TY - JOUR N2 - The paper considers the problem of increasing the generalization ability of classification systems by creating an ensemble of classifiers based on the CNN architecture. Different structures of the ensemble will be considered and compared. Deep learning fulfills an important role in the developed system. The numerical descriptors created in the last locally connected convolution layer of CNN flattened to the form of a vector, are subjected to a few different selection mechanisms. Each of them chooses the independent set of features, selected according to the applied assessment techniques. Their results are combined with three classifiers: softmax, support vector machine, and random forest of the decision tree. All of them do simultaneously the same classification task. Their results are integrated into the final verdict of the ensemble. Different forms of arrangement of the ensemble are considered and tested on the recognition of facial images. Two different databases are used in experiments. One was composed of 68 classes of greyscale images and the second of 276 classes of color images. The results of experiments have shown high improvement of class recognition resulting from the application of the properly designed ensemble. L1 - http://journals.pan.pl/Content/123185/PDF/2636_BPASTS_2022_70_3.pdf L2 - http://journals.pan.pl/Content/123185 PY - 2022 IS - 3 EP - e141004 DO - 10.24425/bpasts.2022.141004 KW - CNN KW - ensemble of classifiers KW - face recognition KW - feature selection A1 - Szmurło, Robert A1 - Osowski, Stanislaw VL - 70 DA - 17.05.2022 T1 - Ensemble of classifiers based on CNN for increasing generalization ability in face image recognition SP - e141004 UR - http://journals.pan.pl/dlibra/publication/edition/123185 T2 - Bulletin of the Polish Academy of Sciences Technical Sciences ER -