# Multispectral Image Enhancement Based on the Dark Channel Prior and Bilateral Fractional Differential Model

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

- (1)
- By synthesizing the multiband information of multispectral remote sensing images, our algorithm obtains more accurate and clearer images than single-band remote sensing images.
- (2)
- A bilateral fractional differential model is firstly proposed and effectively improves the edge and textural details of multispectral images.
- (3)
- By expanding the bands and optimizing the transmittance of the dark channel prior model, an improved image with higher contrast and brightness is further obtained.

## 2. Related Work

- (1)
- The method of combining the dark channel prior algorithm with the fractional differential algorithm is applied to multispectral remote sensing image enhancement for the first time. Based on improving the overall brightness and detail characteristics of the image, the information of the spatial dimension and spectral dimension is combined to make full use of all wave bands of multispectral images.
- (2)
- Unlike the previous fractional differential algorithm, we propose a new fractional differential framework that enhances the edge and textural details of images. Considering the influence of the spatial distance on the pixel autocorrelation, we modify the fractional differential coefficients and propose the spatial domain fractional mask. Furthermore, according to the ability of the pixel similarity to judge the image edge, we propose the intensity domain fractional mask. Then the two above domains are fused to compose the bilateral fractional differential framework, and the enhanced images can be obtained by the framework. This framework can fully combine the information of spatial domain and intensity domain to maintain the details of the image in the smooth region and texture region and improve the image definition.
- (3)
- An improved dark channel prior algorithm, which extends the RGB channels to multiple channels and optimizes the transmittance through our proposed spatial domain fractional differentiation algorithm to make full use of the ground features of different spectral bands and enhance the overall brightness and contrast of an image, is proposed.

## 3. Proposed Method

#### 3.1. Dark Channel Prior

#### 3.2. Improved Dark Channel Prior

#### 3.3. G-L Fractional Differential Model

#### 3.4. Bilateral Fractional Differential Model

#### 3.4.1. Spatial Domain Fractional Differential Model

#### 3.4.2. Intensity Domain Fractional Differential Model

Algorithm 1. Multispectral image enhancement based on IDCP_BFD. | |

Input: A original multispectral image $R=\left\{{R}^{1},{R}^{2},\cdots ,{R}^{z}\right\}$ of the test dataset. | |

- (1)
**Improved Dark Channel Prior step:**- Through (13), the multispectral image
**R**is reverse as**I**to be used for the subsequent algorithm. - The enhanced multispectral image $J=\left\{{J}^{1},{J}^{2},\cdots ,{J}^{z}\right\}$ is generated via (14)–(22).
- (2)
**Bilateral Fractional Differential step:**
For λ = 1, to z doLet ${J}^{\lambda}$ be any bands in the multispectral image. | |

**Spatial domain Fractional Differential**- Enhance the ${J}^{\lambda}$ by using the spatial domain fractional differential normalized mask of Table 3.
- Implement histogram equalization and calculate the spatial domain enhanced image ${J}_{\alpha}^{\lambda \prime}$ via the (34).
| |

**Intensity domain Fractional Differential**- Enhance the ${J}^{\lambda}$ by using the intensity domain fractional differential normalized mask of Table 4.
- Perform contrast limited adaptive histogram equalization and calculate the intensity domain enhanced image ${J}_{\beta}^{\lambda \prime}$ via the (39).
| |

end for● Calculate the final enhanced image S by considering (40). Output: The enhanced result of the original image. |

## 4. Experimental Results and Analysis

_{k}represents the proportion of pixels with gray value k in the image. The larger the information entropy, the richer the information of the image.

