Finding the production date of a pre-modern manuscript is commonly a long process in historical research, requiring days of work from highly specialised experts. In this paper, we present an automatic dating method based on modelling both the language and the image data. By creating a statistical model over the changes in the pen strokes and short character sequences in the transcribed text, a combination of multiple estimators give a distribution over the time line for each manuscript. We have evaluated our estimation scheme on the medieval charter collection "Svenskt Diplomatariums huvudkartotek" (SDHK), including more than 5300 transcribed charters from the period 1135 - 1509. Our system is capable of achieving a median absolute error of 12 years, where the only human input is a transcription of the charter text. Since reading and transcribing the text is a skill that many researchers and students have, compared to the more specialized skill of dating medieval manuscripts based on palaeographical expertise, we find our novel approach suitable for helping individual researchers to date collections of manuscript pages. For larger collections, transcriptions could also be collected using crowd sourcing. 

In this article, the so-called Quill Feature method is tested as a means for scribal attribution in mediaeval manuscripts and placed in a palaeographic context. The Quill Features encompass a range of measurable features such as stroke curvature and variation in the width of strokes that graphs are comprised of, giving consideration to the angle at which they are drawn. The construction of the mediaeval pen, the quill, creates variation in the width of strokes, most importantly due to direction and force of pressure. It has been assumed that individual patterns can be discerned in this variation. The manuscript chosen for investigation is Cod. Ups. C 61, which contains a number of St Birgitta's Revelations and other material connected with this saint. Earlier researchers claimed that three or four scribes produced this manuscript, and this assertion was tested with the Quill Feature method. It transpired that the resultant values, calculated through measuring the investigated features, were very similar for the pages produced by the same scribe, which means that the method identifies important regularities in the scribal hands. Nevertheless, some of the investigated hands showed great similarities and were difficult for the computer to separate. Finally, the strengths and weaknesses of the digital methods of script analysis are discussed. The main advantage of these approaches is that a computer can measure characteristics of the script that cannot be discerned with the naked eye, and with a high degree of exactness. One weakness is that the measurements obtained are given in numbers, and can thus be difficult to apply to the script signs. Furthermore, one single method cannot provide a full description of script characteristics, so a combination of methods, both digital and manual, is desirable.

In this paper we propose a novel approach for manuscript dating based on shape statistics. Our goal was to develop a strategy well suited for a large scale dating effort where heterogeneous collections of thousands of manuscripts could be automatically processed. The proposed method takes the gray scale image as input. Uses the stroke width transform and a statistical model of the gradient image to find ink boundaries. Finally, a distribution over common shapes, quantified using shape context descriptors, is produced for each manuscript image. The proposed method is binarization free, rotational invariant and requires minimal segmentation. Also, we propose parameters estimation schemes, simplifying the deployment process. We evaluate our work on the 10000+ manuscripts collection “Svenskt diplomatariums huvudkarotek”, consisting of charters from the medieval period of todays Sweden. The images, originally intended for web viewing, were of low quality and had compression artifacts, but could still be dated with a median absolute error of < 19 years. Due to unsupervised feature learning and regression, we get these results using only 5% of the labels in the estimator training