臺灣博碩士論文加值系統

構圖是攝影中非常重要的一部分，能夠影響我們對照片的看法。一個很好的 構圖可以讓我們專注於攝影師想要在攝影主題中傳達的內容。在本篇論文，我們 討論構圖的種類，有關影像美學校正的論文以及攝影構圖過程中可能存在的缺失。
在我們的方法中，我們使用一些特徵來對影像的構圖進行分類，然後根據該 構圖類型的定義來校正影像。例如，我們使用有關線段的特徵來糾正不當的水平 構圖影像。因為我們了解影像的原始構圖，所以我們根據構圖定義修正影像時， 能避免使用無意義的特徵。
我們的方法具有良好的校正效果，水平構圖影像校正的成功率為 87.39%， 垂直構圖影像校正成功率為 85.57%，對角線構圖影像成功校正率為 71.02%。對 於太陽構圖的校正，成功率為 76.31%。與其他直接進行美學校正的論文相比，我 們的方法更切合攝影師攝影時的拍攝情景，可以在攝影師欲傳達的構圖主題和攝 影美學間達到平衡。

Composition is a very important part of photography, it affects our view of a photo. A good composition can let us focus on what the photographer wants to convey in the theme. In this thesis, we discuss the composition classification, the aesthetic correction of images and the possible defects in the composition of photography.
In our method, we use some features to classify the composition of image, and then correct the image composition according to this composition type. For example, we used features about lines in correcting images with improper horizontal composition. Because of we know the composition of the image, we correct the image according to the definition of the image composition, and avoid using useless features.
Our Method has a good effect of correction, the correction rate of horizontal composition correction is 87.39%, the correction rate of vertical composition correction is 85.57%, and the correction rate of diagonal line composition correction is 71.02%. For the sun-like composition correction, the correction rate is 76.31%.
Compared with the other papers that deal directly with aesthetic correction, our method is more suitable for the photographer’s use, and it can balance the photo to enhance aesthetics and ideas of the photographer.

中文摘要 1
Abstract 2
致謝 3
Table of Contents 3
List of Figures 8
List of Tables 10
Introduction 1
Overview 1
Motivation 1
System Description 2
Thesis organization 3
Related Works 4
Principal Types of Photographic Compositions 4
Sun-like Composition 4
Horizontal Composition 4
Vertical Composition 5
Symmetry Composition 5
Diagonal Line Composition 6
Rule of Thirds Composition 6
Vanishing Point Composition 7
Reviews of Photographic Compositions 7
Normal Improper Photographic Compositions 10
Preprocess for Correcting Improper Image Compositions 11
Classifying Photographic Composition 11
Features for Photographic Compositions 12
Decision Tree 13
Features for Horizontal Compositions, Vertical Composition and Diagonal Line Composition 15
Hough Transform for Extracting Lines 15
If we make the above transform to n points in same line, those n points in original image space will be transformed to n sine curves in - space, and those sine curves intersect in a point. It means when we find the curves of intersecting point in - space, we can determine the straight line in original image space. Slope for Grouping Lines 16
Grouping Lines for Correcting Horizontal Composition, Vertical Composition and Diagonal Line Composition 17
Features for Sun-like Composition 18
Salient Region for Extracting the Main Object 18
Binarization for Extracting the Main Object 19
Center and Area for Correcting Sun-like Composition 20
Correcting Improper Images 22
Correction of Horizontal Composition 22
Correction of Vertical Composition 23
Correction of Diagonal Line Composition 25
Correction of Sun-like composition 27
Experimental Results and Discussions 30
Experimental Setup 30
Results of Correction of Horizontal Composition 31
Results of Correction of Vertical Composition 35
Results of Correction of Diagonal Line Composition 38
Results of Correction of Sun-like Composition 42
Conclusions and Future Works 47
Conclusions 48
Future Works 48
References 49