#### 4.1. Dataset

#### 4.2. Parameter Analysis

#### 4.3. Comparative Experiments of Other Methods Based on Differential Filtering

#### 4.4. Comparative Experiments Based on Dark Channel Prior

#### 4.5. Comparative Experiments of Other Methods Based on Image Enhancement

## 5. Discussion

## 6. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Fu, X.; Wang, J.; Zeng, D.; Huang, Y.; Ding, X. Remote Sensing Image Enhancement Using Regularized-Histogram Equalization and DCT. IEEE Geosci. Remote Sens. Lett.
**2015**, 12, 2301–2305. [Google Scholar] [CrossRef] - Lu, X.; Li, X. Multiresolution Imaging. IEEE Trans. Cybern.
**2014**, 44, 149–160. [Google Scholar] [CrossRef] [PubMed] - Lu, X.; Wang, Y.; Yuan, Y. Graph-Regularized Low-Rank Representation for Destriping of Hyperspectral Images. IEEE Trans. Geosci. Remote Sens.
**2013**, 51, 4009–4018. [Google Scholar] [CrossRef] - Wang, J.; Yang, Y.; Chen, Y.; Han, Y. LighterGAN: An Illumination Enhancement Method for Urban UAV Imagery. Remote Sens.
**2021**, 13, 1371. [Google Scholar] [CrossRef] - Hagag, A.; Hassan, E.S.; Amin, M.; Abd El-Samie, F.E.; Fan, X. Satellite multispectral image compression based on removing sub-bands. Optik
**2017**, 131, 1023–1035. [Google Scholar] [CrossRef] - Iqbal, M.Z.; Ghafoor, A.; Siddiqui, A.M. Satellite Image Resolution Enhancement Using Dual-Tree Complex Wavelet Transform and Nonlocal Means. IEEE Geosci. Remote Sens. Lett.
**2013**, 10, 451–455. [Google Scholar] [CrossRef] - Lee, E.; Kim, S.; Kang, W.; Seo, D.; Paik, J. Contrast Enhancement Using Dominant Brightness Level Analysis and Adaptive Intensity Transformation for Remote Sensing Images. IEEE Geosci. Remote Sens. Lett.
**2013**, 10, 62–66. [Google Scholar] [CrossRef] - Pyka, K. Wavelet-Based Local Contrast Enhancement for Satellite, Aerial and Close Range Images. Remote Sens.
**2017**, 9, 25. [Google Scholar] [CrossRef] [Green Version] - Lisani, J.; Michel, J.; Morel, J.; Petro, A.B.; Sbert, C. An Inquiry on Contrast Enhancement Methods for Satellite Images. IEEE Trans. Geosci. Remote Sens.
**2016**, 54, 7044–7054. [Google Scholar] [CrossRef] - Liu, J.; Zhou, C.; Chen, P.; Kang, C. An Efficient Contrast Enhancement Method for Remote Sensing Images. IEEE Geosci. Remote Sens. Lett.
**2017**, 14, 1715–1719. [Google Scholar] [CrossRef] - Liu, C.; Sui, X.; Kuang, X.; Liu, Y.; Gu, G.; Chen, Q. Optimized Contrast Enhancement for Infrared Images Based on Global and Local Histogram Specification. Remote Sens.
**2019**, 11, 849. [Google Scholar] [CrossRef] [Green Version] - Febin, I.P.; Jidesh, P.; Bini, A.A. A Retinex-Based Variational Model for Enhancement and Restoration of Low-Contrast Remote-Sensed Images Corrupted by Shot Noise. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2020**, 13, 941–949. [Google Scholar] [CrossRef] - Jang, J.H.; Kim, S.D.; Ra, J.B. Enhancement of Optical Remote Sensing Images by Subband-Decomposed Multiscale Retinex With Hybrid Intensity Transfer Function. IEEE Geosci. Remote Sens. Lett.
**2011**, 8, 983–987. [Google Scholar] [CrossRef] - Ye, X.; Yang, H.; Li, C.; Jia, Y.; Li, P. A Gray Scale Correction Method for Side-Scan Sonar Images Based on Retinex. Remote Sens.
**2019**, 11, 1281. [Google Scholar] [CrossRef] [Green Version] - Song, M.; Qu, H.; Zhang, G.; Tao, S.; Jin, G. A Variational Model for Sea Image Enhancement. Remote Sens.
**2018**, 10, 1313. [Google Scholar] [CrossRef] [Green Version] - Chaudhuri, D.; Kushwaha, N.K.; Samal, A. Semi-Automated Road Detection From High Resolution Satellite Images by Directional Morphological Enhancement and Segmentation Techniques. