their position can be assessed once a quality requirement is included. The proposed procedure allows: (i) to quantify the properties of the ergodic stationary field, and (ii) to assess the second order parameters of the studied random variables. The paper ends illustrating the methodology with an application that considers the inspection of a concrete beam with a capacitive technique. The results indicate that the proposed methodology is useful to identify the parameters of a random field by reducing the number of NDT measures under given quality requirements.