跳到主要內容

臺灣博碩士論文加值系統

(44.210.83.132) 您好!臺灣時間:2024/05/25 19:23
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:覃自強
研究生(外文):Tzu-Chiang Tan
論文名稱:入侵人物車輛預警與辨識系統開發
論文名稱(外文):Invades the character vehicles early warning and the identification system development
指導教授:郝樹聲郝樹聲引用關係
指導教授(外文):Shu-Sheng Hao
學位類別:碩士
校院名稱:國防大學理工學院
系所名稱:電子工程碩士班
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:中文
論文頁數:81
中文關鍵詞:人車追蹤人臉辨識
外文關鍵詞:people trackingface recognition
相關次數:
  • 被引用被引用:1
  • 點閱點閱:160
  • 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:3
本論文提出一套利用影像處理方法建構以個人電腦為平台的即時影像追蹤及預警辨識系統,主要應用於營區庫房之門禁管制及管理,國軍目前正朝縮減兵力的政策執行,一套入侵人物預警及辨識系統可有效的精簡查核之衛哨人力,而且在時效上,更具有即時警告突發狀況的能力。
在本篇論文中,為求即時因此摒除複雜的演算法,嘗試利用基本的影像處理方法來達到監控的目的,本系統是使用影像相減法,得到每個影像區塊的移動資訊,分離出移動物體與背景,再利用影像前置處理方法消除雜訊、增強主體遮罩,以利擷取目標區域做後續的追蹤及辨識。
自動追蹤的系統部分,則是將擷取出移動物體的區域資訊,利用卡曼濾波器及核心函數演算法做追蹤及估測,本文也探討這兩種演算法的追蹤正確率及速率。最後將此擷取的人臉樣版資訊,利用統計的辨識分析方法,對此影像進行特徵的擷取,再與資料庫內人臉資訊做比對,達到辨識及警告的效果。
本文研究的主要目的是在探討常用的影像追蹤及辨識演算法的執行效率,根據本文所提出的理論實際進行不同的追蹤及辨識方法的研究,在不同情況下得到移動物體追蹤及辨識所需要的時間以及正確率,並將自動追蹤及辨識系統做整合,以期建立一套即時追縱及辨識之系統,待人車進入畫面時先行追蹤,再根據擷取的人臉做辨識,若非資料庫內人臉圖片則實施警告標示,利用最少的人力達到監視的效果。
In this thesis, we use two algorithms called Kalman filter and Core Function to track human and car in the scene. For security application purpose, we also apply recognition algorithms to detect the extracted persons. Our final goal is to construct a security system using in campus or other places. Once detecting an intruder or a car, this system will make an alarm to notify the guard. The speed of processing is the main concern in our research. We adopt the frame difference to accelerate the computation speed. With this simple computation, we can roughly extract the moving area. After using image pre-processing algorithms, we can not only denoise the area but also separate the object from the background. After the foreground has been extracted, it can be use as a mask for extracting the target area.
Successive targets are tracked by the Kalman filter and Core Function algorithms. In the thesis, we compare their performance such as tracking accuracy and computation speed. For human, we apply the face recognition algorithms on them to identify the specific person. We use statistical methods such as 2D Principal Component Analysis (2DPCA) and 2D Linear Discriminate Analysis (2DLDA) to recognize the person. In the recognition process, we have built a testing database beforehand. After the features of the human have been extracted, they are comparing with the features in the test database to look for the most similar person.
In summary, we use the frame difference method to extract the interesting area relate to the car or human. We adopt both the Kalman filter and Core Function as the tracking algorithms. In order to recognize the human, we apply several statistical algorithms to detect the person. We are also comparing the speed and accuracy about our proposed methods.