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2012**, 5, 1538–1544. [Google Scholar] [CrossRef] - Bandeira, A. Random Laplacian Matrices and Convex Relaxations. Found. Comput. Math.
**2015**, 18, 345–379. [Google Scholar] [CrossRef] [Green Version] - Chen, S.; Zhao, F. The Adaptive Fractional Order Differential Model for Image Enhancement Based on Segmentation. Int. J. Pattern Recognit. Artif. Intell.
**2018**, 32, 1854005. [Google Scholar] [CrossRef] - Pu, Y.; Zhou, J.; Yuan, X. Fractional Differential Mask: A Fractional Differential-Based Approach for Multiscale Texture Enhancement. IEEE Trans. Image Processing
**2010**, 19, 491–511. [Google Scholar] [CrossRef] - Hemalatha, S.; Margret Anouncia, S. G-L fractional differential operator modified using auto-correlation function: Texture enhancement in images. Ain Shams Eng. J.
**2018**, 9, 1689–1704. [Google Scholar] [CrossRef] - Wadhwa, A.; Bhardwaj, A. Enhancement of MRI images of brain tumor using Grünwald Letnikov fractional differential mask. Multimed. Tools Appl.
**2020**, 79, 25379–25402. [Google Scholar] [CrossRef] - He, K.; Sun, J.; Tang, X. Single Image Haze Removal Using Dark Channel Prior. IEEE Trans. Pattern Anal. Mach. Intell.
**2011**, 33, 2341–2353. [Google Scholar] [CrossRef] [PubMed] - Xuan, D.; Guan, W.; Yi, P.; Weixin, L.; Jiangtao, W.; Wei, M.; Yao, L. Fast efficient algorithm for enhancement of low lighting video. In Proceedings of the 2011 IEEE International Conference on Multimedia and Expo, Barcelona, Spain, 11–15 July 2011; pp. 1–6. [Google Scholar]
- Caballero, R.; Berbey-Alvarez, A. Underwater Image Enhancement Using Dark Channel Prior and Image Opacity. In Proceedings of the 2019 7th International Engineering, Sciences and Technology Conference (IESTEC), Panama City, Panama, 9–11 October 2019; pp. 556–561. [Google Scholar]
- Im, J.; Yoon, I.; Hayes, M.H.; Paik, J. Dark channel prior-based spatially adaptive contrast enhancement for back lighting compensation. In Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, BC, Canada, 26–31 May 2013; pp. 2464–2468. [Google Scholar]
- Sonkar, P.K.; Raj, K. Single Image Dehazing Using Dark Channel Prior With Median Filter and Contrast Enhancement. In Proceedings of the 2020 IEEE International Conference for Innovation in Technology (INOCON), Bangluru, India, 6–8 November 2020; pp. 1–6. [Google Scholar]
- Yang, H.; Chen, P.; Huang, C.; Zhuang, Y.; Shiau, Y. Low Complexity Underwater Image Enhancement Based on Dark Channel Prior. In Proceedings of the 2011 Second International Conference on Innovations in Bio-inspired Computing and Applications, Shenzhen, China, 16–18 December 2011; pp. 17–20. [Google Scholar]
- Yang, H.; Wang, J. Color image contrast enhancement by co-occurrence histogram equalization and dark channel prior. In Proceedings of the 2010 3rd International Congress on Image and Signal Processing, Yantai, China, 16–18 October 2010; pp. 659–663. [Google Scholar]
- Wei, C.; Wang, W.; Yang, W.; Liu, J. Deep Retinex Decomposition for Low-Light Enhancement. arXiv
**2018**, arXiv:1808.04560v1. [Google Scholar] - Tian, L.; Du, Q.; Younan, N.; Kopriva, I. Multispectral image enhancement with extended offset-sparsity decomposition. In Proceedings of the 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Beijing, China, 10–15 July 2016; pp. 4383–4386. [Google Scholar]
- Bhandari, A.K.; Gadde, M.; Kumar, A.; Singh, G.K. Comparative analysis of different wavelet filters for low contrast and brightness enhancement of multispectral remote sensing images. In Proceedings of the 2012 International Conference on Machine Vision and Image Processing (MVIP), Coimbatore, India, 14–15 December 2012; pp. 81–86. [Google Scholar]
