FTIR spectral imaging enhancement for teacher’s facial expressions recognition in the intelligent learning environment

Abstract Fourier transform infrared (FTIR) spectral imaging is a valuable tool for rapid identification of the real/fake human face in a single spectral image. However, it often suffers from the problems of band overlap and random noise. In this work, we have developed a spectral resolution enhancement algorithm via the contourlet transforms regularization for FTIR spectral imaging. First, we apply the contourlet transform to the observed FTIR spectrum and the ground-truth one, and compare their distributions of the contourlet coefficients. It shows that the distribution of the ground-truth spectrum is sparser than that of the observed one. Then, an infrared spectral resolution enhancement model is built to regularize the distribution of the observed FTIR spectrum by total variation regularization. Furthermore, an alternating scheme is exploited to estimate the instrument function and the latent FTIR spectrum, and it shows excellent convergence and stability. Experimental results show the effectiveness of the proposed approach in suppressing random noise and preserving spectral details in the FTIR spectral resolution enhancement problems. Thus, the proposed method leads the high-resolution FTIR spectrum as a more efficient tool for the recognition of teacher's facial expressions in the intelligent learning environment.

[1]  Meng Li,et al.  Deconvolution methods based on φHL regularization for spectral recovery. , 2015, Applied optics.

[2]  Hanping Hu,et al.  Human behavior during Flash Crowd in web surfing , 2014 .

[3]  Yi Chang,et al.  Anisotropic Spectral-Spatial Total Variation Model for Multispectral Remote Sensing Image Destriping , 2015, IEEE Transactions on Image Processing.

[4]  Hai Liu,et al.  Nonlocal low-rank-based blind deconvolution of Raman spectroscopy for automatic target recognition. , 2018, Applied optics.

[5]  B. Billinghurst,et al.  High resolution synchrotron radiation Fourier transform infrared spectrum of the COH-bending mode in methanol-D 1 (CH 2 DOH) , 2017 .

[6]  Kun Zhang,et al.  Robust head pose estimation using Dirichlet-tree distribution enhanced random forests , 2016, Neurocomputing.

[7]  Wenbing Tao,et al.  A Constrained Radial Agglomerative Clustering Algorithm for Efficient Structure From Motion , 2018, IEEE Signal Processing Letters.

[8]  Yangjie Wei,et al.  Multi-feature fusion for thermal face recognition , 2016 .

[9]  Qian Du,et al.  Fusing Local and Global Features for High-Resolution Scene Classification , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[10]  Sanya Liu,et al.  Blind spectral deconvolution algorithm for Raman spectrum with Poisson noise , 2014 .

[11]  Ronei J. Poppi,et al.  Direct analysis of the main chemical constituents in Chenopodium quinoa grain using Fourier transform near-infrared spectroscopy , 2015 .

[12]  Naixue Xiong,et al.  Analysis and Design of Digital Chaotic Systems With Desirable Performance via Feedback Control , 2015, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[13]  Hao Zhang,et al.  FBRDLR: Fast blind reconstruction approach with dictionary learning regularization for infrared microscopy spectra , 2018 .

[14]  Tom Goldstein,et al.  The Split Bregman Method for L1-Regularized Problems , 2009, SIAM J. Imaging Sci..

[15]  P. Ekman,et al.  Facial Expressions of Emotion , 1979 .

[16]  Sanya Liu,et al.  Blind image restoration with sparse priori regularization for passive millimeter-wave images , 2016, J. Vis. Commun. Image Represent..

[17]  Paul Benjamin Crilly.  Increased throughput for process chromatography using constrained deconvolution , 1992, IEEE Trans. Ind. Electron..

[18]  Kun Zhang,et al.  Towards Improving Social Communication Skills With Multimodal Sensory Information , 2014, IEEE Transactions on Industrial Informatics.

[19]  Tao Huang,et al.  Blind Spectral Signal Deconvolution with Sparsity Regularization: An Iteratively Reweighted Least-Squares Solution , 2017, Circuits Syst. Signal Process..

