合成孔徑雷達(synthetic aperture radar, SAR)影像經常被使用在監測地面的目
標物，因為其主動提供能量源以及對雲雨的穿透特性，在任何時間或天氣
狀況之下，都可以得到地面觀測的結果。因此利用前後期影像配合變遷偵
測技術，可以監測自然環境的事件或者人為活動對於環境的改變。理論上
合成孔徑訊號可以用空間混沌理論模型來表示，因為一個像素解析度內的
散射訊號是同調訊號的總合，而渾沌理論模型的特性是碎形維度。在這個
研究中，使用兩種估算碎形維度的方法，一種是差分盒維度數(differential
box-counting, DBC) ，另一種是改進 z 方向的碎形維度。我們提出一個簡單
的基於空間混沌模型理論的合成孔徑雷達影像變遷偵測方法。首先先計算
各個時期合成孔徑雷達影像的碎形維度之差值，偵測建物和植被復育區域
造成的改變。接著各自使用固定誤警率 (constant false alarm rate, CFAR)和
支持向量機 (support vector machine, SVM) 去分類變遷和未變遷區域。實驗
結果顯示出差分盒維度數和改進的方法對於偵測建物區域的準確度是差不
多的。然而，植被的復育部分，改進的碎形方法相較於差分盒維度數有較
好的分類準確度。最後的結果也顯示支持向量機相較於固定誤警率，不管
在建物區域或者植被復育區域都有比較好的分類效能。前後期影像的碎形
數值直接交由支持向量機作變遷偵測分類比前後期影像強度差值的準確度
還高。

Due to their all-weather, all-time and penetration characteristics, synthetic aperture radar (SAR) images are frequently used to monitor ground targets. As a result, environmental changes via natural events or human activities can be observed by applying a change detection technique. Theoretically, SAR signals can be characterized as chaotic phenomena since the scattering of signals within a resolution cell can be summed coherently. Accordingly, an SAR signal can be represented by spatial chaotic model (SCM) and characterized by its fractal dimension. In this study, two approaches for estimating fractal dimensions are conducted, which are estimated by the differential box-counting (DBC) and improved fractal dimension methods in the z-direction. Based on the spatial chaotic model, a simplified SAR image change detection procedure is proposed. This method first calculates the differences in fractal dimensions among multitemporal SAR images to detect the changes in building and grass-recovery areas. Both the constant false alarm rate (CFAR) and support vector machine (SVM) are applied to classify the changed and unchanged areas, respectively. The experimental results reveal that both the DBC and improved fractal dimension methods are similar for detecting changes in building areas. However, regarding the changes in grass recovery areas, the improved fractal dimension method outperforms the DBC method. The results also show that the SVM performs better than the CFAR for both building and grass areas. SVM with all fractal dimensions has higher accuracy than only the difference of fractal
dimensions difference between SAR images.

[1] Bruzzone, L. and S. B. Serpico, 1997: An Iterative Technique for the Detection of Land-cover Transitions in Multispectral Remote-Sensing Images, IEEE Transactions on Geoscience and Remote Sensing, 35, 858-867.
[2] Chavez, P. S. and D. J. Mackinnon, 1994: Automatic Detection of Vegetation Changes in the Southwestern United States Using Remotely Sensed Images, Photogrammetric Engineering and Remote Sensing, 60, 571-583.
[3] Grover, K. D., S. Quegan, and C. da Costa Freitas, 1999: Quantitative Estimation of Tropical Forest Cover by SAR, IEEE Transactions on Geoscience and Remote Sensing, 37, 479-490.
[4] Hame, T., I. Heiler, and J. San Miguel-Ayanz, 1998: An Unsupervised Change Detection and Recognition System for Forest, International Journal of Remote Sensing, 19, 1079-1099.
[5] Ridd, M. K. and J. Liu, 1998: A Comparison of Four Algorithms for Change Detection in an Urban Environment, Remote Sensing of Environment, 63, 95-100. [6] Quegan, S., T. Le Toan, J. J. Yu, F. Ribbes, and N. Floury, 2000: Multitemporal ERS SAR Analysis Applied to Forest Mapping, IEEE Transactions on Geoscience and Remote Sensing, 38, 741-753.
