What’s Wrong with the Murals at the Mogao Grottoes: A Near-Infrared Hyperspectral Imaging Method

Although a significant amount of work has been performed to preserve the ancient murals in the Mogao Grottoes by Dunhuang Cultural Research, non-contact methods need to be developed to effectively evaluate the degree of flaking of the murals. In this study, we propose to evaluate the flaking by automatically analyzing hyperspectral images that were scanned at the site. Murals with various degrees of flaking were scanned in the 126th cave using a near-infrared (NIR) hyperspectral camera with a spectral range of approximately 900 to 1700 nm. The regions of interest (ROIs) of the murals were manually labeled and grouped into four levels: normal, slight, moderate, and severe. The average spectral data from each ROI and its group label were used to train our classification model. To predict the degree of flaking, we adopted four algorithms: deep belief networks (DBNs), partial least squares regression (PLSR), principal component analysis with a support vector machine (PCA + SVM) and principal component analysis with an artificial neural network (PCA + ANN). The experimental results show the effectiveness of our method. In particular, better results are obtained using DBNs when the training data contain a significant amount of striping noise.

[1]  Giles M. Foody,et al.  Feature Selection for Classification of Hyperspectral Data by SVM , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Rolf B. Saager,et al.  Polarization-Sensitive Hyperspectral Imaging in vivo: A Multimode Dermoscope for Skin Analysis , 2014, Scientific Reports.

[3]  Yuval Garini,et al.  Spectral imaging: Principles and applications , 2006, Cytometry. Part A : the journal of the International Society for Analytical Cytology.

[4]  R. Cavalli,et al.  Remote hyperspectral imagery as a support to archaeological prospection , 2007 .

[5]  Alan R. Gillespie,et al.  Sub-pixel artifact detection using remote sensing , 2003 .

[6]  Rainer Künnemeyer,et al.  Method of Wavelength Selection for Partial Least Squares , 1997 .

[7]  Shaohui Mei,et al.  An accurate SVM-based classification approach for hyperspectral image classification , 2013, 2013 21st International Conference on Geoinformatics.

[8]  David Lognoli,et al.  Detection and characterization of biodeteriogens on stone cultural heritage by fluorescence lidar. , 2002, Applied optics.

[9]  Maria Udén.  Swedish case study , 2004 .

[10]  Yongchao Zhao,et al.  A High-Order Statistical Tensor Based Algorithm for Anomaly Detection in Hyperspectral Imagery , 2014, Scientific reports.

[11]  Rafael Abella,et al.  Characterization of trace gases' fluctuations on a ‘low energy’ cave (Castañar de Íbor, Spain) using techniques of entropy of curves , 2011 .

[12]  Pierre Genthon,et al.  Microclimates of l'Aven d'Orgnac and other French limestone caves (Chauvet, Esparros, Marsoulas) , 2006 .

[13]  C. Freitas,et al.  The role and importance of cave microclimate in the sustainable use and management of show caves , 2010 .

[14]  Bor-Chen Kuo,et al.  A Kernel-Based Feature Selection Method for SVM With RBF Kernel for Hyperspectral Image Classification , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[15]  Hui Zhang,et al.  Investigation of the renewed diseases on murals at Mogao Grottoes , 2013, Heritage Science.

[16]  Iacopo Mochi,et al.  Fluorescence lidar imaging of the cathedral and baptistery of Parma , 2003 .

[17]  Craig Barnes,et al.  The relationship between local climate and radon concentrations in the Temple of Baal, Jenolan Caves, Australia , 2003 .

[18]  Renfu Lu,et al.  Hyperspectral and multispectral imaging for evaluating food safety and quality , 2013 .

[19]  Stephen Marshall,et al.  Effective Feature Extraction and Data Reduction in Remote Sensing Using Hyperspectral Imaging [Applications Corner] , 2014, IEEE Signal Processing Magazine.

[20]  Gang Wang,et al.  Deep Learning-Based Classification of Hyperspectral Data , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[21]  Stephen Marshall,et al.  Quantitative Prediction of Beef Quality Using Visible and NIR Spectroscopy with Large Data Samples Under Industry Conditions , 2015 .

[22]  Shutao Li,et al.  Novel Two-Dimensional Singular Spectrum Analysis for Effective Feature Extraction and Data Classification in Hyperspectral Imaging , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[23]  Giovanna Cecchi,et al.  Fluorescence lidar technique for the remote sensing of stone monuments , 2000 .

[24]  S Svanberg,et al.  Hyperspectral fluorescence lidar imaging at the Colosseum, Rome: elucidating past conservation interventions. , 2008, Optics express.

[25]  Jason Weston,et al.  Semisupervised Neural Networks for Efficient Hyperspectral Image Classification , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[26]  R. Chiari,et al.  Fluorescence lidar monitoring of historic buildings. , 1998, Applied optics.

[27]  Gamal ElMasry,et al.  Near-infrared hyperspectral imaging for predicting colour, pH and tenderness of fresh beef , 2012 .

[28]  Susanna Bracci,et al.  Lidar remote sensing of stone cultural heritage: detection of protective treatments , 2001 .

[29]  Zhenfeng Shao,et al.  A Novel Hierarchical Semisupervised SVM for Classification of Hyperspectral Images , 2014, IEEE Geoscience and Remote Sensing Letters.

[30]  Sune Svanberg,et al.  Fluorescence Lidar Multispectral Imaging for Diagnosis of Historical Monuments, Ö–vedskloster: A Swedish Case Study , 2007 .

[31]  Ashok Samal,et al.  Partial least squares analysis of near-infrared hyperspectral images for beef tenderness prediction , 2008 .

[32]  Lei Guo,et al.  Object Detection in Optical Remote Sensing Images Based on Weakly Supervised Learning and High-Level Feature Learning , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[33]  Maryam Imani,et al.  Principal component discriminant analysis for feature extraction and classification of hyperspectral images , 2014, 2014 Iranian Conference on Intelligent Systems (ICIS).

[34]  Arif Mahmood,et al.  Hyperspectral Face Recognition With Spatiospectral Information Fusion and PLS Regression , 2015, IEEE Transactions on Image Processing.

[35]  Shiming Xiang,et al.  Aircraft Detection by Deep Belief Nets , 2013, 2013 2nd IAPR Asian Conference on Pattern Recognition.

[36]  H Edner,et al.  Fluorescence lidar imaging of historical monuments. , 2001, Applied optics.

[37]  Iacopo Mochi,et al.  Lithotypes characterization with a fluorescence lidar imaging system using a multi-wavelength excitation source , 2003, SPIE Remote Sensing.

[38]  Sune Svanberg,et al.  Documentation of soiled and biodeteriorated facades: A case study on the Coliseum, Rome, using hyperspectral imaging fluorescence lidars , 2009 .

[39]  Fabio Del Frate,et al.  Pixel Unmixing in Hyperspectral Data by Means of Neural Networks , 2011, IEEE Transactions on Geoscience and Remote Sensing.