- Bhandari, A.K.; Kumar, A.; Singh, G.K. SVD Based Poor Contrast Improvement of Blurred Multispectral Remote Sensing Satellite Images. In Proceedings of the 2012 Third International Conference on Computer and Communication Technology, Allahabad, India, 23–25 November 2012; pp. 156–159. [Google Scholar]
- Suresh, S.; Das, D.; Lal, S. A Framework for Quality Enhancement of Multispectral Remote Sensing Images. In Proceedings of the 2017 Ninth International Conference on Advanced Computing (ICoAC), Chennai, India, 14–16 December 2017; pp. 9–14. [Google Scholar]
- Wang, M.; Zheng, X.; Feng, C. Color constancy enhancement for multi-spectral remote sensing images. In Proceedings of the 2013 IEEE International Geoscience and Remote Sensing Symposium—IGARSS, Melbourne, VIC, Australia, 21–26 July 2013; pp. 864–867. [Google Scholar]
- Lu, S.-l.; Zou, L.-j.; Shen, X.-h.; Wu, W.-y.; Zhang, W. Multi-spectral remote sensing image enhancement method based on PCA and IHS transformations. J. Zhejiang Univ. SCIENCE A
**2011**, 12, 453–460. [Google Scholar] [CrossRef] - Venkatakrishnamoorthy, T.; Reddy, G.U. Cloud enhancement of NOAA multispectral images by using independent component analysis and principal component analysis for sustainable systems. Comput. Electr. Eng.
**2019**, 74, 35–46. [Google Scholar] [CrossRef] - Mulla, A.; Baviskar, J.; Mohhamed, R.; Baviskar, A. Adaptive Band Specific Image Enhancement Scheme for Segmented Satellite Images. In Proceedings of the 2015 International Conference on Pervasive Computing (ICPC), Pune, India, 8–10 January 2015; pp. 1–5. [Google Scholar]
- Yang, C.; Lu, L.; Lin, H.; Guan, R.; Shi, X.; Liang, Y. A Fuzzy-Statistics-Based Principal Component Analysis (FS-PCA) Method for Multispectral Image Enhancement and Display. IEEE Trans. Geosci. Remote Sens.
**2008**, 46, 3937–3947. [Google Scholar] [CrossRef] - Cafagna, D. Fractional calculus: A mathematical tool from the past for present engineers [Past and present]. IEEE Ind. Electron. Mag.
**2007**, 1, 35–40. [Google Scholar] [CrossRef] - Matlob, M.A.; Jamali, Y. The Concepts and Applications of Fractional Order Differential Calculus in Modelling of Viscoelastic Systems: A primer. Crit. Rev. Biomed. Eng.
**2017**, 47, 249–276. [Google Scholar] [CrossRef] - Che, J.; Shi, Y.; Xiang, Y.; Ma, Y. The fractional differential enhancement of image texture features and its parallel processing optimization. In Proceedings of the 2012 5th International Congress on Image and Signal Processing, Chongqing, China, 16–18 October 2012; pp. 330–333. [Google Scholar]
- Zuiderveld, K. Contrast Limited Adaptive Histogram Equalization. In Graphics Gems; Heckbert, P.S., Ed.; Academic Press: San Diego, CA, USA, 1994; pp. 474–485. [Google Scholar]
- Guo, X.; Li, Y.; Ling, H. LIME: Low-Light Image Enhancement via Illumination Map Estimation. IEEE Trans. Image Processing
**2017**, 26, 982–993. [Google Scholar] [CrossRef] - Suresh, S.; Lal, S.; Reddy, C.S.; Kiran, M.S. A Novel Adaptive Cuckoo Search Algorithm for Contrast Enhancement of Satellite Images. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2017**, 10, 3665–3676. [Google Scholar] [CrossRef] - Luo, X.; Zeng, T.; Zeng, W.; Huang, J. Comparative analysis on landsat image enhancement using fractional and integral differential operators. Computing
**2020**, 102, 247–261. [Google Scholar] [CrossRef] - Demirel, H.; Ozcinar, C.; Anbarjafari, G. Satellite Image Contrast Enhancement Using Discrete Wavelet Transform and Singular Value Decomposition. IEEE Geosci. Remote Sens. Lett.
**2010**, 7, 333–337. [Google Scholar] [CrossRef]

