# A Study for Texture Feature Extraction of High-Resolution Satellite Images Based on a Direction Measure and Gray Level Co-Occurrence Matrix Fusion Algorithm

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Gray Level Co-Occurrence Matrix and Two-Order Statistical Parameters

_{1}distanced from a fixed spatial location relationship (size and direction) to another gray value g

_{2}. Assume that f(i, j) is a 2D gray-scale image, where S is the set of pixels with a certain spatial relation in the region and P refers to the GLCM, which can be expressed as

#### 2.2. Direction Measure

#### 2.3. Fusion of Direction Measure and Gray Level Co-Occurrence Matrix

#### 2.3.1. Weight Factor of Fusion Feature

_{j}is the gray value of point j. Angle θ = 45° × t, (t = 0, 1, 2, 3, which makes c = 3). ${W}_{tj}$ is the weight of X

_{j}that belongs to the direction measure of d(2t + 1)in the θ direction. The algorithm for determining ${W}_{tj}$ is expressed as

#### 2.3.2. Fusion Feature Calculation

_{θ}. The fusion feature of Q is calculated as

#### 2.3.3. Steps of Fusion Feature

## 3. Experimental Results and Analysis

#### 3.1. High-Resolution Remote Sensing Image Classification

#### 3.1.1. GaoFen-2 Data

#### 3.1.2. QuickBird Data

#### 3.1.3. GeoEye-1 Data

#### 3.2. Direction Measure of Image

#### 3.3. Image Classification with or without Distinct Directionality

## 4. Discussion

## Acknowledgments

## Author Contributions

## Conflicts of Interest

## References

- Kitada, K.; Fukuyama, K. Land-use and land-cover mapping using a gradable classification method. Remote Sens.
**2012**, 4, 1544–1558. [Google Scholar] [CrossRef] - Racoviteanu, A.; Williams, M.W. Decision tree and texture analysis for mapping debris-covered glaciers in the Kangchenjunga area, Eastern Himalaya. Remote Sens.
**2012**, 4, 3078–3109. [Google Scholar] [CrossRef] - Myint, S.W.; Lam, N.S.N.; Tyler, J.M. Wavelets for urban spatial feature discrimination: Comparisons with fractal, spatial autocorrelation, and spatial co-occurrence approaches. Photogramm. Eng. Remote Sens.
**2004**, 70, 803–812. [Google Scholar] [CrossRef] - Haralick, R.M.; Shanmugam, K.; Dinstein, I. Textural Features for Image Classification. IEEE Trans. Syst. Man Cybern.
**1973**, SCM-3, 610–621. [Google Scholar] [CrossRef] - Davis, L.S. A survey of edge detection techniques. Comput. Graph. Image Process.
**1975**, 4, 248–270. [Google Scholar] [CrossRef] - Huang, D.S.; Wunsch, D.C.; Levine, D.S.; Jo, K.H. Advanced Intelligent Computing Theories and Applications. With Aspects of Theoretical and Methodological Issues (ICIC). In Proceedings of the Fourth International Conference on Intelligent Computing, Shanghai, China, 15–18 September 2008. [Google Scholar]
- Zhang, Y. Optimisation of building detection in satellite images by combining multispectral classification and texture filtering. ISPRS J. Photogramm. Remote Sens.
**1999**, 54, 50–60. [Google Scholar] [CrossRef] - Sahoo, M. Biomedical Image Fusion and Segmentation using GLCM. In Proceedings of the 2nd National Conference-Computing, Communication and Sensor Network (CCSN), Orissa, India, 29–30 October 2011; pp. 34–39. [Google Scholar]
- Kekre, H.; Thepade, S.D.; Sarode, T.K.; Suryawanshi, V. Image retrieval using texture features extracted from GLCM, LBG and KPE. Int. J. Comput. Theory Eng.
**2010**, 2, 695–700. [Google Scholar] [CrossRef] - De Almeida, C.W.D.; De Souza, R.M.C.R.; Candeias, A.L.B. Texture Classification Based on Co-Occurrence Matrix and Self-Organizing Map. In Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, Istanbul, Turkey, 10–13 October 2010; pp. 2487–2491. [Google Scholar]
- Srinivasan, G.N.; Shobha, G. Segmrntation Techniques for ATDR. NAUN Int. J. Comput.
