# Hyperspectral Super-Resolution Technique Using Histogram Matching and Endmember Optimization

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Problem Formulation

**Z**are described by an additive mixture:

**E**act as non-orthogonal bases to represent

**Z**in a lower dimensional space ${\mathbb{R}}^{\mathrm{p}}$ with rank {

**Z**} ≤ p.

**I**is a spectrally downsampled version of

**Z**:

**S**of the hyperspectral camera and the spectral response function

**R**of the conventional camera form part of the camera specifications and are assumed to be known. The spectral response function of the Nikon D700 offered in the work of Jun et al. [29] is used as

**R**for this research.

**H**, is defined as:

**H**are defined not as a factorization of

**EA**and spatial downsampling operator, but as another form of matrix, is that the pixel location of the super-resolution result is not the same as that of

**H**.

**L**is defined as:

**R**and the low-resolution HSI,

**H**. The constraint condition of this paper follows that of Lanaras et al. [28]. The main assumption of the constraint is that the endmembers are reflectance spectra of individual materials, and the abundances are proportions of those endmembers. As a consequence, the factorization is subject to the following constraints:

**A**. Here 1 denotes a vector of 1s compatible with the dimensions of

**A**.

## 3. Proposed Solution

#### 3.1. Overall Scheme

**L**, obtained from an HSI. The histogram-matching process is a minimization of grayscale transformation T in the following equation:

**L**’s histogram and ${c}_{1}$ is the cumulative distribution function of

**I**’s histogram for all intensities k on a gray scale. Since the histogram equalization is defined on a gray scale, it has to be iteratively performed on each channel of a RGB image. T is a function that finds the index on ${c}_{0}$ that has the value most similar to the value of ${c}_{1}$ at a particular index $g$ on grayscale. After the function T is defined for g on all intensities on a gray scale, histogram equalization is performed by finding and mapping a value corresponding to

**L**in the input image

**I**, using T. In Step 3 of the proposed algorithm, an estimate of

**Z**, or equivalently

**E**and

**A**, is needed. From the given super-resolution problem, the following constrained least-squares problem is formulated as:

**A**and constrained to non-negative values.

#### 3.2. Overall Algorithm and Implementation

**H**and

**I**, which are low-resolution HSI and high-resolution RGB images, respectively. Because

**Z**will be reconstructed using endmember abundance with the same resolution as

**I**, the resolution of

**I**has to be an integer multiple of the resolution of

**H**with upsampling rate

**S**. Additionally, RGB camera sensitivity

**C**is required to reconstruct an RGB image from

**H**.

**L**from

**H**, using

**C**. The camera sensitivity for each RGB channel is multiplied by the spectral information of

**H**of each pixel, and normalized into an 8-bit-precision RGB image

**L**, for further histogram equalization with

**I**. Then, the histogram of

**I**is matched with that of

**L**, so that

**I**has the same color distribution as

**L**. The histogram equalization is performed using a MATLAB built-in function [30]. The next step is to find the initial values, ${E}^{\left(0\right)}$ and ${A}^{\left(0\right)}$, to optimize

**Z**, which consists of the endmember vector

**E**and the per-pixel abundance vector

**A**. Simplex identification via split augmented Lagrangian (SISAL) [31] initializes endmember

**E**and sparse unmixing by variable splitting and augmented Lagrangian (SUnSAL) [32] initializes

**A’**, respectively. SISAL is an algorithm for unsupervised hyperspectral linear unmixing and finds the minimum volume simplex containing the hyperspectral vectors, by augmented Lagrangian optimizations. SUnSAL is an eigen decomposition-based hyperspectral unmixing algorithm. The MATLAB code for SISAL and SUnSAL is available at the author’s webpage [33]. The low-resolution per-pixel abundance

**A’**is upsampled with

**S**and will be used as the initial point of the low-resolution step. The optimization is performed with a projected gradient method for 7a. The equation for the projected gradient method is:

