Presentation is loading. Please wait.

Presentation is loading. Please wait.

Chapter 2 Getting to Know Your Data Yubao (Robert) Wu Georgia State University.

Similar presentations


Presentation on theme: "Chapter 2 Getting to Know Your Data Yubao (Robert) Wu Georgia State University."— Presentation transcript:

1 Chapter 2 Getting to Know Your Data Yubao (Robert) Wu Georgia State University

2 Chapter 2 Getting to Know Your Data Data Objects and Attribute Types Basic Statistical Descriptions of Data Data Visualization Measuring Data Similarity and Dissimilarity

3 Printer Forensics ? Printer Forensics Based on Page Document's Geometric Distortion Yubao Wu, Xiangwei Kong, Xingang You, and Yiping Guo IEEE International Conference on Image Processing (ICIP), 2009

4 Printer Forensics scan Digital Image Feature extraction  Darkness  Dots around the letters SVM

5 print scan PDF FileScanned Image Printer Forensics Based on Page Document's Geometric Distortion Yubao Wu, Xiangwei Kong, Xingang You, and Yiping Guo IEEE International Conference on Image Processing (ICIP), 2009

6 Least Squares – Slope of Each Line X-coordinate of “e” in one line Y-coordinate Printer Forensics Based on Page Document's Geometric Distortion Yubao Wu, Xiangwei Kong, Xingang You, and Yiping Guo IEEE International Conference on Image Processing (ICIP), 2009

7 Slope of Each Line Line Number Matlab - plot Why are the slopes monotonically increasing? Printer Forensics Based on Page Document's Geometric Distortion Yubao Wu, Xiangwei Kong, Xingang You, and Yiping Guo IEEE International Conference on Image Processing (ICIP), 2009

8 IdealIn practice Age of the Printer !

9 Experimental Results Each data point = 1 Page = 1 Scanned Image 100% Accuracy Scatterpl ot

10 Feature Extraction == Data Reduction Original Scan Image 2550 x 3508 pixels 41 lines; 72 “e” in each line 41 x 72 = 2911 data points 41 slopes for the 41 lines 2550 pixels 3508 pixels 41 double values

11 Matlab Plot Function plot(vX,vY,'kx'); hold on; Matlab default settingsModify the Settings

12 Matlab Plot Function plot(X,Y,'kx'); hold on; Matlab Default Settings Modify the Settings hFigHandle = figure(1); set(gca,'FontSize',40); plot(X,Y,'kx','LineWidth',4,'MarkerE dgeColor','k','MarkerFaceColor','w', 'MarkerSize',20); hold on; xlim([1 30]); ylim([-0.05 0.05]); set(hFigHandle, 'Position', [10 50 1180 1000]);

13 Text Cloud

14 Fisher’s Iris Data Scatterplo t

15 Fisher’s Iris Data

16 Microsoft Bing MapsGoogle Maps

17 ColorBrewer 2.0 How to Choose Colors? Matlab LineSpec (Line Specification) SpecifierColor rRed gGreen bBlue cCyan mMagenta yYellow kBlack wWhite Microsoft PowerPoint http://colorbrewer2.org/

18 How to Choose Qualitative Colors?

19 Qualitative Colors

20 http://colorbrewer2.org/

21 Sequential Colors

22 http://colorbrewer2.org/

23 https://d3js.org/ Data-driven documents http://circos.ca/

24 https://d3js.org/http://circos.ca/ The Open Graph Viz Platform


Download ppt "Chapter 2 Getting to Know Your Data Yubao (Robert) Wu Georgia State University."

Similar presentations


Ads by Google