In recent years many experimental specialists have been carried out analysis of some pictures to say if they are authentic work of art. A lot of this work are checked for possible forgery work in museum and, on the other hand, some people want to know if they have a picture of a famous painter. The work of Spanish painter Francisco Goya is one of more questioned, that is due to the large number of pictures he made. The work, that it is presented in this paper, can help to expert people to authenticate that kind of pictures because it is another sort of verification. This paper describes the development of a fully-automated computerized analysis system of Goya's pictures, which enables locate some small prints (graphisms) in the authentic Goya's work. The proposed methodology works on four steps: digital image acquisition on gray level, segmentation, pattern recognition and localization of small prints in the picture. The original work of art is digitalized in gray level because strokes are the same in all colour or in gray level and the more important thing of this paper is the study of lines. To have digitalized gray level images as input is better about time of process and storage capacity. Segmentation step looks for regions in all of gray levels but considering that a person, in general, can discriminate only between 64 levels of gray. All of segmented regions are scaled at the same size but keeping their measurement. The out of process are the location of some small print (graphism) if the input digital image is a Goya's picture. An artificial backpropagation neural network is used as pattern recognition to say if graphism segmented is a Goya's small print or no.