## 4. Experiment

#### 4.1. Baseline Study for Spatial Information Mismatch

**H**and

**I**. The method of Lanaras et al. implicitly assumes that there is no pixel mismatch or color difference between high resolution RGB image and low resolution HSI, and the effects of this assumption are investigated using images with pixel mismatch and color difference. A public hyperspectral database, called the Harvard dataset [34], was used for the evaluation of the proposed algorithm. Because the purpose of the Harvard dataset [34] is to establish the basic statistical structure of HSIs of real-world scenes, the dataset is in accord with the condition in which the proposed algorithm will be used. The Harvard dataset [34] has 50 indoor and outdoor images recorded under daylight illumination, and 27 images recorded under artificial or mixed illumination. The spatial resolution of the images is 1392 × 1040 pixels, with 31 spectral bands of width 10 nm, from 420 to 720 nm. The original HSIs are used as ground truth for the evaluation.

**Z**, with respect to the ground truth image $\widehat{\mathrm{Z}}$:

#### 4.2. Proposed Method Evaluation

## 5. Conclusions

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Adão, T.; Hruška, J.; Pádua, L.; Bessa, J.; Peres, E.; Morais, R.; Sousa, J. Hyperspectral Imaging: A Review on UAV-Based Sensors, Data Processing and Applications for Agriculture and Forestry. Remote Sens.
**2017**, 9, 1110. [Google Scholar] [CrossRef] - Mahesh, S.; Jayas, D.S.; Paliwal, J.; White, N.D.G. Hyperspectral imaging to classify and monitor quality of agricultural materials. J. Stored Prod. Res.
**2015**, 61, 17–26. [Google Scholar] [CrossRef] - Brackx, M.; Van Wittenberghe, S.; Verhelst, J.; Scheunders, P.; Samson, R. Hyperspectral leaf reflectance of Carpinus betulus L. saplings for urban air quality estimation. Environ. Pollut.
**2017**, 220, 159–167. [Google Scholar] [CrossRef] [PubMed] - Balzarolo, M.; Peñuelas, J.; Filella, I.; Portillo-Estrada, M.; Ceulemans, R. Assessing Ecosystem Isoprene Emissions by Hyperspectral Remote Sensing. Remote Sens.
**2018**, 10, 1086. [Google Scholar] [CrossRef] - Fossi, A.P.; Ferrec, Y.; Roux, N.; D’almeida, O.; Guerineau, N.; Sauer, H. Miniature and cooled hyperspectral camera for outdoor surveillance applications in the mid-infrared. Opt. Lett.
**2016**, 41, 1901. [Google Scholar] [CrossRef] [PubMed] - Vanegas, F.; Bratanov, D.; Powell, K.; Weiss, J.; Gonzalez, F. A Novel Methodology for Improving Plant Pest Surveillance in Vineyards and Crops Using UAV-Based Hyperspectral and Spatial Data. Sensors
**2018**, 18, 260. [Google Scholar] [CrossRef] [PubMed] - Ray, A.; Kopelman, R.; Chon, B.; Briggman, K.; Hwang, J. Scattering based hyperspectral imaging of plasmonic nanoplate clusters towards biomedical applications. J. Biophotonics
**2016**, 9, 721–729. [Google Scholar] [CrossRef] - Weijtmans, P.J.C.; Shan, C.; Tan, T.; Brouwer de Koning, S.G.; Ruers, T.J.M. A Dual Stream Network for Tumor Detection in Hyperspectral Images. In Proceedings of the 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), Venice, Italy, 8–11 April 2019; pp. 1256–1259. [Google Scholar]
- Mielke, C.; Rogass, C.; Boesche, N.; Segl, K.; Altenberger, U. EnGeoMAP 2.0—Automated Hyperspectral Mineral Identification for the German EnMAP Space Mission. Remote Sens.