誌謝 ii
摘要 iii
ABSTRACT iv
目錄 v
表目錄 viii
圖目錄 ix
1. 緒論 1
1.1 研究動機與目的 1
1.2 研究系統流程 2
1.3 論文架構 5
2. 影像前置處理 6
2.1 色彩模式轉換 6
2.2 灰階影像 10
2.3 影像二值化 10
2.4 空間濾波 11
2.5 侵蝕、膨脹、斷開與閉合 12
2.6 影像投影 14
3. 物體偵測及追蹤 17
3.1 移動物體偵測 18
3.1.1 連續影像相減法(Temporal Differencing) 19
3.1.2 光流偵測法(Optical Flow) 20
3.1.3 背景相減法((Background Subtraction) 21
3.2 特徵區域選取 24
3.3 移動估測追蹤 28
3.3.1 卡曼濾波器演算法(Kalman Filter) 28
3.3.1.1卡曼濾波器在移動物體追蹤中的應用 33
3.3.2 核心函數物體追蹤演算法(Kernel-Based Object Tracking ) 36
3.3.2.1核心函數之定義 37
3.3.2.2目標色彩分佈密度函數的表示 38
3.3.2.3 Bhattacharyya相似係數 40
3.3.2.4平均位移法求得最高的相關係數 43
3.3.2.5平均位移法之實作流程 45
4. 預警及辨識 48
4.1 特徵擷取 48
4.1.1 二維主成分分析法(2DPCA) 48
4.1.2 二維線性鑑別分析法(2DLDA) 51
4.1.3 獨立成份分析(ICA) 54
4.2 特徵比對 55
4.2.1 歐基里德距離(Euclidean distance) 55
4.2.2 曼哈頓距離(City Block) 56
4.2.3 相關係數(Correlation Coefficient) 57
5. 實驗過程與分析 60
5.1 物體追蹤方面 60
5.1.1 室外單人走動 60
5.1.2 室外二人並排走動 62
5.1.3 室外多人走動 63
5.1.4 室外車輛移動 65
5.1.5 移動人物被遮蔽 66
5.1.6 效能分析 68
5.2 人臉辨識方面 70
6. 結論與未來工作 75
參考文獻 77
自傳 81
[1]Denman, S., Fookes, J., Cook, C., Davoren, C. and Mamic, A., “Multi-view Intelligent Vehicle Surveillance System,” IEEE International Conference on Video and Signal Based Surveillance AVSS '06, 2006.
[2]Kettnaker, V. and Zabih, R., “Bayesian multi-camera surveillance in Computer Vision and Pattern Recognition,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp.253-259, 1999.
[3]Chang, T.H. and Gong, S., “Tracking multiple people with a multi-camera system in Multi-Object Tracking,” IEEE Workshop on Multi-Object Tracking Proceedings, pp.19-25, 2001.
[4]Besada, M. and Casar, R., “Design of an A-SMGCS prototype at Barajas airport: airport surveillance sensors bias estimation in Information Fusion,” International Conference on 2005 8th, 2005.
[5]Stauffer, C. and W.E.L. Grimson, “Learning patterns of activity using real-time tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.747-757, Aug., 2000.
[6]Yutao, F. and Gelenbe E.,“ Adaptive object tracking and video compression with dynamic contours,” Proceedings of the Fifteenth National in Radio Science Conference. NRSC '98, 1998.
[7]Changick, K. and Jenq-Neng H., “A fast and robust moving object segmentation in video sequences,” International Conference on Image Processing. ICIP 99, pp.131-134, 1999.
[8]Changick, K. and Jenq-Neng H., “Fast and automatic video object segmentation and tracking for content-based applications,” IEEE Transactions on Circuits and Systems for Video Technology, pp.122-129, 2002.
[9]Lipton, A.J., Fujiyoshi, H. and Patil, R.S., “Moving target classification and tracking from real-time video,” Fourth IEEE Workshop on Applications of Computer Vision Proceedings. WACV '98, pp. 8-14, 1998.
[10]Vega-Riveros, J.F. and Jabbour, K., “Review of motion analysis techniques,” IEEE Proceedings on Communications, Speech and Vision, pp.397-404, Dec., 1989.
