跳到主要內容

臺灣博碩士論文加值系統

(3.231.230.177) 您好!臺灣時間:2021/08/02 10:35
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:鄭堯壬
研究生(外文):Yao-Jen Cheng
論文名稱:多相機監控系統中之物件顏色校正與不變量分析
論文名稱(外文):Color Constancy and Invariance Analysis for Multiple Camera Surveillance System
指導教授:謝君偉謝君偉引用關係
學位類別:碩士
校院名稱:元智大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:英文
論文頁數:50
中文關鍵詞:非重疊多相機環境顏色校正顏色辨識行人車輛
外文關鍵詞:non-overlapping camera environmentcolor calibtationcolor classificationpedestrianvehicle
相關次數:
  • 被引用被引用:0
  • 點閱點閱:126
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
顏色校正在許多視覺影像處理的問題中,扮演著相當重要的腳色,因為同一顏色在不同的光照或是不同品牌的相機、攝影機拍攝下會造成其紅、綠、藍三種表示顏色的值的差異。由於這些值的差異會造成影像處理上的誤判,最簡單的例子就是利用背景相減來減出前景就很容易受到光照改變而使得前景剪錯;還有利用身上的顏色來做行人辨識也會受到光線改變而誤判。因此我們需要一個改變的準則來降低這些錯誤。
本系統主要是要找到兩個場景的顏色對應關係,先對紅綠藍三種顏色的像素數目做統計,統計其累加機率密度函數CDF,然後再利用動態時間校正法的來對累加機率密度函數找出兩兩統計圖的最佳對應關係。之後再使用主軸分析法來找出最佳化的對應關係。此外,本系統也辨識背景相減出來的顏色,主要將被辨識的顏色一共有七種,分別是黑、灰、白、紅、綠、藍、黃。辨識階段分為兩個層級:灰階類別層級與非灰階類別層級。其中的門檻值是用訓練出來的。由實驗結果得知,我們的系統可以適用於許多不同條件底下,並且有很好的表現。
Color constancy plays an important role in many multimedia processes. This thesis proposes a novel approach for color constancy in multi-camera environment. The traditional algorithm of color constancy is to separate the illumination and the reflectance components on images and then estimate the illuminant. In this thesis, it will be focused on the color transformations among non-overlapping multiple cameras. According to the algorithm “Dynamic Time Warping”, it produces several corresponding transforms among scenes. By using PCA projection, the high dimensional transforms will be projected into low dimensional space. Thus there will be a cluster in low dimension and then the system can learn the clustering center and the corresponding transform is the optimized transform that we expect. So that the operator can apply the optimal color transform to calibrating the color and lighting changes among multiple cameras. After the calibration of colors, the next task is to identify object in different scenes. Furthermore, object color classification is an attached work in this thesis, especially vehicle colors. There are 7 principal colors, which is black, gray, white, red, green, blue, and yellow. In this system, the classification will be divided into two stages, gray level like decision stage and non-gray level like decision stage. The decision thresholds in each stage are learned from training data. Experimental results reveal the performances in several different conditions.
摘 要 i
Abstract ii
誌 謝 iii
Content iv
List of figure v
List of table vii
Chapter 1 Introduction 1
Chapter 2 System Overview 4
Chapter 3 Color Constancy 6
3.1 Background Subtraction 6
3.2 Remove Foreground Noise 7
3.3 The Formation of Color Correspondence 9
3.4 Dynamic Time Warping (DTW) 11
3.5 Principal Component Analysis (PCA) 13
3.6 Principal Component Analysis for Optimization 14
3.7 Color Histogram 17
3.8 Objects (Vehicles or Pedestrians) Recognition 18
Chapter 4 Object Color Identification 20
4.1 Gray Level Category Decision 22
4.2 Non-Gray Level Category Decision 23
4.2.1 Remove Gray Level Color 24
4.2.2 Equalization 24
4.2.3 K-mean algorithm 27
4.2.4 Non-Gray Level Pixel-based Classification 28
Chapter 5 Experimental Results 30
5.1 Object Recognition Result 30
5.1.1 Outdoor Environment: Highway Scenes 30
5.1.2 Indoor Environment: School Scenes 37
5.2 Color Identification Result 43
Chapter 6 Conclusions 46
References 47
[1]K. Grauman and T. Darrell, “The Pyramid Match Kernel: Discriminative Classification with Sets of Image Features,” Proc. of IEEE International Conference on Computer Vision, vol. 2, pp. 1458-1465, Oct 17-21, 2005.
[2]S. Beucher and C. Lantuejoul, “Use of watersheds in contour detection,” Proc. Int. Workshop Image Processing, Real-Time Edge and Motion Detection/Estimation, Rennes, France, Sept. 17-21, 1979.
[3]R. C. Gonzalez and R. E. Woods, Digital Image Processing, Second Edition.
[4]Ahmed Elgammal, David Harwood, and Larry Davis, “Non-parametric Model for Background Subtraction,” Proc. 6th European Conference on Computer Vision, pp.751-767, 2000.
[5]Thanarat Horprasert, David Harwood, and Larry S. Davis, “A Statistical Approach for Real-time Robust Background Subtraction and Shadow Detection,” IEEE International Conference on Computer Vision, pp.1-19, 1999.