**2016**, 8, 127. [Google Scholar] [CrossRef] - Scafutto, R.D.M.; de Souza Filho, C.R.; Rivard, B. Characterization of mineral substrates impregnated with crude oils using proximal infrared hyperspectral imaging. Remote Sens. Environ.
**2016**, 179, 116–130. [Google Scholar] [CrossRef] - Wang, J.; Hu, P.; Lu, Q.; Shu, R. An airborne pushbroom hyperspectral imager with wide field of view. Chin. Opt. Lett.
**2005**, 3, 689–691. [Google Scholar] - Johnson, W.R.; Wilson, D.W.; Fink, W.; Humayun, M.; Bearman, G. Snapshot hyperspectral imaging in ophthalmology. J. Biomed. Opt.
**2007**, 12, 014036. [Google Scholar] [CrossRef] [PubMed] - Wei, Q.; Bioucas-Dias, J.; Dobigeon, N.; Tourneret, J.Y. Hyperspectral and Multispectral Image Fusion Based on a Sparse Representation. IEEE Trans. Geosci. Remote Sens.
**2015**, 53, 3658–3668. [Google Scholar] [CrossRef][Green Version] - Wang, Q.; Shen, Y.; Zhang, Y.; Zhang, J.Q. A quantitative method for evaluating the performances of hyperspectral image fusion. IEEE Trans. Instrum. Meas.
**2003**, 52, 1041–1047. [Google Scholar] [CrossRef] - Wilson, T.A. Perceptual-based hyperspectral image fusion using multiresolution analysis. Opt. Eng.
**1995**, 34, 3154. [Google Scholar] [CrossRef] - Akgun, T.; Altunbasak, Y.; Mersereau, R.M. Super-resolution reconstruction of hyperspectral images. IEEE Trans. Image Process.
**2005**, 14, 1860–1875. [Google Scholar] [CrossRef] - Zhao, Y.; Yang, J.; Zhang, Q.; Song, L.; Cheng, Y.; Pan, Q. Hyperspectral imagery super-resolution by sparse representation and spectral regularization. EURASIP J. Adv. Signal Process.
**2011**, 2011. [Google Scholar] [CrossRef] - Kwon, H.; Tai, Y.W. RGB-guided hyperspectral image upsampling. In Proceedings of the the IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December 2015; pp. 307–315. [Google Scholar]
- Mitani, Y.; Hamamoto, Y. A Consideration of Pan-Sharpen Images by HSI Transformation Approach. In Proceedings of the SICE Annual Conference 2010, Taipei, Taiwan, 8–21 August 2010; pp. 1283–1284. [Google Scholar]
- Aiazzi, B.; Baronti, S.; Selva, M. Improving component substitution pansharpening through multivariate regression of MS+Pan data. IEEE Trans. Geosci. Remote Sens.
**2007**, 45, 3230–3239. [Google Scholar] [CrossRef] - Chen, Y.; Zhang, G. A Pan-Sharpening Method Based on Evolutionary Optimization and IHS Transformation. Math. Probl. Eng.
**2017**, 2017, 1–8. [Google Scholar] [CrossRef][Green Version] - Li, Y.; Xie, W.; Li, H. Hyperspectral image reconstruction by deep convolutional neural network for classification. Pattern Recognit.
**2017**, 63, 371–383. [Google Scholar] [CrossRef] - Yuan, Y.; Zheng, X.; Lu, X. Hyperspectral Image Superresolution by Transfer Learning. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2017**, 10, 1963–1974. [Google Scholar] [CrossRef] - Wycoff, E.; Chan, T.H.; Jia, K.; Ma, W.K.; Ma, Y. A non-negative sparse promoting algorithm for high resolution hyperspectral imaging. In Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, BC, Canada, 26–31 May 2013; pp. 1409–1413. [Google Scholar]
- Huang, B.; Member, A.; Song, H.; Cui, H.; Peng, J.; Xu, Z. Spatial and Spectral Image Fusion Using Sparse Matrix Factorization. IEEE Trans. Geosci. Remote Sens.