[20]  Ting-Zhu Huang,et al.  Image restoration using total variation with overlapping group sparsity , 2013, Inf. Sci..

[21]  Tianxu Zhang,et al.  Blind spectrum reconstruction algorithm with L0-sparse representation , 2015 .

[22]  Hai Liu,et al.  Spectral Deconvolution and Feature Extraction With Robust Adaptive Tikhonov Regularization , 2013, IEEE Transactions on Instrumentation and Measurement.

[23]  Jianwen Sun,et al.  Adaptive total variation-based spectral deconvolution with the split Bregman method. , 2014, Applied optics.

[24]  Yan Li,et al.  High spectral specificity of local chemical components characterization with multichannel shift-excitation Raman spectroscopy , 2015, Scientific Reports.

[25]  P. Ziółkowski,et al.  Improvement of spatial resolution for local Seebeck coefficient measurements by deconvolution algorithm. , 2009, The Review of scientific instruments.

[26]  Ji Zhao,et al.  Non-rigid visible and infrared face registration via regularized Gaussian fields criterion , 2015, Pattern Recognit..

[27]  P. Jansson.  Deconvolution of images and spectra , 1997 .

[28]  Steven Kay,et al.  Autoregressive Modeling of Raman Spectra for Detection and Classification of Surface Chemicals , 2012, IEEE Transactions on Aerospace and Electronic Systems.

[29]  Qian Du,et al.  Robust Hyperspectral Image Classification by Multi-Layer Spatial-Spectral Sparse Representations , 2016, Remote. Sens..

[30]  Wesley E. Snyder,et al.  Restoration of ultrasonic NDE images , 1993, IEEE Trans. Ind. Electron..

[31]  Laura K. Guerrero,et al.  Associations between Nonverbal Behaviors and Initial Impressions of Instructor Competence and Course Content in Videotaped Distance Education Courses , 1998 .

[32]  Tao Huang,et al.  Infrared spectrum blind deconvolution algorithm via learned dictionaries and sparse representation. , 2016, Applied optics.

[33]  Limin Shao,et al.  Self-Weighted Correlation Coefficients and Their Application to Measure Spectral Similarity , 2009, Applied spectroscopy.

[34]  V. Lorenz-Fonfria,et al.  Maximum Entropy Deconvolution of Infrared Spectra: Use of a Novel Entropy Expression without Sign Restriction , 2005, Applied spectroscopy.

[35]  D. Mazur,et al.  Application of infrared spectroscopy in the identification of Ewing sarcoma: A preliminary report , 2017 .

[36]  Yi Hong Ong,et al.  Fast reconstruction of Raman spectra from narrow‐band measurements based on Wiener estimation , 2013 .

[37]  Hai Liu,et al.  A MAP-based algorithm for spectroscopic semi-blind deconvolution. , 2012, The Analyst.

[38]  Tianxu Zhang,et al.  Multi-order blind deconvolution algorithm with adaptive Tikhonov regularization for infrared spectroscopic data , 2015 .

[39]  Hu Zhu,et al.  Spectral semi-blind deconvolution with hybrid regularization , 2014 .

[40]  Hu Zhu,et al.  Spectral restoration using semi-blind deconvolution method with detail-preserving regularization , 2015 .

[41]  Hai Liu,et al.  Joint Baseline-Correction and Denoising for Raman Spectra , 2015, Applied spectroscopy.

[42]  O. Busto,et al.  Identification of olive oil sensory defects by multivariate analysis of mid infrared spectra. , 2015, Food chemistry.

[43]  Jiayi Ma,et al.  Infrared and visible image fusion via gradient transfer and total variation minimization , 2016, Inf. Fusion.

[44]  Hu Zhu,et al.  Spectral semi-blind deconvolution with least trimmed squares regularization , 2014 .

[45]  Jiaxiong Fang,et al.  Correction of FTIR acquired photodetector response spectra from mid-infrared to visible bands using onsite measured instrument function , 2018, Infrared Physics & Technology.

