Approach to Identification of Changes from Local Surface Normal Analysis of RTI Data in Application to Cultural Heritage

Saha, Sunita; Lewis, David A.; Sitnik, Robert

doi:10.1145/3552464.3555685

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To characterize the directionality of change on a mock-up painting before and after damage, Manfredi et al (2013Manfredi et al ( , 2014 suggested a methodology that looks at the documentation of topographic changes using surface normals [9,10]. Following the surface normal deviation approach, Saha et al (2022) proposed a semi-quantitative segmentation approach based on local normal distribution; however, the work clearly stated the need for image registration to quantify changes accurately [11]. To overcome the limitations of reconstruction models Dulecha et.al.…”

Section: Introductionmentioning

confidence: 99%

Appearance segmentation and documentation applied to cultural heritage surfaces

Saha¹,

Σιάτου²,

Degrigny³

et al. 2023

View full text Add to dashboard Cite

This paper describes the application of a novel supervised segmentation technique used for conservation documentation based on visible appearance changes of Cultural Heritage (CH) metal surfaces. The technique employs a linear discriminant analysis model to classify Reflectance Transformation Imaging (RTI) reconstruction coefficients. The Hemispherical Harmonics (HSH) reconstruction coefficients for each pixel are first calculated and then normalized. This normalization enhances the robustness and invariance of the application, making it possible to apply it for documenting different surfaces at different time intervals. This paper presents three cases related to surface appearance changes due to corrosion that provokes low or high topographic changes. Tarnishing of silver and filiform corrosion on steel are examined. For each case study, a supervised data set is constructed, teaching the algorithm to recognize as distinct a specified appearance characteristic (such as corrosion, metal, etc.) by comparing it to the reconstruction coefficients of each pixel through Linear Discriminant Analysis (LDA). A simplified color map visualizes the segmented information. The calculated results are afterward verified by visible inspection from conservation-restoration experts. The method can segment surfaces with changes in micro-geometry, creating accurate cartographies of the object's condition. However, limitations are met on surfaces with minimal topography and high specularity.

show abstract