[11]Haritaoglu, I., D. Harwood, and Davis L.S., “ : real-time surveillance of people and their activities,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 809-830, Aug., 2000.
[12]Salmond, D., “Target tracking: introduction and Kalman tracking filters,” IEEE in Target Tracking: Algorithms and Applications (Ref. No. 2001/174), pp. 1-16, 2001.
[13]Welch, G. and Bishop, G., “An Introduction to Kalman filters,” Department of Computer Science University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3175, 2004.
[14]Comaniciu, D. and Meer P., “Mean shift: a robust approach toward feature space analysis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.603-619, 2002.
[15]Comaniciu, D., Ramesh, V. and Meer P., “Kernel-based object tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.564-577, May, 2003.
[16]鍾宜岑, “應用於動態背景中的移動物體影像之偵測與即時追蹤系統” ,碩士論文,國立交通大學,新竹,第21-34頁,2007。
[17]簡隆至, “即時移動物體偵測及自動追蹤系統” ,碩士論文,國立臺灣科技大學,台北,第47-76頁,2004。
[18]Comaniciu, D. and Meer, P., “Mean shift analysis and applications,” The Proceedings of the Seventh IEEE International Conference on Computer Vision, pp. 1197-1203, 1999.
[19]Kailath, T., “The Divergence and Bhattacharyya Distance Measures in Signal Selection,” IEEE Transactions on Communication Technology, pp. 52-60, Feb., 1967.
[20]Ying, W. and S. Pengfei. “Image PCA: A New Approach for Face Recognition. In Acoustics,” IEEE International Conference on Acoustics, Speech and Signal Processing. ICASSP 2007, pp. I-1241-I-1244, 2007.
[21]Ki-Chung, C., Seok Cheol K. and K. Sang Ryong., “Face recognition using principal component analysis of Gabor filter responses,” Proceedings. International Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, pp.53-57, 1999.
[22]Martinez, A.M. and Kak, A.C., “PCA versus LDA,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.228-233, Feb., 2001.
[23]Huiyuan, W., et al. “Face Recognition Combing Principal Component Analysis and Fractional-step Linear Discriminant analysis,” 2006 8th International Conference on Signal Processing, 2006.
[24]Xu, A., Jin, X., Jiang, Y. and Guo, P., “Complete Two-Dimensional PCA for Face Recognition,” IEEE International Conference on Pattern Recognition, Beijing, China, Vol. 3, pp. 481-484, 2006.
[25]Jian, Y., et al., “Two-dimensional PCA: a new approach to appearance-based face representation and recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.131-137, Jan., 2004.
[26]Kongsontana, S. and Rangsanseri, Y., “Face Recognition Using 2DLDA Algorithm,” IEEE Eighth International Symposium on Signal Processing and Its Applications, Bangkok, Thailand, Vol. 2, pp. 675-678, Aug., 2005.
[27]Ye, J., Ravi, J. and Li, Q., “Two-Dimensional Linear Discriminant Analysis,” Advances in Neural Information Processing Systems, Vol. 17, pp. 131-137, 2004.
[28]Aleix Martinez, M. and Avinash Kak, C., “PCA versus ICA,” IEEE Trans on Pattern Anal. Machine Intell, VOL. 23, NO. 2, Feb, 2001.
[29]Liwei, W., Yan Z. and Jufu F., “On the Euclidean distance of images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.1334-1339, Aug., 2005.
[30]繆紹綱,數位影像處理,全華科技書局,台北,2005。
[31]王精忠, “車牌辨識系統之研究” ,碩士論文,大同大學通訊工程研究所,台北,第23-26頁,2005。
[32]Yambor, W. S., “Analysis of PCA-Based and Fisher Discriminant-Based Image Recognition Algorithms,” Master Dissertation, Colorado State University, U.S.A., pp. 32-34, July, 2000.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top