[7] Chang, C. P., J. Y. Yen, A. Hooper, F. M. Chou, Y. A. Chen, C. S. Hou, W. C. Hung, and M. S. Lin, 2010: Monitoring of Surface Deformation in Northern Taiwan Using DInSAR and PSInSAR Techniques, Terre. Atoms. Ocean. Sci., 21, 447-461.
63
[8] Bovolo, F. and L. Bruzzone, 2005: A Detail-Preserving Scale-Driven Approach to Change Detection in Multitemporal SAR Images, IEEE Transactions on Geoscience and Remote Sensing, 43, 2963-2972.
[9] Bruzzone, L. and D. F. Prieto, 2000: Automatic Analysis of the Difference Image for unsupervised Change Detection, IEEE Transactions on Geoscience and Remote Sensing, 38, 1171-1182.
[10] Lu, D., P. Mausel, E. Brondizio, and E. Moran, 2004: Change Detection Techniques, International Journal of Remote Sensing, 25, 2365-2401.
[11] Lee, J. S., 1980: Digital Image Enhancement and Noise Filtering by Use of local Statistics, IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-2, 165-168.
[12] Frost, V. S., J. A. Stiles, K. S. Shanmugan, and J. C. Holtzman, 1982: A Model for Radar Images and Its Application to Digital Filtering of Multiplicative Noise, IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-4, 157-165.
[13] Lopes, A., E. Nerzy, R. Touzi, and H. Laur, 1990: Maximum a Posteriori Speckle Filtering and First Texture Models in SAR Images, IEEE International Geoscience and Remote Sensing Symposium Proceedings (IGARSS 1990), 3, 2409-2412.
[14] Solbø, S. and T. Eltoft, 2004: Homomorphic Wavelet-based Statistical Despeckling of SAR Images, IEEE Transactions on Geoscience and Remote Sensing, 42, 711-721.
[15] Singh, P. and R. Shree, 2017: Statistical Quality Analysis of Wavelet Based
SAR Images in Despeckling Process, Asian Journal of Electrical Sciences, 6, 1-18.
[16] Bazi, Y., L. Bruzzone, and F. Melgani, 2005: An Unsupervised Approach Based on the Generalized Gaussian Model to Automatic Change Detection in Multitemporal SAR Images, IEEE Transactions on Geoscience and Remote Sensing, 43, 874–887.
[17] Tzeng, Y. C. and K. S. Chen, 2007: Change Detection in Synthetic Aperture Radar Images Using a Spatially Chaotic Model, Optical Engineering, 46, 1-9.
[18] Chou, N. S., Y. C. Tzeng, K. S. Chen, C. T. Wang and K. C. Fan, 2009: On the application of a spatial chaotic model for detecting landcover changes in synthetic aperture radar images, Journal of Applied Remote Sensing, 3.
[19] Sarkar, N. and B. B. Chaudhuri, 1994: An Efficient Differential Box-Counting Approach to Compute Fractal Dimension of Images, IEEE Transactions on Systems, Man, and Cybernetics, 24, 115-120.
[20] Wieland, M., W. Liu and F. Yamazaki, 2016: Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes. Remote Sens., 8, 792, doi:10.3390/rs8100792.
[21] Zeng, Y., J. Zhang, and J. L. Van Genderen, 2008: Change Detection Approach to SAR and Optical Image Integration. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXVII.
[22] Forouzanfar, M. and H. Abrishami-Moghaddam, 2010: Ultrasound Speckle
Reduction in the Complex Wavelet Domain, in Principles of Waveform Diversity and Design, M. Wicks, E. Mokole, S. Blunt, R. Schneible, and V. Amuso (eds.), SciTech Publishing, Section B - Part V, 558-77.
[23] Jayaraman et al., 2009: Digital Image Processing, Tata McGraw Hill Education, 272.
[24] Rosin, P. and J. Collomosse, 2012: Image and Video-Based Artistic Stylisation, Springer, 92.
[25] Paolini, L., et al., 2006: Radiometric correction effects in Landsat multi‐ date/multi‐sensor change detection studies, International Journal of Remote Sensing, 27, 685‐704.
[26] Nielsen, A., 1998: Multivariate Alteration Detection (MAD) and MAF Postprocessing in Multispectral, Bitemporal Image Data: New Approaches to Change Detection Studies, Remote Sensing of Environment, 64, 1-19.