**Figure 3.**Original multispectral data 1: (

**a**) Band 1; (

**b**) Band 2; (

**c**) Band 3; (

**d**) Band 4; (

**e**) Band 5; (

**f**) Band 6; (

**g**) Band 7; (

**h**) 7-bands composite image.

**Figure 4.**Original multispectral data 2: (

**a**) Band 1; (

**b**) Band 2; (

**c**) Band 3; (

**d**) Band 4; (

**e**) Band 5; (

**f**) Band 6; (

**g**) Band 7; (

**h**) 7-bands composite image.

**Figure 5.**Original multispectral data 3: (

**a**) Band 1; (

**b**) Band 2; (

**c**) Band 3; (

**d**) Band 4; (

**e**) Band 5; (

**f**) Band 6; (

**g**) Band 7; (

**h**) 7-bands composite image.

**Figure 6.**Information entropy with different parameters in the multispectral image dataset: (

**a**) brightness adjustment parameter $K$; (

**b**) fractional differential order $v$.

**Figure 10.**The histograms corresponding to enhanced data 1 by using different input bands combinations: (

**a**) original DCP of 3-Bands; (

**b**) IDCP of 3-Bands; (

**c**) IDCP of 4-Bands; (

**d**) IDCP of 5-Bands; (

**e**) IDCP of 6-Bands; (

**f**) IDCP of 7-Bands.

**Figure 11.**The quantitative results of different input bands combinations used in 30 multispectral data: (

**a**) entropy; (

**b**) average gradient.

**Figure 12.**Enhanced image of data 2 by using: (

**a**) Retinex-Net; (

**b**) LIME; (

**c**) ACSEA; (

**d**) Proposed IDCP_BFD.

$0$ | $\frac{{v}^{2}-v}{2}$ | $0$ | $0$ | $0$ | $0$ | $0$ | $0$ | $\frac{{v}^{2}-v}{2}$ |

$0$ | $-v$ | $0$ | $\frac{{v}^{2}-v}{2}$ | $-v$ | $1$ | $0$ | $-v$ | $0$ |

$0$ | $1$ | $0$ | $0$ | $0$ | $0$ | $1$ | $0$ | $0$ |

**Table 2.**The 3 × 3 G-L fractional differential masks and 3 × 3 spatial domain fractional differential masks in (

**A**) negative x-direction; (

**B**) negative y-direction; (

**C**) upper-right diagonal direction.

(A) | |||||

$0$ | ${a}_{1}$ | $0$ | $\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}{a}_{1}$ | $\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}{a}_{1}$ | $\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}{a}_{1}$ |

$0$ | ${a}_{0}$ | $0$ | $\frac{{w}_{2}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $\frac{{w}_{2}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ |

$0$ | $1$ | $0$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $1$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ |

(B) | |||||

$0$ | $0$ | $0$ | $\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}{a}_{1}$ | $\frac{{w}_{2}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ |

${a}_{1}$ | ${a}_{0}$ | $1$ | $\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}{a}_{1}$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $1$ |

$0$ | $0$ | $0$ | $\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}{a}_{1}$ | $\frac{{w}_{2}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ | $\frac{{w}_{1}}{3{w}_{1}+2{w}_{2}}{a}_{0}$ |

(C) | |||||

$0$ | $0$ | ${a}_{1}$ | $\frac{{w}_{3}{a}_{1}}{2{w}_{3}+2{w}_{4}+{w}_{5}}$ | $\frac{{w}_{4}{a}_{1}}{2{w}_{3}+2{w}_{4}+{w}_{5}}$ | $\frac{{w}_{5}{a}_{1}}{2{w}_{3}+2{w}_{4}+{w}_{5}}$ |

$0$ | ${a}_{0}$ | $0$ | $\frac{{w}_{1}{a}_{0}}{2{w}_{1}+{w}_{2}}$ | $\frac{{w}_{2}{a}_{0}}{2{w}_{1}+{w}_{2}}$ | $\frac{{w}_{4}{a}_{1}}{2{w}_{3}+2{w}_{4}+{w}_{5}}$ |

$1$ | $0$ | $0$ | $1$ | $\frac{{w}_{1}{a}_{0}}{2{w}_{1}+{w}_{2}}$ | $\frac{{w}_{3}{a}_{1}}{2{w}_{3}+2{w}_{4}+{w}_{5}}$ |

$\frac{{w}_{5}}{2{w}_{3}+2{w}_{4}+{w}_{5}}{a}_{2}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{2{w}_{3}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\frac{{w}_{5}}{2{w}_{3}+2{w}_{4}+{w}_{5}}{a}_{2}$ |

$\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{{w}_{2}}{2{w}_{1}+{w}_{2}}+\frac{2{w}_{2}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{2{w}_{1}}{2{w}_{1}+{w}_{2}}+\frac{3{w}_{1}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{{w}_{2}}{2{w}_{1}+{w}_{2}}+\frac{2{w}_{2}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ |

