臺灣博碩士論文加值系統

分析平面影像的資訊是電腦視覺主要的研究領域，其中紋理的識別與分類一直是重要的課題。均勻分布的紋理可以是自然或是人工布料的質地，而在本篇論文裡將研究如何去分類流行服飾上面均勻分布的幾何圖案。這樣一個分類器有助於在影像分割上快速擷取人們的衣著資訊，方便進一步分析服裝款式。傳統的紋理分類方法基本上有LBP、HoG，以及其他變形，近來則是以類神經網路、CNN為主流。雖然CNN具有高準確度，但是相對於傳統的方法，需要大量的資料來去訓練，同時過多的參數需要調配也是缺點之一。本篇論文以改進LTP為主軸進行研究，並以旋轉不變性、尺度不變性為目標將演算法流程加以改良，旨在產生更容易區別分類的特徵向量。在實驗的步驟，共同比較了其他演算法彼此的效能差異，包含面對不同性質的紋理資料集的辨識準確率，以及運算的時間複雜度。而在大部分的情況下，我們的方法都具有不錯的效率與表現。

Texture classification and recognition is an important topic in computer vision research area. In this work, we aim at studying classification of the evenly distributed geometrical pattern printed on the fabrics. This research is helpful to get the clothing information rapidly when performing image segmentation and making the subsequent clothing-style analysis more convenient. Traditional texture analysis features include Local Binary Patterns、Histogram of Oriented Gradients, and other transformation methods. Convolutional neural networks are popular methods in recent year. However, to extract features from CNN, it needs a huge amount of data compared to traditional methods. Also, it is sensitive to parameters. In this paper, we improve LTP and take it as our main work to study. To deal with the problem of rotation invariance and scale invariance, the proposed algorithm generates a feature vector with better discriminability. In experiments, we compare the performance of our method with others on different characteristic texture database. Also, we analyze the time complexity of different algorithms. In general, our method has superior performance and efficiency.

摘要 ................................................................................................................................................... i
Abstract ............................................................................................................................................ ii
致謝 ................................................................................................................................................ iii
目錄 ................................................................................................................................................. iv
圖目錄 ............................................................................................................................................. vi
表目錄 ...........................................................................................................................................viii
第一章 緒論 ................................................................................................................................. 1
1.1 研究動機 ............................................................................................................................ 1
1.2 相關研究 ............................................................................................................................ 1
1.3 系統流程 ............................................................................................................................ 2
1.3.1 紋理特徵擷取流程 .................................................................................................... 2
1.3.2 紋理特徵實驗流程 .................................................................................................... 3
1.3.3 紋理特徵應用流程 .................................................................................................... 3
1.4 論文架構 ............................................................................................................................ 4
第二章 相關文獻 ......................................................................................................................... 5
2.1 FLS架構 ............................................................................................................................ 5
2.2 Local Binary Pattern .......................................................................................................... 5
2.3 Local Ternary Pattern ........................................................................................................ 6
2.4 Uniform Pattern Histogram ............................................................................................... 7
2.5 Rotation Invariant Pattern Histogram ............................................................................... 8
2.6 Orthogonal Symmetric Local Ternary Pattern ................................................................. 9
2.7 Scale Invariant Local Ternary Pattern ............................................................................ 10
2.8 Spatial Pyramid Matching ............................................................................................... 10
2.9 Polar Coordinate Transform ............................................................................................ 10
2.10 Histogram of Gradient ................................................................................................... 11
第三章 特徵擷取與應用 ........................................................................................................... 12
3.1 紋理特徵擷取 .................................................................................................................. 12
3.1.1 離散小波轉換 ........................................................................................................... 12
v
3.1.2 SRLTP(同步旋轉區域三元模式) ........................................................................... 14
3.1.3 離散傅立葉轉換 ...................................................................................................... 15
3.1.4 主成分分析 ............................................................................................................... 16
3.1.5 支持向量機 ............................................................................................................... 18
3.2 紋理特徵應用 .................................................................................................................. 20
3.2.1 人臉偵測 ................................................................................................................... 20
3.2.2 Grabcut前景切割 .................................................................................................... 22
第四章 實驗結果討論 ............................................................................................................... 23
4.1 資料集介紹 ...................................................................................................................... 23
4.1.1 UMD資料集 ............................................................................................................ 23
4.1.2 Brodatz資料集 ......................................................................................................... 23
4.1.3 Outex資料集 ............................................................................................................ 24
4.1.4 DTD與服飾紋理圖相混和資料集(DTD+) ......................................................... 25
4.2 演算法準確率驗證.......................................................................................................... 26
4.2.1 實驗設置 ................................................................................................................... 26
4.2.2 實驗評比 ................................................................................................................... 26
4.2.3 實驗結果 ................................................................................................................... 27
4.3 實驗結果分析 .................................................................................................................. 32
4.3.1 SRLTP之於旋轉性質紋理資料集 ........................................................................ 32
4.3.2 SRLTP之於非旋轉性質紋理資料集 .................................................................... 33
4.3.3 各演算法準確率比較 .............................................................................................. 34
4.3.4 各演算法複雜度比較 .............................................................................................. 35
4.4 多尺度圖像與2D-DWT比較 ....................................................................................... 38
4.5 SRLTP閥值之於準確率之影響 .................................................................................... 39
第五章 結論與未來研究 ........................................................................................................... 41
參考文獻 ........................................................................................................................................ 42

