簡易檢索 / 詳目顯示

研究生: 劉韋辰
weichen Liu
論文名稱: 基於FPGA之單移動目標物 歷史軌跡方向即時辨識系統
FPGA-Based Architecture for Moving Direction Recognition of a Target Object Using Historical Trajectories
指導教授: 許陳鑑
Hsu, Chen-Chien
學位類別: 碩士
Master
系所名稱: 電機工程學系
Department of Electrical Engineering
論文出版年: 2011
畢業學年度: 99
語文別: 中文
論文頁數: 73
中文關鍵詞: 電腦視覺光流移動估測歷史軌跡FPGA
英文關鍵詞: Computer Vision, Optical Flow, Motion Estimation, Historical Trajectories, FPGA
論文種類: 學術論文
相關次數: 點閱:197下載:7
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 本論文主要係利用FPGA晶片及CMOS影像感測器裝置,實現一具有視覺方向感測認知之嵌入式智慧型系統,作法上係以單移動目標物歷史軌跡方向辨識為基礎,提出一種類似於人類視覺隱含有時間概念的演算方式,利用在不同時階(Time-Scale)所紀錄之物體移動所造成的歷史軌跡(Historical Trajectories),完成移動方向辨識計算,實驗結果證明本方法具有足夠的精確度,可以達成利用視覺指令控制人機介面在應用上的需求。論文整體架構大致上可分為:(1)以軟體實現整個辨識系統的演算法、(2)將系統實現於Altera FPGA開發版之晶片設計,希望藉由優異的演算方式,突顯其利於實現於嵌入式裝置的優勢。

    The purpose of this paper is to implement a vision-based embedded system of a moving-direction recognition system using FPGA chip and CMOS image sensor. Based on historical trajectories of a single moving object, an algorithm to recognize the moving direction of the object is proposed in this thesis similar to the concept of time scale of human vision, where historical trajectories of the object in different time scale are recorded to determine the moving direction. Experimental shows that this method achieves a good level of accuracy for various embedded applications in controlling human-machine interfaces. The architecture of the thesis can be divided into two parts: (1) Simulate the proposed algorithm by software. (2) Implement the algorithm on Altera FPGA development board.

    中文摘要 i 英文摘要 ii 誌謝 iii 目錄 iv 表目錄 vii 圖目錄 viii 第一章 緒論 1  1.1 研究動機與背景 1  1.2 研究目的 1  1.3 研究方法 3  1.4 研究限制 3  1.5 研究步驟 4 第二章 文獻探討與回顧 6  2.1 Mean-Shift算法 7  2.2 光流法(Optical Flow) 8   2.2.1 Horn-Schunck算法 8   2.2.2 Lucas-Kanade算法(線性最佳近似解) 8    (1) Lucas-Kanade算法原理 16    (2) LK金字塔 19 第三章 系統架構設計及軟體模擬 20  3.1 背景知識(色彩空間的轉換) 20  3.2 影像前處理 21   3.2.1 影像侵蝕 21   3.2.2 影像膨脹 23  3.3 歷史軌跡方向辨識系統流程 26   3.3.1 色彩空間轉換與即時影像相減 27   3.3.2 中值門檻濾波(Median Filtering) 28   3.3.3 影像開運算(Opening Operation) 29   3.3.4 動態能量遞減法(Momentum Decreasing) 30   3.3.5 動態方向辨識(Moving Direction Identification) 33 第四章 硬體設計平台 37  4.1 多媒體開發板及周邊元件介紹 37  4.2 CMOS影像擷取模組介紹 39   4.2.1 CMOS影像擷取模組特性 40   4.2.2 D5M影像擷取模組腳位說明 40   4.2.3 D5M影像擷取格式 41   4.2.4 D5M影像擷取模組之資料傳輸格式 43  4.3 LTM顯示模組 44   4.3.1 LTM顯示模組之特性 45   4.3.2 LTM顯示模組之腳位說明 46 第五章 歷史軌跡方向辨識系統在FPGA之實現 48  5.1 色彩空間轉換 49  5.2 即時動態偵測 50   5.2.1 即時影像相減模組 51   5.2.2 中值門檻濾波模組 53  5.3 歷史軌跡模組 54  5.4 動態移動方向辨識模組 58   5.4.1 水平掃描計算模組 60   5.4.2 垂直掃描計算模組 63 第六章 實驗結果 65 第七章 結論與未來研究方向 69  7.1 結論 69  7.2 未來研究方向 70 參考文獻 71