$\left(\frac{2{w}_{3}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{2{w}_{1}}{2{w}_{1}+{w}_{2}}+\frac{3{w}_{1}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $8{a}_{0}$ | $\left(\frac{2{w}_{1}}{2{w}_{1}+{w}_{2}}+\frac{3{w}_{1}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{2{w}_{3}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ |

$\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{{w}_{2}}{2{w}_{1}+{w}_{2}}+\frac{2{w}_{2}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{2{w}_{1}}{2{w}_{1}+{w}_{2}}+\frac{3{w}_{1}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{{w}_{2}}{2{w}_{1}+{w}_{2}}+\frac{2{w}_{2}}{3{w}_{1}+2{w}_{2}}\right){a}_{1}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ |

$\frac{{w}_{5}}{2{w}_{3}+2{w}_{4}+{w}_{5}}{a}_{2}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{2{w}_{3}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{3}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\left(\frac{{w}_{4}}{2{w}_{3}+2{w}_{4}+{w}_{5}}+\frac{{w}_{4}}{{w}_{3}+2{w}_{4}}\right){a}_{2}$ | $\frac{{w}_{5}}{2{w}_{3}+2{w}_{4}+{w}_{5}}{a}_{2}$ |

$\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(2,-2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(2,-1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(2,0\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(2,1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(2,2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ |

$\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(1,-2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(1,-1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(1,0\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(1,1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(1,2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ |

$\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(0,-2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(0,-1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | ${\mathrm{w}}_{p\left(0,0\right)}{a}_{0}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(0,1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(0,2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ |

$\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-1,-2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(-1,-1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(-1,0\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(-1,1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{1}\left(i,j\right)}}{a}_{1}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-1,2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ |

$\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-2,-2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-2,-1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-2,0\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-2,1\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ | $\frac{{\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(-2,2\right)}}{\sum {\mathrm{w}}_{p{\mathsf{\Omega}}_{2}\left(i,j\right)}}{a}_{2}$ |

