臺灣博碩士論文加值系統

本論文以自走車人物追蹤為研究方向，以雙眼網路攝影機作為感測器模擬人類視覺系統，針對人物頭部特徵為追蹤目標，首先利用正規化RGB分析膚色範圍後，鎖定人臉區域作為橢圓比對範圍，隨之使用預先建立好的橢圓模組對人臉區域內的Sobel輪廓進行比對，匹配度最高者即為人頭所在。並使用雙眼視覺重疊視野將影像平面之二維目標物座標透過幾何關係與參數矩陣，計算出目標物的三維座標，即可得知深度資訊。最後為了達到追蹤控制，使用X定位與Y定位，控制法則分別為模糊與比例控制，進行自走車原地旋轉及前後移動，成功追蹤人物移動。

The study focuses on a system which can track people with the autonomous mobile robot. The system uses two calibrated cameras to simulate human vision system, called stereo vision, to acquire human head by two correlated images. First, we apply the normalized RGB color model to catch the human skin which is the matching range of ellipse model. Then, Sobel operator is employed to search the face-silhouette template database. Finally, we calculate the depth information with homography matrix and triangle method to provide the real-time 3D position of human head. We use this information to achieve feedback control of robot. The mobile robot is also applied with a fuzzy reasoning system and PD controller to follow human movement, and the experimental result shows the proposed system perform the tracking task successfully.

摘要 I
ABSTRACT II
目錄 III
圖表索引 VI
緒論 1
1.1 前言 1
1.2 文獻回顧 2
1.2.1 影像追蹤 3
1.2.2 距離偵測 3
1.2.3 移動控制 4
1.3研究目的與動機 4
1.4 論文架構 5
第二章 人物追蹤自走車系統架構 6
2.1 系統架構 6
2.2 車體結構 7
2.2.1 電源供給電路 9
2.2.2 馬達控制 10
2.2.3 影像裝置 12
2.3 軟體系統 13
第三章 人臉膚色偵測 15
3.1 光線補償 16
3.2 色彩空間轉換 18
3.2.1 RGB色彩空間 18
3.2.2 正規化RGB 19
3.2.3 YCrCb色彩空間 19
3.2.4 HSV 色彩空間 19
3.3 膚色檢測 21
3.4 膚色過濾 24
3.4.1 形態學處理 24
3.4.2 區域連通標籤化 26
3.4.3 特徵提取 28
第四章 橢圓人頭判定與追蹤 30
4.1 前置影像處理 31
4.1.1 高斯濾波器 31
4.1.2 Sobel測邊 32
4.2 橢圓模組比對 34
4.2.1 建立橢圓模組 34
4.2.2 橢圓模組匹配 37
4.3 位置估測 39
4.4 樣板比對 42
第五章 雙眼視覺測距 44
5.1 攝影機校正原理 45
5.1.1內部參數 46
5.1.2外部參數 47
5.2.1 內部校正 49
5.2.2 外部校正 51
5.3 內外部參數進行三維重建 52
5.3.1 三維重建原理 52
5.3.2 三維重建模擬 55
第六章 以立體視覺結合模糊理論之定位控制 59
6.1 模糊理論 60
6.1.1 模糊化 61
6.1.2 模糊知識庫 62
6.1.3 模糊推論引擎 62
6.1.4 解模糊化機構 63
6.2 追蹤定位設計 64
6.2.1 建立知識庫 64
6.2.2 模糊控制器模擬 67
第七章 實驗結果 69
7.1 橢圓人頭追蹤 69
7.1.1 以膚色為主體，位置估測為輔助 69
7.1.2 以位置估測為主體，膚色為輔助 72
7.2 雙眼測距 73
7.3 追蹤定位 75
第八章 結論與建議 77
8.1 結論 77
8.2 建議 78
參考文獻 79
附錄一 馬達控制板的指令格式說明 81
附錄二 車體接線圖 83

