簡易檢索 / 詳目顯示

回結果列表
研究生： 蔡瑞堂
Tsai, Ruei-Tang
論文名稱： 以移動相機對運動物體之偵測
Moving objects detection with mobile cameras
指導教授： 田思齊
Tien, Szu-Chi
學位類別： 碩士
Master
系所名稱： 工學院 - 機械工程學系
Department of Mechanical Engineering
論文出版年： 2019
畢業學年度： 107
語文別： 中文
論文頁數： 57
中文關鍵詞： 相機移動估測 、運動目標物偵測 、樣板比對 、隨機取樣一致算法 、自適應二值化
外文關鍵詞： camera ego-motion estimation, moving objects detection, template matching method, Random Sample Consensus algorithm, adaptive thresholding method
相關次數： 點閱：109  下載：6 
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

    • 本研究建立一個以移動相機對運動物體偵測的影像處理系統。此系統於相機與運動目標物皆移動的情況下，達成運動目標物偵測的目的。在影像處理程序方面，我們先將影像裁切為多個區塊後，對連續影像使用樣板比對以估測每個區塊的移動情形。然後對樣板比對結果進行隨機取樣一致算法，以建立背景影像之仿射變換模型做為相機自我移動估測。經由持續對相機自我移動補償後的影像使用幀間差分法去除影像背景，再參考自適應二值化方法將影像分割成背景與運動目標物兩類，達成影像分割的效果。最後利用中值濾波與形態學法中的閉運算濾除樣板比對的誤差與影像雜訊，完成運動目標物的偵測。整體處理程序包含相機自我移動估測與運動目標物偵測，可在相機中斷(7幀/秒)內完成。實驗結果顯示，本論文建議之方法可即時補償相機自我移動，同時完成運動目標物偵測。

    In this study, a real-time image process system for detecting moving objects with a mobile camera is established. The system can detect moving objects when both the camera and the objects are moving. For image processing, any two successive images are cut into multiple blocks first and compared with template-matching method for each corresponding block to estimate their motion. Then, camera ego-motion is estimated by applying RANdom SAmple Consensus (RANSAC) algorithm on template-matching results gotten earlier to find the best affine transformation model of the backgrounds between two successive images. Next, the background is removed by using frame difference method and adaptive thresholding method to distinguish the background and the moving objects. At last, in order to filter out the error of camera ego-motion estimation and noises on image, median filter and morphology are used to intensify the contours of moving objects. It is noted that, the overall process, including camera ego-motion estimation and moving objects detection, can be done within 0.142 seconds (i.e., 7 fps). Experimental result shows that, with the proposed method, the estimation of camera ego-motion and the detection of moving objects can be completed in real time.

    圖目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 表目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 符號表. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 第一章緒論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 第二章影像偵測與處理. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.1 相機自我移動估測. . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.1.1 樣板比對法. . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.1.2 背景特徵提取. . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.2 運動目標物偵測. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.2.1 幀間差分法. . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.2.2 影像二值化. . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.2.3 中值濾波. . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.2.4 形態學法. . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 第三章實驗設備與系統架構. . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.1 硬體. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.2 軟體. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.3 實驗場景與參數設定. . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.3.1 實驗場景. . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.3.2 影像參數設定. . . . . . . . . . . . . . . . . . . . . . . . . . 35 第四章實驗結果與討論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.1 以靜止相機對運動目標物(靜對動)偵測實驗. . . . . . . . . . . . . 42 4.2 以移動相機對運動目標物(動對動)偵測實驗. . . . . . . . . . . . . 44 第五章結論與未來展望. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.1 結論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.2 未來展望. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

    [1] X. Fan et al. A combined 2d-3d vision system for automatic robot picking.Proceedings of the 2014 International Conference on Advanced Mechatronic Systems, pages 513-516, 2014.
    [2] Cheng Xing and Jing-Jing Huang. An improved mosaic method based on sift algorithm for uav sequence images. 2010 International Conference On Computer Design and Applications, 2010.
    [3] G. Spampinato et al. Stereo vision based navigation for automated vehicles in industry. 2009 IEEE Conference on Emerging Technologies Factory Automation, pages 1-4, 2009.
    [4] M. P. Nkosi et al. Autonomous pedestrian detection. pages 1-5, 2015.
    [5] N. Yamamoto and N. Uchida. Improvement of image processing for a collaborative security flight control system with multiple drones. 2018 32nd International Conference on Advanced Information Networking and Applica-
    tions Workshops (WAINA), pages 199-202, 2018.
    [6] H. K. Chavda and M. Dhamecha. Moving object tracking using ptz camera in video surveillance system. 2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), pages 263-266, 2017.
    [7] 蘇庭毅. 以現場可程式化閘陣列實現追蹤移動目標之快速影像處理. Master's thesis, 國立成功大學, 2018.
    [8] JiaWei Tang et al. FPGA-based real-time moving target detection system for unmanned aerial vehicle application. 2010 International Conference on Image Analysis and Signal Processing, 2016.
    [9] Z. Xu, D. Zhang, and L. Du. Moving object detection based on improved three frame di fference and background subtraction. 2017 International Conference on Industrial Informatics - Computing Technology, Intelligent Technology,
    Industrial Information Integration (ICIICII), pages 79-82, 2017.
    [10] Z.Wang and X. Yang. Moving target detection and tracking based on pyramid lucas-kanade optical flow. 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), pages 66-69, 2018.
    [11] C.K. Wang. Design and implementation of a multi-purpose real-time visual tracking system based on modified adaptive background substration and multi-cue template matching. Master's thesis, National Cheng Kung University, 2004.
    [12] P. Anandan. Measuring visual motion from image sequences. University of Massachusetts Amherst, 1987.
    [13] Z. Zhang et al. A method for 2d image texture synthesis based on lbp. 2009 International Conference on Arti ficial Intelligence and Computational Intelligence, 2009.
    [14] R.C. Gonzalez and R.E. Woods. Digital Image Processing. Upper Saddle River, 2008.
    [15] H. Liu and X. Hou. Moving detection research of background frame difference based on gaussian model. 2012 International Conference on Computer Science and Service System, pages 258-261, 2012.
    [16] N. Otsu. A threshold selection method from gray-level histograms. IEEE Trans. Sys., Man., 9(1):62-66, 1979.
    [17] Luc De Vos and Michael Stegherr. Parameterizable vlsi architectures for the full- search block-matching algorithm. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 36(10), 1989.
    [18] L.G Chen H.M Jong and T.D Chiueh. Parallel architectures for 3-step hierarchical search block-matching algorithm. IEEE Transctions on Circuits and System for Video Technology, 4(4):407-416, 1994.
    [19] Tsai Lung-Yu. Image processing (median filter). https://reurl.cc/RZZGD.
    [20] PMI KM33-05A-400N0-0331. http://www.nsk-cs.com.tw/?f=km-mono-stage.
    [21] Logitech HD Webcam C310. https://www.logitech.com/zh-tw/product/hd-webcam-c310.
    [22] Pololu 25D Metal LP 6V with 48CPR Encoder. https://www.pololu.com/product/2285.
    [23] How to set exposure time of Logitech C922 webcam on OpenCV?
    https://reurl.cc/ANrvp.

    下載圖示 校內：2021-08-19公開
    校外：2021-08-19公開
    QR CODE