**2013**, 52, 1693–1703. [Google Scholar] [CrossRef] - Simoes, M.; Bioucas-Dias, J.; Almeida, L.B.; Chanussot, J. A convex formulation for hyperspectral image superresolution via subspace-based regularization. IEEE Trans. Geosci. Remote Sens.
**2015**, 53, 3373–3388. [Google Scholar] [CrossRef] - Fang, L.; Zhuo, H.; Li, S. Neurocomputing Super-resolution of hyperspectral image via superpixel-based sparse representation. Neurocomputing
**2018**, 273, 171–177. [Google Scholar] [CrossRef] - Lanaras, C.; Baltsavias, E.; Schindler, K. Hyperspectral super-resolution by coupled spectral unmixing. In Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December 2015; pp. 3586–3594. [Google Scholar]
- Jiang, J.; Liu, D.; Gu, J.; Susstrunk, S. What is the space of spectral sensitivity functions for digital color cameras? In Proceedings of the IEEE Workshop on Applications of Computer Vision, Tampa, FL, USA, 5–17 January 2013; pp. 168–179. [Google Scholar]
- Enhance Contrast Using Histogram Equalization - MATLAB Histeq. Available online: https://www.mathworks.com/help/images/ref/histeq.html (accessed on 4 September 2019).
- Bioucas-Dias, J.M. A variable splitting augmented Lagrangian approach to linear spectral unmixing. In Proceedings of the 2009 First Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, Grenoble, France, 26–28 August 2009; pp. 1–4. [Google Scholar]
- Iordache, M.; Bioucas-dias, J.M.; Plaza, A.; Member, S. Sparse Unmixing of Hyperspectral Data. IEEE Trans. Geosci. Remote Sens.
**2011**, 49, 2014–2039. [Google Scholar] [CrossRef][Green Version] - Jose, M. Bioucas Dias Welcome to the Home Page of Jose. Available online: http://www.lx.it.pt/~bioucas/ (accessed on 4 September 2019).
- Chakrabarti, A.; Zickler, T. Statistics of Real-World Hyperspectral Images. In Proceedings of the CVPR 2011, Colorado Springs, CO, USA, 20–25 June 2011; pp. 193–200. [Google Scholar]
- Kruse, F.A.; Lefkoff, A.B.; Boardman, J.W.; Heidebrecht, K.B.; Shapiro, A.T.; Barloon, P.J.; Goetz, A.F.H. The spectral image processing system (SIPS)-interactive visualization and analysis of imaging spectrometer data. Remote Sens. Environ.
**1993**, 44, 145–163. [Google Scholar] [CrossRef] - Akhtar, N.; Shafait, F.; Mian, A. Bayesian sparse representation for hyperspectral image super resolution. In Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, 7–12 June 2015; pp. 3631–3640. [Google Scholar]
- Akhtar, N.; Shafait, F.; Mian, A. Sparse Spatio-spectral Representation for Hyperspectral Image Super-resolution. In European Conference on Computer Vision; Springer: Cham, Switzerland, 2014; pp. 63–78. [Google Scholar][Green Version]
- Yokoya, N.; Yairi, T.; Iwasaki, A. Coupled nonnegative matrix factorization unmixing for hyperspectral and multispectral data fusion. IEEE Trans. Geosci. Remote Sens.
**2012**, 50, 528–537. [Google Scholar] [CrossRef]