[1] K. Hirai, M. Hirose, Y. Haikawa andT.Takenaka ,
“The Development of Honda Humanoid Robot,”
Proceedings of IEEE International Conference on
Robotics & Automation, pp.1321-1326, May, 1998.
[2] Ben Tribelhorn and Zachary Dodds , “Evaluating the
Roomba: A low-cost, ubiquitous platform for
robotics research and education,” IEEE International
Conference on Robotics and Automation, pp.10-14,
April, 2007.
[3] B. Ugur Toreyin, Yigithan Dedeoglu, Ugur Gudukbay and
A. Enis Cetin, “Computer Vision-based Method for Real
-time Fire and Flame Detection,” Department of
Electrical and Electronics Engineering, August 2005.
[4] R. Polana and R. Nelson, “Low level recognition of
human motion,” in Proc. IEEE Workshop Motion of Non-
Rigid and Articulated Objects, pp. 77–82, Austin,
1994.
[5] C. A. Pau and A. Barber, “Traffic sensor using a
color vision method,” in Proc. SPIE—Transportation
Sensors and Controls: Collision Avoidance, Traffic
Management, and ITS, vol. 2902, pp. 156–165, 1996.
[6] D.-S. Jang and H.-I. Choi, “Active models for
tracking moving objects,” Pattern Recognit, vol. 33,
no. 7, pp. 1135–1146, 2000.
[7] Alexandre R.J. Francois, “CAMSHIFT Tracker Design
Experiments with Intel OpenCV and SAI,” Institute for
Robotics and Intelligent Systems, July 2004.
[8] Stan Birchfield, “An Elliptical Head Tracker,”
Computer Science Department Stanford University.
[9] Stan Birchfield, “Elliptical Head Tracking Using
Intensity Gradients and Color Histograms,” Computer
Science Department Stanford University.
[10] Rein-Lien Hsu, Anil K Jain and Mohamed Abdel-
Mottaleb, “Face Detection In Color Images,” IEEE
Transactions On Pattern Analysis And Machine
Intelligence, Vol.24,No.5, May 2002.
[11] Shang-Hong Lai, Chang_Wu Fu and Shyang Chang, ”A
Generalized Depth Estimation Algorithm with a Single
Image,”IEEE Transactions On Pattern Analysis And
Machine Intelligence, Vol.14,No.4,April 1992
[12] Tsai, R.Y., “An efficient and accurate camera
calibration technique for 3D machine vision ,” Proc
. International Conference on Computer Vision an
Pattern Recognition, 1986, pp. 364-374.
[13] Zhengyou Zhang, “A Flexible New Technique for Camera
Calibration,” December 2, 1998.
[14] Murray D, Little J., “Using Real-time Stereo Vision
for Mobile Robot Navigation,” Autonomous Robots,
2000, 8(2):161-171.
[15] Zadeh L.A., “Fuzzy Sets,” Information and Control,
Vol.8, pp.338 -357, 1965.
[16] Mamdani E.H., “Application Of Fuzzy Logic To
Apprpximate Reasoning Using Linguistic Synthesis, ”
IEEE Trans. Computers, Vol.C-26,1977.
[17] Soriano, M., B. Martinkauppi, S. Huovinen and M.
Laaksonen, “Skin color modeling under varying
illumination conditions using the skin locus for
selecting training pixels, ” Proceedings of IEEE
Nordic Signal Processing Symposium, Kolm�熳den, Sweden.
[18] R. Brunelli, and T. Poggio, “Template Matching:
Matched Spatial Filters And Beyond,” MIT AI Memo
1549,July 1995.
[19] 任紹棟, “具立體視覺測距之移動機器人碩士論文,” 國立台
灣科技大學, 民國96年。
[20] D. Forsyth and J. Ponce, “Computer Vision-A Modern
Approach,” Prentice Hall, 2003.
[21] 鍾國亮, “影像處理與電腦視覺,” 台北市, 東華書局, 民國
91年。
[22] http://sourceforge.net/projects/opencvlibrary/
[23] 孫宗瀛, 楊英魁, “Fuzzy 控制理論、實作與應用,” 台北
市, 全華圖書, 民國86年。