The fields of conservation science and scientific analysis for cultural heritage research are typically characterized by limited studies of unique artifacts. This research proposes a change to the current approach by use of a standard scientifically-based diagnostic approach similar to that followed in the field of medicine.

The non-invasive identification of the artist’s palette of pigments is the focus of this research in the development of the first diagnostic exam for the assessment of painted artworks. The existing techniques of technical photography and X-ray fluorescence spectroscopy were applied in a dual-technique protocol designed especially to make the analysis fast, reliable, and low-cost. The exam methodology incorporated a new systematic approach for data processing, combining the information from the two complementary techniques for accurate and consistent pigment identification results.

The exam methodology was implemented in three different scenarios: a historical drawing, a fresco mural, and wall paintings, each of which served as a test for the protocol in a field environment.

As a member of an interdisciplinary team in the Qualcomm Institute’s Center of Interdisciplinary Science for Art, Architecture, and Archaeology (CISA3), advanced tools for advanced implementation of the diagnostic exam were developed and informed by the field scenarios. A platform for data visualization was created using an augmented reality tablet device to provide data contextualization for the acquisition and analysis workflow and improve the retention of important metadata. A robotic platform was also adapted from a 3D printer to provide automation accelerating the data acquisition phase of the exam.

Finally, two experiments were designed to test the capabilities of the diagnostic exam on known samples that were more complex than single pigment swatches and more closely resembled an actual artifact. A mock-up oil painting and controlled mixture and layering scenarios were used as test samples to qualitatively evaluate the performance of the exam to deliver accurate results. This research presents a data-driven approach to preventative conservation that is scalable and suited to the variety of users in the field.