Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art

Rohani, Neda; Pouyet, Émeline; Walton, Marc; Cossairt, Oliver; Katsaggelos, Aggelos K.

doi:10.1002/ange.201805135

Cited by 23 publications

(28 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although not implemented in this study, further data processing can also be applied benefitting from access to high resolution SWIR data, such as linear spectral unmixing and semi-quantitative absorption depth analyses, as well as other chemometric approaches [114][115][116][117][118]. The ability to record high spectral resolution image cubes in the SWIR opens possibilities for in-depth materials analysis and mapping for on-site wall paintings, which the SRS hyperspectral system has begun to demonstrate with these archaeological case studies.…”

Section: Materials Identification and Mappingmentioning

confidence: 99%

Acquisition of High Spectral Resolution Diffuse Reflectance Image Cubes (350–2500 nm) from Archaeological Wall Paintings and Other Immovable Heritage Using a Field-Deployable Spatial Scanning Reflectance Spectrometry Hyperspectral System

Radpour

Delaney

Kakoulli

2022

Sensors

View full text Add to dashboard Cite

There is growing interest in bringing non-invasive laboratory-based analytical imaging tools to field sites to study wall paintings in order to collect molecular information on the macroscale. Analytical imaging tools, such as reflectance imaging spectrometry, have provided a wealth of information about artist materials and working methods, as well as painting conditions. Currently, scientific analyses of wall paintings have been limited to point-measurement techniques such as reflectance spectroscopy (near-ultraviolet, visible, near-infrared, and mid-infrared), X-ray fluorescence, and Raman spectroscopy. Macroscale data collection methods have been limited to multispectral imaging in reflectance and luminescence modes, which lacks sufficient spectral bands to allow for the mapping and identification of artist materials of interest. The development of laboratory-based reflectance and elemental imaging spectrometers and scanning systems has sparked interest in developing truly portable versions, which can be brought to field sites to study wall paintings where there is insufficient space or electrical power for laboratory instruments. This paper presents the design and testing of a simple hyperspectral system consisting of a 2D spatial spot scanning spectrometer, which provides high spectral resolution diffuse reflectance spectra from 350 to 2500 nm with high signal to noise and moderate spatial resolution (few mm). This spectral range at high spectral resolution was found to provide robust chemical specificity sufficient to identify and map many artists’ materials, as well as the byproducts of weathering and conservation coatings across the surface of ancient and Byzantine Cypriot wall paintings. Here, we present a detailed description of the hyperspectral system, its performance, and examples of its use to study wall paintings from Roman tombs in Cyprus. The spectral/spatial image processing workflow to make maps of pigments and constituent painting materials is also discussed. This type of configurable hyperspectral system and the imaging processing workflow offer a new tool for the field study of wall paintings and other immovable heritage.

show abstract

Section: Materials Identification and Mappingmentioning

confidence: 99%

Acquisition of High Spectral Resolution Diffuse Reflectance Image Cubes (350–2500 nm) from Archaeological Wall Paintings and Other Immovable Heritage Using a Field-Deployable Spatial Scanning Reflectance Spectrometry Hyperspectral System

Radpour

Delaney

Kakoulli

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Rohani et al (2018) implemented nonlinear unmixing to learn from painted art works. Deep neural network was explored to identify various ingredient pigments in any spectrum.…”

Section: Modern Application Areas Of Hyperspectral Imagingmentioning

confidence: 99%

Critical insights into modern hyperspectral image applications through deep learning

Jaiswal

Sharma

Yadav

2021

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Hyperspectral imaging has shown tremendous growth over the past three decades. Hyperspectral imaging was evolved through remote sensing. Along, with the technological enhancements hyperspectral imaging has outgrown, conquering over other various application areas. In addition to it, data enriched data cubes with abundant spectral and spatial information works as perk for capturing, analyzing, reviewing, and interpreting results from data. This review concentrates on emerging application areas of hyperspectral imaging. Emerging application areas are selected in ways where there is a vast scope for future enhancements by exploiting cutting edge technology, that is, deep learning. Applications of hyperspectral imaging techniques in some selected areas (remote sensing, document forgery, history and archaeology conservation, surveillance and security, machine vision for fruit quality inspection, medical imaging) are focused. The review pivots around the publicly available datasets and features used domain wise. This review can act as a baseline for deep learning and machine vision experts, historical geographers, and scholars by providing them a view of how hyperspectral imaging is implemented in multiple domains along with future research prospects. This article is categorized under: Technologies > Machine Learning Technologies > Prediction

show abstract

“…In addition, machine learning also handles the challenge of processing large amounts of data which is often a major concern in cultural heritage applications. Such methodology has already been reported in the study of illuminated manuscripts where hyperspectral imaging and a deep neural network were combined to perform the spectral unmixing and quantitative estimation of pigment concentrations 27 . Another important work on the degraded medieval manuscript 28 proposed a codebook algorithm to fuse the hyperspectral data and XRF data that successfully revealed the hidden content through the correlated spectral mapping.…”

Section: Introductionmentioning

confidence: 99%

Digital restoration of colour cinematic films using imaging spectroscopy and machine learning

Liu

Catelli

Katsaggelos

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Digital restoration is a rapidly growing methodology within the field of heritage conservation, especially for early cinematic films which have intrinsically unstable dye colourants that suffer from irreversible colour fading. Although numerous techniques to restore film digitally have emerged recently, complex degradation remains a challenging problem. This paper proposes a novel vector quantization (VQ) algorithm for restoring movie frames based on the acquisition of spectroscopic data with a custom-made push-broom VNIR hyperspectral camera (380–780 nm). The VQ algorithm utilizes what we call a multi-codebook that correlates degraded areas with corresponding non-degraded ones selected from reference frames. The spectral-codebook was compared with a professional commercially available film restoration software (DaVinci Resolve 17) tested both on RGB and on hyperspectral providing better results in terms of colour reconstruction.

show abstract

Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art

Cited by 23 publications

References 35 publications

Acquisition of High Spectral Resolution Diffuse Reflectance Image Cubes (350–2500 nm) from Archaeological Wall Paintings and Other Immovable Heritage Using a Field-Deployable Spatial Scanning Reflectance Spectrometry Hyperspectral System

Acquisition of High Spectral Resolution Diffuse Reflectance Image Cubes (350–2500 nm) from Archaeological Wall Paintings and Other Immovable Heritage Using a Field-Deployable Spatial Scanning Reflectance Spectrometry Hyperspectral System

Critical insights into modern hyperspectral image applications through deep learning

Digital restoration of colour cinematic films using imaging spectroscopy and machine learning

Contact Info

Product

Resources

About