**2008**, 2, 165–171. [Google Scholar] - Tuceryan, M. Moment-based texture segmentation. Pattern Recognit. Lett.
**1994**, 15, 659–668. [Google Scholar] [CrossRef] - Liu, J.; Cheng, Y.; Sun, J. Modified FCM SAR image segmentation method based on GMCL feature. Comput. Eng. Des.
**2012**, 33, 3502–3506. [Google Scholar] - Pacifici, F.; Chini, M.; Emery, W.J. A neural network approach using multi-scale textural metrics from very high-resolution panchromatic imagery for urban land-use classification. Remote Sens. Environ.
**2009**, 113, 1276–1292. [Google Scholar] [CrossRef] - Mukherjee, G.; Chatterjee, A.; Tudu, B. Study on the Potential of Combined GLCM Features towards Medicinal Plant Classification. In Proceedings of the 2nd International Conference on Control, Instrumentation, Energy & Communication (CIEC), Kolkata, India, 28–30 January 2016; pp. 98–102. [Google Scholar]
- Li, Z.; Zhu, G.; Dong, T. Application of GLCM—Based texture features to remote sensing image classification. Geol. Explor.
**2011**, 47, 456–461. [Google Scholar] - Huang, X.; Yang, W. Classification of remotely sensed imagery according to the combination of gray scale and texture features based on the dynamic windows. J. Geomat. Sci. Technol.
**2015**, 32, 277–281. [Google Scholar] - Rao, C.N.; Sastry, S.S.; Mallika, K.; Tiong, H.S.; Mahalakshmi, K. Co-occurrence matrix and its statistical features as an approach for identification of phase transitions of mesogens. Int. J. Innov. Res. Sci. Eng. Technol.
**2013**, 2, 4531–4538. [Google Scholar] - Teng, X.; Song, S.; Zhan, Y. Statistical Class Feature in Texture Analysis of Remote Sensing Imagery. Adv. Environ. Sci. Eng.
**2012**, 518–523, 5749–5753. [Google Scholar] [CrossRef] - Liu, F.; Picard, R.W. Periodicity, directionality and randomness: Wold features for image modeling and retrieval. IEEE Trans. Pattern Anal. Mach. Intell.
**1996**, 18, 722–733. [Google Scholar] [CrossRef] - Chen, B.; Zhang, Y.; Chen, L. RS Image classification based on SVM method with texture. Eng. Surv. Mapp.
**2007**, 16, 23–27. [Google Scholar] - Zhao, L.; He, H.; Yin, Z. Texture direction adaptive image interpolation based on Curvelet transform. J. Optoelectron. Laser
**2012**, 23, 799–804. [Google Scholar] [CrossRef] - Zlatopolsky, A.A. Description of texture orientation in remote sensing data using computer program LESSA. Comput. Geosci.
**1997**, 23, 45–62. [Google Scholar] [CrossRef] - Karu, K.; Jain, A.K.; Bolle, R.M. Is there any texture in the image? Pattern Recognit.
**1996**, 29, 1437–1446. [Google Scholar] [CrossRef] - Puissant, A.; Hirsch, J.; Weber, C. The utility of texture analysis to improve per-pixel classification for high to very high spatial resolution imagery. Int. J. Remote Sens.
**2005**, 26, 733–745. [Google Scholar] [CrossRef] - Zheng, S.; Zheng, J.; Shi, M.; Guo, B.; Sen, B.; Sun, Z.; Jia, X.; Li, X. Classification of cultivated Chinese medicinal plants based on fractal theory and gray level co-occurrence matrix textures. J. Remote Sens.
**2014**, 18, 868–886. [Google Scholar] - Melgani, F.; Bruzzone, L. Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens.
**2004**, 42, 1778–1790. [Google Scholar] [CrossRef] - Gumus, E.; Kilic, N.; Sertbas, A.; Ucan, O.N. Evaluation of face recognition techniques using PCA, wavelets and SVM. Expert Syst. Appl.
**2010**, 37, 6404–6408. [Google Scholar] [CrossRef] - Song, X.; Duan, Z.; Jiang, X. Comparison of artificial neural networks and support vector machine classifiers for land cover classification in Northern China using a SPOT-5 HRG image. Int. J. Remote Sens.
**2012**, 33, 3301–3320. [Google Scholar] [CrossRef] - Zhao, Y. The Principle and Method of Analysis of Remote Sensing Application, 1st ed.; Science Press: Beijing, China, 2003; pp. 207–208. [Google Scholar]

Method | Equation | Description |
---|---|---|

Angular second moment (ASM) | $ASM={\displaystyle \sum _{i}^{N}{\displaystyle \sum _{j}^{N}P(i,j{)}^{2}}}$ | |

Contrast (CON) | $CON={\displaystyle \sum _{i}^{N}{\displaystyle \sum _{j}^{N}{(i-j)}^{2}P(i,j)}}$ | |

Correlation (COR) | $COR=\frac{{\displaystyle \sum _{i}^{N}{\displaystyle \sum _{j}^{N}(i-\overline{x})(j-\overline{y})P(i,j)}}}{{\sigma}_{x}{\sigma}_{y}}$ | $\begin{array}{c}\overline{x}={\displaystyle \sum _{i}^{N}i{\displaystyle \sum _{j}^{N}P(i,j)}}\overline{y}={\displaystyle \sum _{j}^{N}j{\displaystyle \sum _{i}^{N}P(i,j)}}\\ {{\sigma}_{x}}^{2}={\displaystyle \sum _{i}^{N}{(i-\overline{x})}^{2}{\displaystyle \sum _{j}^{N}P(i,j)}}\\ {{\sigma}_{y}}^{2}={\displaystyle \sum _{j}^{N}{(j-\overline{y})}^{2}}{\displaystyle \sum _{i}^{N}P(i,j)}\end{array}$ |