［1］ Jianbo Shi and J. Malik, "Normalized cuts and image segmentation," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 888-905, Aug 2000.
［2］ T. Ojala, M. Pietikainen and D. Harwood, "Performance evaluation of texture measures with classification based on Kullback discrimination of distributions," Proceedings of 12th International Conference on Pattern Recognition, Jerusalem, 1994, pp. 582-585 vol.1.
［3］ Xiaoyang Tan and Bill Triggs, Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions, IEEE Transactions on Image Processing, 19(6), pp. 1635-1650, 2010.
［4］ G. Zhao, T. Ahonen, J. Matas and M. Pietikainen, "Rotation-Invariant Image and Video Description With Local Binary Pattern Features," in IEEE Transactions on Image Processing, vol. 21, no. 4, pp. 1465-1477, April 2012.
［5］ He, C., Ahonen, T., Pietikäinen, M.:A Bayesian local binary pattern texture descriptor. In: Proc.International Conference on Pattern Recognition, pp. 1–4 (2008)
［6］ Mingming Huang, Zhichun Mu, Hui Zeng, Shuai Huang, “Local image region description using orthogonal symmetric local ternary pattern”, In: Pattern Recognition Letters, Volume 54, 1 March 2015, Pages 55-62
［7］ S. Liao, G. Zhao, V. Kellokumpu, M. Pietikäinen and S. Z. Li, "Modeling pixel process with scale invariant local patterns for background subtraction in complex scenes," 2010
43
IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, CA, 2010, pp. 1301-1306.
［8］ S. Lazebnik, C. Schmid and J. Ponce, "Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories," 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06), 2006, pp. 2169-2178.
［9］ M. H. Kiapour, X. Han, S. Lazebnik, A. C. Berg and T. L. Berg, "Where to Buy It: Matching Street Clothing Photos in Online Shops," 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 2015, pp. 3343-3351.
［10］ L. Wolf, T. Hassner and Y. Taigman, "Effective Unconstrained Face Recognition by Combining Multiple Descriptors and Learned Background Statistics," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 10, pp. 1978-1990, Oct. 2011.
［11］ G. Zhao, T. Ahonen, J. Matas and M. Pietikainen, "Rotation-Invariant Image and Video Description With Local Binary Pattern Features," in IEEE Transactions on Image Processing, vol. 21, no. 4, pp. 1465-1477, April 2012.
［12］ C. He, T. Zhuo, X. Su, F. Tu and D. Chen, "Local Topographic Shape Patterns for Texture Description," in IEEE Signal Processing Letters, vol. 22, no. 7, pp. 871-875, July 2015.
44
［13］ Y. Xu, S. Huang, H. Ji and C. Fermuller, "Combining powerful local and global statistics for texture description," 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, 2009, pp. 573-580.
［14］ Jolliffe, I. T. Principal Component Analysis. Springer-Verlag. 1986: 487. ISBN 978-0-387-95442-4. doi:10.1007/b98835.
［15］ P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, 2001, pp. I-511-I-518 vol.1.
［16］ Vision Burges, Christopher J. C.; A Tutorial on Support Vector Machines for Pattern Recognition, Data Mining and Knowledge Discovery 2:121–167, 1998
［17］ Stein, Elias;Weiss, Guido,Introduction to Fourier Analysis on Euclidean Spaces, Princeton, N.J.: Princeton University Press, 1971, ISBN 978-0-691-08078-9.
［18］ Stéphane Mallat, A Wavelet Tour of Signal Processing