A. E. Savakis, S. P. Etz, and A. C. Loui, “Evaluation of Image Appeal in Consumer Photography,” Proceedings of the International Society for Optics and Photonics, USA, pp. 111-121, Jan 2000.
R. O. Duda and P. E. Hart, “Use of the Hough Transformation to Detect Lines and Curves in Pictures,” Communications of the ACM, vol. 15, no. 1, pp. 11-15, Jan 1972.
P. V. C. Hough, “Method and Means for Recognizing Complex Patterns,” U.S. Patent 3 069 654, Dec. 18, 1962.
J. R. Quinlan, C4.5: Programs for Machine Learning. Morgan Kaufmann, CA: San Francisco, 1993.
L. Mai, H. Le, Y. Niu, and F. Liu, “Rule of Thirds Detection from Photograph,” Proceedings of the IEEE International Symposium on Multimedia, USA, pp. 91-96, Dec 2011.
L. Bai, X. Wang and Y. Chen, “Landscape Image Composition Analysis Based on Image Processing,” Proceedings of the IEEE International Conference on Computer Science and Automation Engineering, China, vol. 2, pp. 787-790, Jun 2013.
J. H. Huang, “A Fuzzy Logic Approach for Recognition of Photographic Compositions,” M.S. Thesis, Dept. Math. Sci., National Chengchi Univ., Taipei, Taiwan, 2007.
L. G. Liu, R. J. Chen, L. Wolf and D. Cohen-Or, “Optimizing Photo Composition,” Computer Graphics Forum, vol. 29, no. 2, pp. 469-478, May 2010.
L. Zhang, M. Song, Q. Zhao, X. Liu, J. Bu, and C. Chen, “Probabilistic graphlet transfer for photo cropping,” in IEEE Transactions on Image Processing, vol. 22, no. 2, pp. 802-815, Feb 2013.
J. Yan, S. Lin, S. B. Kang and X. Tang, “Learning the Change for Automatic Image Cropping,” The IEEE Conference on Computer Vision and Pattern Recognition, pp. 971-978, June 2013.
Y. Lin, “A Photo Composition Classification System Based on Supervised Learning,” M.S. Thesis, Dept. Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, 2014.
E. Rublee, V. Rabaud, K. Konolige and G. Bradski, “ORB: An Efficient Alternative to SIFT or SURF,” Proceedings of the IEEE International Conference on Computer Vision, Spain, pp. 2564-2571, Nov 2011.
I. Sobel and G. Feldman, “A 3x3 Isotropic Gradient Operator for Image Processing,” Presentation for Stanford Artificial Project, pp.271-272, 1968.
M. Ghane and A. Shahbahrami, “Landscape Image Filtering via Aesthetic Inference,” IOSR Journal of Engineering, vol. 2, no. 8, pp. 33-38, Aug 2012.
M. M. Cheng, G. X. Zhang, N. J. Mitra, X. Huang, and S. M. Hu, “Global Contrast Based Salient Region Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, no. 3, pp. 569-582, Mar 2015.
N. Otsu, “A Threshold Selection Method from Gray-level Histograms,” IEEE transactions on systems, man, and cybernetics, vol. 9, no. 1, pp. 62-66, Jan 1979.
S. Suzuki, “Topological Structural Analysis of Digitized Binary Images by Border Following,” Proceedings of Computer Vision, Graphics, and Image Processing, vol. 30, no. 1, pp. 32-46, Apr 1985.
E. Rosten and T. Drummond, “Machine Learning for High-speed Corner Detection,” Proceedings of the 9th European Conference on Computer Vision, Austria, pp. 430-443, May 2006.
M. Calonder, V. Lepetit, C. Strecha, and P. Fua, “BRIEF: Binary Robust Independent Elementary Features,” Proceedings of the 11th European Conference on Computer Vision, Greece, pp. 778-792, Sep 2010.
P. L. Rosin, “Measuring Corner Properties,” Computer Vision and Image Understanding, vol. 73, no. 2, pp. 291-307, Feb 1999.
M. A. Fischler and R. C. Bolles, “Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography,” Communications of the ACM, vol. 24, no. 6, pp. 381-395, Jun 1981.
J. R. Quinlan, “Induction of Decision Trees,” Machine Learning, vol. 1, no. 1, pp. 81-106, March 1986.
U. M. Fayyad and K. B. Irani, “Multi-interval Discretization of Continuous-valued Attributes for Classification Learning,” Proceedings of the 13th International Joint Conference on Artificial Intelligence, France, pp. 1022-1029, Sep 1993.
G. H. John and P. Langley, “Estimating Continuous Distributions in Bayesian Classifiers,” Proceedings of the 11th Conference on Uncertainty in Artificial Intelligence, Canada, pp. 338-345, Aug 1995.
D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning Internal Representations by Error Propagation,” Parallel Distributed Processing: Explorations in the Microstructure of Cognition: Foundations, MIT Press, MA: Cambridge, vol. 1, pp. 318-362, 1986.