International audienceThis article presents a new semi-automatic method for charge and mass identification of charged nuclear fragments using either ∆E − E correlations between measured energy losses in two successive detectors or correlations between charge signal amplitude and rise time in a single silicon detector, derived from digital pulse shape analysis techniques. In both cases different nuclear species (defined by their atomic number Z and mass number A) can be visually identified from such correlations if they are presented as a two-dimensional histogram (’identification matrix’), in which case correlations for different species populate different ridge lines (’identification lines’) in the matrix. The proposed algorithm is based on the identification matrix’s properties and uses as little information as possible on the global form of the identification lines, making it applicable to a large variety of matrices. Particular attention has been paid to the implementation in a suitable graphical environment, so that only two mouse-clicks are required from the user to calculate all initialization parameters. Example applications to recent data from both INDRA and FAZIA telescopes are presented