    [1] Karl Schwerdt and James L. Crowly, “Robus Face Tracking using Color,” IEEE International Conference on Automatic Face and Gesture Recognition, vol.26, pp. 90-95, Jun. 2000.
    [2] Gary R. and Santa Clara, “Computer Vision Face Tracking for Use in a Perceptual User Interface,” Intel Technology Journal, vol.2, pp. 9-15, 1998.
    [3] 吳怡明(98年),手勢辨識應用於遙控音樂播放系統,國立台灣科技大學電機工程系碩士論文。
    [4] 洪兆欣(95年),以軌跡辨識為基礎之手勢辨識系統,國立中央大學資訊工程研究所碩士論文。
    [5] 陳奕豪(99年),基於人臉追蹤之人機介面,義守大學資訊工程研究所碩士論文。
    [6] 劉東樺(98年),以適應性膚色偵測與動態歷史歸機影像為基礎之即時手勢辨識系統,大同大學資訊工程研究所碩士論文。
    [7] Ishfaq Ahmad, Weiquo Zheng, Jiancong Luo and Ming Liou, “A Fast Adaptive Motion Estimation Algorithm,” IEEE Transaction on Circuits and Systems For Video Technology, vol.16, no.3, Mar. 2006.
    [8] Berthold K.P. Horn and Brian G. Schunck, “Determining Optical Flow,” Artificial Intelligence, vol. 17, pp. 185-203, 1981.
    [9] Philippe Thevenaz, Thierry Blu and Michael Unser, “Interpolation Revisited,” IEEE Transactions on Medical Image, vol. 19, pp. 739-755, Jul. 2000.
    [10] Gary Bradski and Adrian Kaehler, Learning OpenCV, O'Reilly Media, 1rd Edition, Sep. 2008.
    [11] Daniel Ashlock, Eun-Youn Kim and Wendy Ashlock, “Fingerprint Analysis of the Noisy Prisoner’s Dilemma,” IEEE Congress on Evolutionary Computation, vol. 1, pp. 41-48, Sep. 2007.
    [12] Changchang Wu, Friedrich Fraundorfer, Jan-Michael Frahm, Jack Snoeyink and Marc Pollefeys, “Image Localization in Satellite Imagery with Feature-Based Indexing,” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Science. vol.37, pp.197-202, 2008.
    [13] Fukunaga, Keinosuke, Larry D. Hostetler, “The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition,” IEEE Transactions on Information Theory, vol.21, no.1, pp.32-40, 1975.
    [14] Bruce D. Lucas and Takeo Kanade, “An iterative image registration technique whit an application to stereo vision,” International Joint Conference on Artificial Intelligence, vol.2, pp. 674-679, 1981.
    [15] Zoltan Prohaszka, “Some Improvements to the Lucas-Kanade Optical Flow Detector in Motion Tracking,” Serbian-Hungarian Joint Symposium on Intelligent Systems, pp. 503-512, 2006.
    [16] Andres Bruhn, Joachim Weickert and Christoph Schnorr, “Lucas/Kanade Meets Horn/ Schunck: Combining Local and Global Optical Flow Methods,” Internation Journal of Computer Vision, vol.61, pp. 211-231, 2005.
    [17] Lorenz Gerstmayr, An Improvement of the Lucas-Kanade Optical-Flow Algorithm for every Circumstance, Computer Engineering Group Faculty of Technology University of Bielefeld, Jun. 2008.
    [18] R. C. Gonzalez and R. E. Woods, Digital Image Processing, Prentice Hall, 3rd Edition, 2008.
    [19] Peter J. Ashenden, An Embedded Systems Approach Using VHDL, 2008.
    [20] 鄭信源(2003年出版,儒林圖書) Verilog硬體描述語言數位電路設計實務。
    [21] 林灶生(2008年出版,全華圖書) Verilog FPGA晶片設計。
    [22] Altera 多媒體發展平台DE2-70網址,URL:http://university.altera.com/materials/boards/
    [23] 友晶科技網址,URL:http://www.terasic.com.tw/
    [24] Terasic, TRDB_D5M_UserGuide, Document Version 1.0, 2008.
    [25] Terasic, TRDB_LTM_UserGuide, Document Version 1.22, 2007.
    [26] Pramod Chandraiah, Hans Gunar Schirner, Nirupama Srinivas and Rainer Doemer, System-On Chip Modeling and Design, CECS Technical Report, Jun. 2004.
    [27] Udaya Kamath and Rajita Kaundin, System-on-Chip Designs Strategy for Success, Wipro Technologies, Jun. 2001.
    [28] Tianguang Zhang, Haiyan Wu and Alexander Borst, “An FPGA Implementation of Insect-Inspired Motion Detector for High-Speed Vision Systems,” IEEE International Conference on Robotics and Automation, vol.52, pp. 335-340, May. 2008.
    [29] Javier Diaz, Eduardo Ros, Francisco Pelayo, Eva M. Ortigosa, and Sonia Mota, “FPGA-Based Real-Time Optical-Flow System,” IEEE Transaction on Circuits and Systems For Video Technology, vol.16, pp. 274-279, Feb. 2006.
    [30] Zhaoyi Wei, Dah-Jye Lee and Brent E. Nelson, “FPGA-based Real-time Optical Flow Algorithm Design and Implementation,” Journal of Multimedia, vol.2, no.5, pp.38-45, Sep. 2007.
    [31] Kumara Ratnayake and Aishy Amer, “An FPGA-Based Implementation of Spatio-Temporal Object Segmentation,” IEEE International Conference on Image Processing, vol.1, pp. 3265-3268, Oct. 2006.

    下載圖示
    QR CODE