[27] Nielsen, A. A., 2002: Multiset canonical correlations analysis and multispectral, truly multitemporal remote sensing data, IEEE Transactions on Image Processing, 11, 293-305.
[28] Dai, X. and S. Khorram, 1998: The effects of image misregistration on the accuracy of remotely sensed change detection, IEEE Transactions on Geoscience and Remote Sensing, 36, 1566-1577.
[29] Debotosh, B., 2014: Adaptive polar transform and fusion for human face image processing and evaluation, Human-Centric Computing and Information Science, 4, 1–18.
[30] Lee, J. S., et al., 1999: Polarimetric SAR speckle filtering and its
implication for classification, IEEE Transactions on Geoscience and Remote Sensing, 37, 2363‐2373.
[31] Martinez, C. L. and X. Fabregas, 2003: Polarimetric SAR speckle noise model, IEEE Transactions on Geoscience and Remote Sensing, 41, 2232-2242.
[32] Zaman, M. R. and C. R. Moloney, 1993: A Comparison of Adaptive Filters for Edge ‐ preserving Smoothing of Speckle Noise, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 1993).
[33] Abarbanal, H. D. I., 1996: Analysis of Observed Chaotic Data, Springer-Verlag New York.
[34] Chen, W. S., S. Y. Yuan, and C. M. Hsieh, 2003: Two Algorithms to Estimate Fractal Dimension of Gray-level Images, Opt. Eng., 42, 2452–2464.
[35] Chiu S., 2005: A constant false alarm rate (CFAR) detector for RADARSAT-2 along-track interferometry, Can. J. Remote Sensing, 31, 1, 73-84.
[36] Chervonenkis, A. Y., 2013: Early history of support vector machines, Empirical Inference, Springer, 13-20.
[37] Boser, B. E., I. Guyon, and V. N. Vapnik, 1992: A training algorithm for optimal margin classifiers, In Proceedings of the Fifth Annual Workshop of Computational Learning Theory, 5, 144-152.
[38] Cortes, C. and V. N. Vapnik, 1995: Support-vector networks, Machine Learning, 20, 273–297.
[39] Burges, C. J. C., 1998: A Tutorial on Support Vector Machines for Pattern Recognition, Data Mining and Knowledge Discovery 2, 121-167.
[40] Hastie, T., R. Tibshirani and J. Friedman, 2009: The elements of Statistical
Learning: Data Mining, Inference, and Prediction, Springer, 134.
[41] MacQueen, J. B., 1967: Some Methods for classification and Analysis of Multivariate Observations. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, University of California Press, 281–297.
[42] Goodfellow, I. J., et al., 2014:Generative Adversarial Networks. CoRR abs, 1406-2661.
[43] Kohonen , T., 1990: The Self-organizing map, Proceedings of the IEEE, 78, 9, 1464-1480.
[44] Grossberg, S., 2013: Adaptive resonance theory: how a brain learns to consciously attend, learn, and recognize a changing World. Neural Networks 37, 1-47.
[45] Ayyaz M. N., I. Javed and W. Mahmood, 2012: Handwritten Character Recognition Using Multiclass SVM Classification with Hybrid Feature Extraction, Pakistan Journal of Engineering & Applied Science, 10, 57-67.
[46] Akbarizadeh, G., 2012: A New Statistical-Based Kurtosis Wavelet Energy Feature for Texture Recognition of SAR Images, IEEE Transactions on Geoscience and Remote Sensing, 50, 4358-4368.
[47] Yekkehkhany, B., A. Safari, S. Homayouni, and M. Hasanlou, 2014: A Comparison Study of Different Kernel Functions for SVM-Based Classification of Multi-Temporal Polarimetry SAR Data, The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, XL-2/W3, 281-285.
[48] Qu, J., C. Wang, and Z. Wang, 2003: Structure-Context Based Fuzzy Neural Network Approach for Automatic Target Detection, Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, 2, 767-769.
[49] Green, D. M. and J. A. Swets, 1996: Signal Detection Theory and Psychophysics, Wiley, New York.
[50] Staelin, D. H., 1966: Measurements and Interpretation of the Microwave Spectrum of the Terrestrial Atmosphere near 1 Centimeter Wavelength, Journal of Geophysical Research, 71, 2875-2881.