Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art. Rohani, N., Pouyet, E., Walton, M., Cossairt, O., & Katsaggelos, A. K. Angewandte Chemie International Edition, 57(34):10910–10914, aug, 2018. Paper doi abstract bibtex Nonlinear unmixing of hyperspectral reflectance data is one of the key problems in quantitative imaging of painted works of art. The approach presented is to interrogate a hyperspectral image cube by first decomposing it into a set of reflectance curves representing pure basis pigments and second to estimate the scattering and absorption coefficients of each pigment in a given pixel to produce estimates of the component fractions. This two-step algorithm uses a deep neural network to qualitatively identify the constituent pigments in any unknown spectrum and, based on the pigment(s) present and Kubelka–Munk theory to estimate the pigment concentration on a per-pixel basis. Using hyperspectral data acquired on a set of mock-up paintings and a well-characterized illuminated folio from the 15th century, the performance of the proposed algorithm is demonstrated for pigment recognition and quantitative estimation of concentration.
@article{Neda2018a,
abstract = {Nonlinear unmixing of hyperspectral reflectance data is one of the key problems in quantitative imaging of painted works of art. The approach presented is to interrogate a hyperspectral image cube by first decomposing it into a set of reflectance curves representing pure basis pigments and second to estimate the scattering and absorption coefficients of each pigment in a given pixel to produce estimates of the component fractions. This two-step algorithm uses a deep neural network to qualitatively identify the constituent pigments in any unknown spectrum and, based on the pigment(s) present and Kubelka–Munk theory to estimate the pigment concentration on a per-pixel basis. Using hyperspectral data acquired on a set of mock-up paintings and a well-characterized illuminated folio from the 15th century, the performance of the proposed algorithm is demonstrated for pigment recognition and quantitative estimation of concentration.},
author = {Rohani, Neda and Pouyet, Emeline and Walton, Marc and Cossairt, Oliver and Katsaggelos, Aggelos K.},
doi = {10.1002/anie.201805135},
issn = {14337851},
journal = {Angewandte Chemie International Edition},
keywords = {deep neural network classification,heritage science,nonlinear unmixing Kubelka–Munk theory,visible hyperspectral imaging},
month = {aug},
number = {34},
pages = {10910--10914},
pmid = {29940088},
title = {{Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art}},
url = {https://onlinelibrary.wiley.com/doi/10.1002/anie.201805135},
volume = {57},
year = {2018}
}
Downloads: 0
{"_id":"gWL2MaTmgcpXHv9J5","bibbaseid":"rohani-pouyet-walton-cossairt-katsaggelos-nonlinearunmixingofhyperspectraldatasetsforthestudyofpaintedworksofart-2018","author_short":["Rohani, N.","Pouyet, E.","Walton, M.","Cossairt, O.","Katsaggelos, A. K."],"bibdata":{"bibtype":"article","type":"article","abstract":"Nonlinear unmixing of hyperspectral reflectance data is one of the key problems in quantitative imaging of painted works of art. The approach presented is to interrogate a hyperspectral image cube by first decomposing it into a set of reflectance curves representing pure basis pigments and second to estimate the scattering and absorption coefficients of each pigment in a given pixel to produce estimates of the component fractions. This two-step algorithm uses a deep neural network to qualitatively identify the constituent pigments in any unknown spectrum and, based on the pigment(s) present and Kubelka–Munk theory to estimate the pigment concentration on a per-pixel basis. Using hyperspectral data acquired on a set of mock-up paintings and a well-characterized illuminated folio from the 15th century, the performance of the proposed algorithm is