**Figure 1.**Overall framework of the proposed solution for hyperspectral super-resolution when spatial information of an Red-Green-Blue (RGB) image and a hyperspectral image (HSI) is mismatched.

**Figure 2.**Example of pixel mismatch environment created by image processing: (

**a**) Original RGB image reconstructed from an HSI from the Harvard dataset [34] with camera sensitivity for a Nikon D700, (

**b**) mismatched image where 50 pixels on the upper and left sides are cut off; (

**c**) color mismatched RGB image reconstructed using a Nokia N900.

**Figure 3.**RGB camera sensitivity for RGB image reconstruction: camera sensitivity for (

**a**) a Nokia N900 and (

**b**) a Nikon D700.

**Figure 4.**Example of a per-pixel RMSE (root–mean–square error) image from the Harvard dataset [34]: (

**a**) Image where translation is applied by cutting off and (

**b**) an image in which the histogram is not matched with that of an RGB image.

**Figure 5.**Example of pixel mismatch representation using (

**a**) cutting off (${P}_{n}$ indicates the number of pixels cut off from the left and upper sides) and (

**b**) shear transformation.

**Figure 6.**Example of per-pixel RMSE images and reconstructed RGB images of high-resolution HSIs: (

**a**) Per-pixel RMSE image where a shear transformation with ${C}_{x}=$ 0.3 is applied, (

**b**) a per-pixel RMSE image where translation with ${P}_{n}=$ 80 is applied, (

**c**) RGB image an where shear transformation with ${C}_{x}=$ 0.3 is applied, and (

**d**) RGB image where translation with ${P}_{n}=$80 is applied.

Requires:H (low-resolution hyperspectral image)I (high-resolution RGB image)C (RGB camera sensitivity)S (upsampling rate) |

Reconstruct L by applying C to HMatch histogram of I to that of LInitialize ${E}^{\left(0\right)}$ with SISAL and ${A}^{\prime \left(0\right)}$ with SUnSAL from HInitialize ${A}^{\left(0\right)}$ by upsampling ${A}^{\prime \left(0\right)}$ with Sk ← 0 while not converged dok ← k + 1 ${A}^{\left(k\right)}$ ← ${A}^{\left(k-1\right)}S$; Estimate ${E}^{\left(k\right)}$ with (8a) and (8b) end whilereturn $Z={E}^{\left(k\right)}{A}^{\left(k\right)}$ |

Method | RMSE | SAM | ||
---|---|---|---|---|

Average | Median | Average | Median | |

Original results for the entire dataset | 1.7 | 1.5 | 2.9 | 2.7 |

Translation by cutting off | 11.6 | 10.36 | 7.72 | 8.11 |

Histogram not matched | 8.88 | 8.55 | 8.06 | 7.15 |

Method of Field of View Mismatch | Lanaras et al. [28] | Proposed Method | |||||||
---|---|---|---|---|---|---|---|---|---|

RMSE | SAM | RMSE | SAM | ||||||

Average | Median | Average | Median | Average | Median | Average | Median | ||

Without Transformation | 2.68 | 2.11 | 5.58 | 5.47 | 2.68 | 2.57 | 6.69 | 5.32 | |

Shear Transformation Using Affine Matrix | ${C}_{x}=$0.1 | 9.79 | 7.70 | 7.16 | 7.35 | 3.64 | 2.96 | 6.73 | 6.03 |

${C}_{x}=$0.2 | 11.76 | 9.33 | 7.88 | 8.00 | 4.4 | 3.56 | 6.9 | 6.26 | |

${C}_{x}=$0.3 | 12.89 | 10.62 | 8.37 | 8.68 | 5.19 | 4.30 | 7.2 | 6.18 | |

Translation by Cutting off Upper and Left Sides | ${P}_{n}=20$ | 9.12 | 6.93 | 6.71 | 6.83 | 3.56 | 2.87 | 6.65 | 5.65 |

${P}_{n}=40$ | 10.95 | 8.17 | 7.31 | 7.40 | 4.25 | 3.27 | 6.86 | 5.80 | |

${P}_{n}=60$ | 12.02 | 9.27 | 7.80 | 8.03 | 4.93 | 3.68 | 7.08 | 5.98 | |

${P}_{n}=80$ | 12.81 | 9.99 | 8.24 | 8.52 | 5.60 | 4.13 | 7.24 | 6.20 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Kim, B.; Cho, S.
Hyperspectral Super-Resolution Technique Using Histogram Matching and Endmember Optimization. *Appl. Sci.* **2019**, *9*, 4444.
https://doi.org/10.3390/app9204444

**AMA Style**

Kim B, Cho S.
Hyperspectral Super-Resolution Technique Using Histogram Matching and Endmember Optimization. *Applied Sciences*. 2019; 9(20):4444.
https://doi.org/10.3390/app9204444

**Chicago/Turabian Style**

Kim, Byunghyun, and Soojin Cho.
2019. "Hyperspectral Super-Resolution Technique Using Histogram Matching and Endmember Optimization" *Applied Sciences* 9, no. 20: 4444.
https://doi.org/10.3390/app9204444