“Synergy effect” and its application in LED-multispectral imaging for improving image quality

Abstract This paper proposes and demonstrates the multi-wavelength “synergy effect” in light emitting diode (LED)-multispectral images obtained by frequency-division modulation, which can be used to improve the image quality of each waveband. Taking three wavelengths in LED-multispectral imaging as an example, the following experiment was designed: The three different center wavelength ( λ 1 , λ 2 and λ 3 ) LEDs modulated by sinusoidal signals with different frequencies are used as the light source to acquire image sequences. The time series composed of corresponding pixels of each frame are demodulated by Fourier transform (FT) to obtain the three single-waveband images I S λ 1 , I S λ 2 and I S λ 3 respectively. Similarly, at the same driving intensity, the above three LEDs are separately used as the light source to acquire image sequences, and three groups of images are obtained respectively. Then FT is used to perform demodulation in the same way and three single-waveband images I A λ 1 , I A λ 2 and I A λ 3 are obtained. By comparison, it can be concluded that the image quality of each waveband obtained by demodulating the three-wavelength images has been improved, which proves the “synergy effect” and provides a reference for improving the quality of multispectral images.

[1]  Gang Li,et al.  Employment of the appropriate range of sawtooth-shaped-function illumination intensity to improve the image quality , 2018, Optik (Stuttgart).

[2]  Yong He,et al.  SSC and pH for sweet assessment and maturity classification of harvested cherry fruit based on NIR hyperspectral imaging technology , 2018, Postharvest Biology and Technology.

[3]  Panagiotis Tsakanikas,et al.  Exploiting multispectral imaging for non-invasive contamination assessment and mapping of meat samples. , 2016, Talanta.

[4]  Lothar Lilge,et al.  A multi‐wavelength, laser‐based optical spectroscopy device for breast density and breast cancer risk pre‐screening , 2017, Journal of biophotonics.

[5]  Arab Ali Chérif,et al.  Multi-spectral palmprint recognition based on oriented multiscale log-Gabor filters , 2016, Neurocomputing.

[6]  Alessandro Torricelli,et al.  Seven-wavelength time-resolved optical mammography extending beyond 1000 nm for breast collagen quantification. , 2009, Optics express.

[7]  Kishor M. Bhurchandi,et al.  No-reference image quality assessment algorithms: A survey , 2015 .

[8]  Donald B. Hondongwa,et al.  A Review of the Pinned Photodiode for CCD and CMOS Image Sensors , 2014, IEEE Journal of the Electron Devices Society.

[9]  Bo Jin,et al.  Robust Tensor Preserving Projection for Multispectral Face Recognition , 2014 .

[10]  Ling Lin,et al.  [Single-trial estimation of dynamic spectrum]. , 2011, Guang pu xue yu guang pu fen xi = Guang pu.

[11]  Jon Y. Hardeberg,et al.  How Are LED Illumination Based Multispectral Imaging Systems Influenced by Different Factors? , 2014, ICISP.

[12]  Jeffrey M Roth,et al.  Ultrasensitive and high-dynamic-range two-photon absorption in a GaAs photomultiplier tube. , 2002, Optics letters.

[13]  Xiao-Yuan Jing,et al.  Multi-spectral low-rank structured dictionary learning for face recognition , 2016, Pattern Recognit..

[14]  Brian W. Pogue,et al.  Interpreting hemoglobin and water concentration, oxygen saturation, and scattering measured in vivo by near-infrared breast tomography , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[15]  Rosalind W. Picard,et al.  Non-contact, automated cardiac pulse measurements using video imaging and blind source separation , 2022 .

[16]  Junichi Nakamura,et al.  Image Sensors and Signal Processing for Digital Still Cameras , 2005 .

[17]  A. Yodh,et al.  Diffuse optics for tissue monitoring and tomography , 2010, Reports on progress in physics. Physical Society.

[18]  Ling Lin,et al.  Optimized lighting method of applying shaped-function signal for increasing the dynamic range of LED-multispectral imaging system. , 2018, The Review of scientific instruments.

[19]  Huimin Lu,et al.  Multispectral palmprint recognition using multiclass projection extreme learning machine and digital shearlet transform , 2014, Neural Computing and Applications.

[20]  R. Cubeddu,et al.  Clinical trial of time-resolved scanning optical mammography at 4 wavelengths between 683 and 975 nm. , 2004, Journal of biomedical optics.

[21]  Renfu Lu,et al.  Hyperspectral Imaging-Based Classification and Wavebands Selection for Internal Defect Detection of Pickling Cucumbers , 2013, Food and Bioprocess Technology.

[22]  F. Roy,et al.  CMOS image sensor: Process impact on dark current , 2014, 2014 IEEE International Reliability Physics Symposium.