Entropy (ENT) | $ENT=-{\displaystyle \sum _{i}^{N}{\displaystyle \sum _{j}^{N}P(i,j)\mathrm{lg}P(i,j)}}$ |

Class | Method One | Method Two |
---|---|---|

Water | 0.8681 | 0.9535 |

Forest land | 0.8543 | 0.8665 |

Arable land | 0.8374 | 0.9976 |

Residential land | 0.7016 | 0.7564 |

Roads | 0.7113 | 0.6788 |

Bare land | 0.5974 | 0.6399 |

OA/% | 86.92 | 92.43 |

Kappa coefficient | 0.83 | 0.87 |

Class | Method One | Method Two |
---|---|---|

Water | 0.8425 | 0.9792 |

Forest land | 0.8517 | 0.8595 |

Arable land | 0.8526 | 0.9842 |

Residential land | 0.6937 | 0.7611 |

Roads | 0.7012 | 0.7057 |

Bare land | 0.6843 | 0.7368 |

OA/% | 85.70 | 93.26 |

Kappa coefficient | 0.82 | 0.89 |

Class | Method One | Method Two |
---|---|---|

Water | 0.8714 | 0.9946 |

Forest land | 0.8623 | 0.8705 |

Arable land | 0.8435 | 0.9860 |

Residential land | 0.7123 | 0.7768 |

Roads | 0.6930 | 0.6964 |

Bare land | 0.6551 | 0.6646 |

OA/% | 88.51 | 96.75 |

Kappa coefficient | 0.84 | 0.93 |

Arable Land | Forest Land | |||||||
---|---|---|---|---|---|---|---|---|

0° | 45° | 90° | 135° | 0° | 45° | 90° | 135° | |

Direction measure | 133 | 94 | 12 | 67 | 64 | 75 | 61 | 70 |

Weight factor | 0.017 | 0.032 | 0.864 | 0.087 | 0.257 | 0.208 | 0.311 | 0.224 |

Class | ASM | CON | COR | ENT |
---|---|---|---|---|

Arable land | 0.0447 | 3.8586 | 0.0481 | 4.2301 |

0.0411 | 3.0875 | 0.0389 | 4.5788 | |

0.0439 | 3.1024 | 0.0431 | 4.4029 | |

0.0392 | 2.9304 | 0.0379 | 4.5967 | |

Forest land | 0.2670 | 2.6334 | 0.1446 | 2.4026 |

0.2040 | 2.2045 | 0.1384 | 2.4786 | |

0.2134 | 2.1342 | 0.1265 | 2.7665 | |

0.2587 | 2.4563 | 0.1323 | 2.5876 |

Class | ASM | CON | COR | ENT |
---|---|---|---|---|

Arable land | 0.0556 | 6.2471 | 0.0470 | 5.4875 |

0.0549 | 6.0083 | 0.0413 | 6.0032 | |

0.0551 | 6.1267 | 0.0434 | 5.8976 | |

0.5127 | 5.9872 | 0.0405 | 6.1233 | |

Forest land | 0.3456 | 1.2545 | 0.2854 | 2.2321 |

0.3208 | 1.2074 | 0.2475 | 2.2482 | |

0.3306 | 0.1944 | 0.2409 | 2.4874 | |

0.3418 | 1.2475 | 0.2463 | 2.3677 |

Method | Forest Land | Arable Land |
---|---|---|

Method one | 90.57 | 91.30 |

Method two | 91.45 | 95.46 |

© 2017 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**

Zhang, X.; Cui, J.; Wang, W.; Lin, C. A Study for Texture Feature Extraction of High-Resolution Satellite Images Based on a Direction Measure and Gray Level Co-Occurrence Matrix Fusion Algorithm. *Sensors* **2017**, *17*, 1474.
https://doi.org/10.3390/s17071474

**AMA Style**

Zhang X, Cui J, Wang W, Lin C. A Study for Texture Feature Extraction of High-Resolution Satellite Images Based on a Direction Measure and Gray Level Co-Occurrence Matrix Fusion Algorithm. *Sensors*. 2017; 17(7):1474.
https://doi.org/10.3390/s17071474

**Chicago/Turabian Style**

Zhang, Xin, Jintian Cui, Weisheng Wang, and Chao Lin. 2017. "A Study for Texture Feature Extraction of High-Resolution Satellite Images Based on a Direction Measure and Gray Level Co-Occurrence Matrix Fusion Algorithm" *Sensors* 17, no. 7: 1474.
https://doi.org/10.3390/s17071474