demonstrated for pigment recognition and quantitative estimation of concentration.","author":[{"propositions":[],"lastnames":["Rohani"],"firstnames":["Neda"],"suffixes":[]},{"propositions":[],"lastnames":["Pouyet"],"firstnames":["Emeline"],"suffixes":[]},{"propositions":[],"lastnames":["Walton"],"firstnames":["Marc"],"suffixes":[]},{"propositions":[],"lastnames":["Cossairt"],"firstnames":["Oliver"],"suffixes":[]},{"propositions":[],"lastnames":["Katsaggelos"],"firstnames":["Aggelos","K."],"suffixes":[]}],"doi":"10.1002/anie.201805135","issn":"14337851","journal":"Angewandte Chemie International Edition","keywords":"deep neural network classification,heritage science,nonlinear unmixing Kubelka–Munk theory,visible hyperspectral imaging","month":"aug","number":"34","pages":"10910–10914","pmid":"29940088","title":"Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art","url":"https://onlinelibrary.wiley.com/doi/10.1002/anie.201805135","volume":"57","year":"2018","bibtex":"@article{Neda2018a,\nabstract = {Nonlinear unmixing of hyperspectral reflectance data is one of the key problems in quantitative imaging of painted works of art. The approach presented is to interrogate a hyperspectral image cube by first decomposing it into a set of reflectance curves representing pure basis pigments and second to estimate the scattering and absorption coefficients of each pigment in a given pixel to produce estimates of the component fractions. This two-step algorithm uses a deep neural network to qualitatively identify the constituent pigments in any unknown spectrum and, based on the pigment(s) present and Kubelka–Munk theory to estimate the pigment concentration on a per-pixel basis. Using hyperspectral data acquired on a set of mock-up paintings and a well-characterized illuminated folio from the 15th century, the performance of the proposed algorithm is demonstrated for pigment recognition and quantitative estimation of concentration.},\nauthor = {Rohani, Neda and Pouyet, Emeline and Walton, Marc and Cossairt, Oliver and Katsaggelos, Aggelos K.},\ndoi = {10.1002/anie.201805135},\nissn = {14337851},\njournal = {Angewandte Chemie International Edition},\nkeywords = {deep neural network classification,heritage science,nonlinear unmixing Kubelka–Munk theory,visible hyperspectral imaging},\nmonth = {aug},\nnumber = {34},\npages = {10910--10914},\npmid = {29940088},\ntitle = {{Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art}},\nurl = {https://onlinelibrary.wiley.com/doi/10.1002/anie.201805135},\nvolume = {57},\nyear = {2018}\n}\n","author_short":["Rohani, N.","Pouyet, E.","Walton, M.","Cossairt, O.","Katsaggelos, A. K."],"key":"Neda2018a","id":"Neda2018a","bibbaseid":"rohani-pouyet-walton-cossairt-katsaggelos-nonlinearunmixingofhyperspectraldatasetsforthestudyofpaintedworksofart-2018","role":"author","urls":{"Paper":"https://onlinelibrary.wiley.com/doi/10.1002/anie.201805135"},"keyword":["deep neural network classification","heritage science","nonlinear unmixing Kubelka–Munk theory","visible hyperspectral imaging"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://sites.northwestern.edu/ivpl/files/2023/06/IVPL_Updated_publications-1.bib","dataSources":["ePKPjG8C6yvpk4mEK","ya2CyA73rpZseyrZ8","D8k2SxfC5dKNRFgro","7Dwzbxq93HWrJEhT6","qhF8zxmGcJfvtdeAg","fvDEHD49E2ZRwE3fb","H7crv8NWhZup4d4by","DHqokWsryttGh7pJE","vRJd4wNg9HpoZSMHD","sYxQ6pxFgA59JRhxi","w2WahSbYrbcCKBDsC","XasdXLL99y5rygCmq","3gkSihZQRfAD2KBo3","t5XMbyZbtPBo4wBGS","bEpHM2CtrwW2qE8FP","teJzFLHexaz5AQW5z"],"keywords":["deep neural network classification","heritage science","nonlinear unmixing kubelka–munk theory","visible hyperspectral imaging"],"search_terms":["nonlinear","unmixing","hyperspectral","datasets","study","painted","works","art","rohani","pouyet","walton","cossairt","katsaggelos"],"title":"Nonlinear Unmixing of Hyperspectral Datasets for the Study of Painted Works of Art","year":2018}