[6]Omar Javed, Khurram Shafique, and Mubarak Shah, “A Hierarchical Approach to Robust Background Subtraction using Color and Gradient Information,” Proceedings of IEEE Motion and Video Computing, pp.22-27, 2002.
[7]Kyungnam Kim, Thanarat H. Chalidabhongse, David Harwood, and Larry Davis, “Background Modeling and Subtraction by Codebook Construction,” IEEE International Conference on Image Processing, vol. 5, pp.3061-3064, 2004.
[8]Kyungnam Kim, Thanarat H. Chalidabhongse, David Harwood, and Larry Davis, “Real-time foreground - background segmentation using codebook model,” Real-time Imaging, vol. 11, no. 3, pp.172-185, 2005.
[9]L. D. Stefano and A. Bugareli, “A Simple And Efficient Connected Components Labeling Algorithm,” Proceedings of International Conference on Image Analysis and Processing, pp. 322-327, Setp. 1999.
[10]Omar Javed, Khurram Shafique, and Mubarak Shah, “Appearance Modeling for Tracking in Multiple Non-overlapping Cameras,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp.26-33, 2005.
[11]Michael D. Grossberg and Shree K. Nayar, “Determining the Camera Response from Images: What Is Knowable?” IEEE Transactions on Pattern Recognition and Machine Intelligence, vol. 25, 1455-1467, Nov. 2003.
[12]G. Buchsbaum, “A spatial processor model for object colour perception,” Journal of the Franklin Institute, 1980.
[13]J. van de Weijer and Th. Gevers, “Color Constancy based on the Grey-Edge Hypothesis,” IEEE International Conference on Image Processing, vol.2, pp. 722-725, Setp. 2005.
[14]Massimo Piccardi and Eric Dahai Cheng, “Multi-Frame Moving Object Track Matching Based on an Incremental Major Color Spectrum Histogram Matching Algorithm,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 3, pp.19, 2005.
[15]Graham D. Finlayson, Steven D. Hordley, and Mark S. Drew, “Removing Shadows from Images,” Proc. European Conf. Computer Vision, pp. 823–836, 2002.
[16]Graham D. Finlayson, Mark S. Drew, and Brian V. Funt, “Spectral sharpening: Sensor Transformations for Improved Color Constancy,” The Optical Society of America, vol. 11, May 1994.
[17]Graham D. Finlayson, Clement Fredembach, and Mark S. Drew, “Detecting Illumination in Images,” International Conference in Computer Vision, pp. 1-8, Oct 2007.
[18]Yaser Sheikh, Xin Li, and Mubarak Shah, “Trajectory Association across Non-overlapping Moving Cameras in Planar Scenes,” IEEE Conf. of Computer Vision and Pattern Recognition, pp. 1-7, Jun. 2007.
[19]Saad M. Khan and Mubarak Shah, “A Multiview Approach to Tracking People in Crowded Scenes using a Planar Homography Constraint,” European Conference of Computer Vision, vol. 4, pp.133-146, 2006.
[20]Dimitrios Makris, Tim Ellis, and James Black, “Bridging the Gaps between Cameras,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition,vol. 2, pp. 205-210, Jun. 2004.
[21]Omar Javed, Sohaib Khan, Zeeshan Rasheed, and Mubarak Shah, “Camera Handoff: Tracking in Multiple Uncalibrated Stationary Cameras,” IEEE Workshop on Human Motion, 2000.
[22]Sohaib Khan, Omar Javed, Zeeshan Rasheed, and Mubarak Shah, “Human Tracking in Multiple Cameras,” International Conference on Computer Vision, vol. 1, pp.331, Jul. 2001.
[23]Sohaib Khan and Mubarak Shah, “Consistent Labeling of Tracked Objects in Multiple Cameras with Overlapping Fields of View,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 10, Oct. 2003.
[24]Peter Vincent Gehler, Carsten Rother, Andrew Blake, Tom Minka, and Toby Sharp, “Bayesian Color Constancy Revisited,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Jun. 2008.
[25]Eamonn J. Keogh and Michael J. Pazzani, “Derivative Dynamic Time Warping,” in First SIAM International Conference on Data Mining, 2001.
[26]Eamonn J. Keogh and Chotirat Ann Ratanamahatana, “Exact Indexing of Dynamic Warping,” Proceedings of the 28th international conference on Very Large Data Bases, 2002.
[27]S. Chu, E. Keogh, D. Hart, and M. Pazzani, “Iterative Deepening Dynamic Time Warping for Time Series,” In Proceedings of SIAM International Conference. on Data Mining, 2002.
[28]S. Kullback, Information theory and statistics, Dover Books on Mathematics.
[29]H Ibrahim and NSP Kong, “Brightness Preserving Dynamic Histogram Equalization for Image Contrast Enhancement,” IEEE Transactions on Consumer Electronics In Consumer Electronics, Vol. 53, No. 4, pp. 1752-1758 ,Nov. 2007.
[30]Gang Wui, Amir Rahimi, Kingshy Goh, Tomy Tsai, Ankur Jain, Yi Wu, Edward Y. Chang, and Yuan-Fang Wang, “Identifying Color in Motion in Video Sensors,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 561-569 , Jun. 2006.
[31]Joost van de Weijer, Cordelia Schmid, and Jakob Verbeek, “Learning Color Names from Real-World Images,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Jun. 2007.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top