[46]  K. Runtz,et al.  Precise navigation using adaptive FIR filtering and time domain spectral estimation , 1994 .

[47]  Xiangyan Liu,et al.  An Infrared Scanning and Tracking System for Detecting Mid-Wave Infrared Spectral Characteristics of Moving Targets , 2014, Applied spectroscopy.

[48]  Boštjan Likar,et al.  Deconvolution in Acousto-Optical Tunable Filter Spectrometry , 2010, Applied spectroscopy.

[49]  Nicu Sebe,et al.  Facial expression recognition from video sequences: temporal and static modeling , 2003, Comput. Vis. Image Underst..

[50]  Sheng Zhong,et al.  Remote Sensing Image Stripe Noise Removal: From Image Decomposition Perspective , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[51]  Diane M. Christophel.  The relationships among teacher immediacy behaviors, student motivation, and learning , 1990 .

[52]  Xiaoyu Cui,et al.  Stepwise method based on Wiener estimation for spectral reconstruction in spectroscopic Raman imaging. , 2017, Optics express.

[53]  Sanya Liu,et al.  Spectral blind deconvolution with differential entropy regularization for infrared spectrum , 2015 .

[54]  Tianxu Zhang,et al.  Blind deconvolution using the similarity of multiscales regularization for infrared spectrum , 2015 .

[55]  Hai Liu,et al.  Parametric semi-blind deconvolution algorithm with Huber–Markov regularization for passive millimeter-wave images , 2013 .

[56]  M. Knapp,et al.  Nonverbal Communication: Issues and Appraisal , 1978 .

[57]  Hai Liu,et al.  Richardson–Lucy blind deconvolution of spectroscopic data with wavelet regularization , 2015 .

[58]  Houzhang Fang,et al.  Blind image deconvolution with spatially adaptive total variation regularization. , 2012, Optics letters.

[59]  Que Pei-wen,et al.  Sparsity enhancement for blind deconvolution of ultrasonic signals in nondestructive testing application. , 2008, The Review of scientific instruments.

[60]  VÍCTOR A. LÓRENZ-FONFRÍA,et al.  The Role and Selection of the Filter Function in Fourier Self-Deconvolution Revisited , 2009, Applied spectroscopy.

[61]  Dianne P. O'Leary,et al.  The Use of the L-Curve in the Regularization of Discrete Ill-Posed Problems , 1993, SIAM J. Sci. Comput..

[62]  Houzhang Fang,et al.  Semi-Blind Spectral Deconvolution with Adaptive Tikhonov Regularization , 2012, Applied spectroscopy.

[63]  Sanya Liu,et al.  Blind Poissonian reconstruction algorithm via curvelet regularization for an FTIR spectrometer. , 2018, Optics express.

[64]  Xi Chen,et al.  Intrinsic decomposition from a single spectral image. , 2017, Applied optics.

[65]  Jianbin Zheng,et al.  Comparison of wavelet transform and Fourier self-deconvolution (FSD) and wavelet FSD for curve fitting , 2000 .

[66]  Nelson M. Hoffman,et al.  Monte-Carlo error analysis in x-ray spectral deconvolution , 1985 .

[67]  Hai Liu,et al.  Fast Blind Instrument Function Estimation Method for Industrial Infrared Spectrometers , 2018, IEEE Transactions on Industrial Informatics.

[68]  Michael K. Ng,et al.  Blind Deconvolution Using Generalized Cross-Validation Approach to Regularization Parameter Estimation , 2011, IEEE Transactions on Image Processing.

[69]  Naixue Xiong,et al.  A general hybrid model for chaos robust synchronization and degradation reduction , 2015, Inf. Sci..

[70]  Fang Qian,et al.  A fully automated algorithm of baseline correction based on wavelet feature points and segment interpolation , 2017 .

[71]  Zhuowen Tu,et al.  Robust Point Matching via Vector Field Consensus , 2014, IEEE Transactions on Image Processing.