Original Images | Metrics | ||||
---|---|---|---|---|---|

μ | C | H | AG | ||

Data 1 | Band1 | 113.73 | 50.06 | 5.54 | 4.33 |

Band2 | 59.25 | 21.06 | 5.06 | 2.91 | |

Band3 | 71.15 | 44.76 | 5.68 | 4.24 | |

Band4 | 72.48 | 41.51 | 5.47 | 4.07 | |

Band5 | 115.07 | 116.69 | 6.54 | 6.94 | |

Band6 | 190.00 | 1.47 | 4.81 | 0.73 | |

Band7 | 69.97 | 55.95 | 6.03 | 4.83 | |

Data 2 | Band1 | 115.15 | 50.06 | 5.73 | 4.37 |

Band2 | 59.21 | 19.70 | 5.19 | 2.85 | |

Band3 | 69.62 | 42.37 | 5.83 | 4.18 | |

Band4 | 74.93 | 47.92 | 5.91 | 4.36 | |

Band5 | 111.22 | 109.63 | 6.65 | 6.84 | |

Band6 | 187.27 | 2.01 | 5.07 | 0.85 | |

Band7 | 67.38 | 56.89 | 6.20 | 4.91 | |

Data 3 | Band1 | 88.51 | 25.78 | 5.50 | 3.30 |

Band2 | 45.43 | 11.72 | 4.91 | 2.19 | |

Band3 | 49.57 | 33.09 | 5.56 | 3.57 | |

Band4 | 80.82 | 95.66 | 6.39 | 6.04 | |

Band5 | 92.72 | 60.55 | 5.88 | 4.98 | |

Band6 | 171.47 | 2.62 | 5.62 | 0.95 | |

Band7 | 50.89 | 43.52 | 5.86 | 4.18 | |

Average in 30 multispectral data | Band1 | 97.84 | 34.04 | 5.61 | 3.79 |

Band2 | 51.12 | 15.94 | 5.07 | 2.61 | |

Band3 | 59.92 | 39.95 | 5.76 | 4.08 | |

Band4 | 76.11 | 52.51 | 5.92 | 4.48 | |

Band5 | 106.76 | 93.46 | 6.37 | 6.27 | |

Band6 | 174.41 | 2.24 | 5.12 | 0.86 | |

Band7 | 61.53 | 53.72 | 6.04 | 4.73 |

Test Images | Methods | Metrics | ||||
---|---|---|---|---|---|---|

μ | C | H | AG | Time(s) | ||

Data 1 | Sobel | 87.68 | 121.56 | 6.06 | 5.74 | 2.21 |

Laplacian | 96.13 | 2530.04 | 7.19 | 32.68 | 2.15 | |

Yipufei [19] | 96.13 | 2270.00 | 7.17 | 30.85 | 2.14 | |

MGL [20] | 96.13 | 2090.31 | 7.13 | 29.81 | 2.22 | |

Wadhwa et al. [21] | 96.02 | 95.72 | 5.97 | 5.47 | 7.24 | |

Proposed BFD | 114.69 | 3202.81 | 7.55 | 40.53 | 6.67 | |

Data 2 | Sobel | 97.10 | 108.57 | 6.12 | 5.67 | 2.24 |

Laplacian | 105.90 | 2311.94 | 7.22 | 31.72 | 2.09 | |

Yipufei [19] | 105.90 | 2063.72 | 7.20 | 29.92 | 2.21 | |

MGL [20] | 105.90 | 1899.01 | 7.15 | 28.90 | 2.18 | |

Wadhwa et al. [21] | 105.74 | 91.62 | 6.01 | 5.34 | 7.24 | |

Proposed BFD | 120.37 | 3090.68 | 7.56 | 40.22 | 6.72 | |

Data 3 | Sobel | 75.20 | 102.33 | 5.80 | 4.81 | 2.23 |

Laplacian | 83.70 | 1372.77 | 6.82 | 24.18 | 2.16 | |

Yipufei [19] | 83.70 | 1227.02 | 6.79 | 22.75 | 2.21 | |

MGL [20] | 83.70 | 1145.76 | 6.76 | 22.19 | 2.14 | |

Wadhwa et al. [21] | 83.54 | 54.96 | 5.69 | 4.23 | 7.15 | |

Proposed BFD | 106.19 | 2160.91 | 7.23 | 32.92 | 6.65 | |

Average in 30 multispectral data | Sobel | 84.78 | 117.11 | 6.02 | 5.73 | 2.24 |

Laplacian | 93.75 | 2054.57 | 7.11 | 30.07 | 2.16 | |

Yipufei [19] | 93.75 | 1842.40 | 7.09 | 28.41 | 2.21 | |

MGL [20] | 93.75 | 1703.19 | 7.05 | 27.52 | 2.16 | |

Wadhwa et al. [21] | 93.56 | 89.58 | 5.98 | 5.28 | 7.22 | |

Proposed BFD | 112.21 | 2866.97 | 7.49 | 38.71 | 6.67 |

Test Images | Methods | Metrics | ||||
---|---|---|---|---|---|---|

μ | C | H | AG | Time(s) | ||

Data 2 | Retinex-Net | 160.20 | 51.46 | 4.95 | 3.83 | 2.82 |

LIME | 199.94 | 151.33 | 6.39 | 7.88 | 2.43 | |

ACSEA | 149.04 | 64.35 | 6.27 | 5.22 | 434.29 | |

Proposed | 120.37 | 3090.68 | 7.56 | 40.22 | 6.72 | |

Average in 30 multispectral data | Retinex-Net | 159.43 | 54.43 | 4.87 | 3.83 | 2.73 |

LIME | 196.01 | 146.38 | 6.36 | 7.76 | 2.29 | |

ACSEA | 161.09 | 99.05 | 6.34 | 6.52 | 450.71 | |

Proposed | 112.21 | 2866.97 | 7.49 | 38.71 | 6.67 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 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 (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Chen, W.; Jia, Z.; Yang, J.; Kasabov, N.K.
Multispectral Image Enhancement Based on the Dark Channel Prior and Bilateral Fractional Differential Model. *Remote Sens.* **2022**, *14*, 233.
https://doi.org/10.3390/rs14010233

**AMA Style**

Chen W, Jia Z, Yang J, Kasabov NK.
Multispectral Image Enhancement Based on the Dark Channel Prior and Bilateral Fractional Differential Model. *Remote Sensing*. 2022; 14(1):233.
https://doi.org/10.3390/rs14010233

**Chicago/Turabian Style**

Chen, Weijie, Zhenhong Jia, Jie Yang, and Nikola K. Kasabov.
2022. "Multispectral Image Enhancement Based on the Dark Channel Prior and Bilateral Fractional Differential Model" *Remote Sensing* 14, no. 1: 233.
https://doi.org/10.3390/rs14010233