At this paper we present a comprehensive method of classification, recognition and identification of Byzantine fresco images through digital image processing. The methodology is interdisciplinary, as it is examined from Byzantine art and its distinct characteristics, as well as from technologies that contribute to the identification and design background using machine learning. The whole process is extensively presented at all stages. The known painters from this era are Michael Astrapas and Eutychios, Manuel Panselinos and Georgios Kalliergis, as their signatures, or their names, have been found on frescoes in monuments from 13th and 14th centuries. Those artists are the representatives of the Macedonian School. The stages followed are: the creation of an integrated database consisting of 123 color images of frescoes and corresponding informational material, the digital analysis of images from 13th and 14th century Byzantine frescoes with additional images from the post-Byzantine period of the Cretan School, the creation of image management systems that do not have digital imprinting elements via Python where items are organized into three main entities, Picture, Work, Artist, and the use of Alex Net from CNNs for accurate image categorization. The system was trained on all data in the learning set with an identification rate of 85%. From the results of the plots, the model behaves satisfactorily and overfitting was avoided. This study can be used as an additional tool for art scholars in the identification and analysis of images by combining the historical analysis of frescoes through Byzantine art with computer vision.