【正文】 the basic steps of edge detection in this topic. In the implementation methods of each step, pare the advantages and disadvantages of Roberts, Sobel, Prewitt, Canny and other operators. To get the clear and true contour, I use the Hough transform method. In addition, in order to further improve the clarity of the images obtained, the calibration of the CCD is done. Finally the performance of various parts of the image processing system are analyzed and summarized. Training Objectives: using VC and MATLAB software programming skillfully, mastering the knowledge of image processing and OpenCV and cultivating the analyzing and problemsolving skills at the same time, to provide solid foundation for future scientific research. Keywords: Edge detection ， Canny operator， CCD calibration， MATLAB， OpenCV 目錄 摘 要 .............................................................................................................. I ABSTRACT .................................................................................................. II 1 緒論 ............................................................................................................. 1 圖像處理技術(shù)研究的背景和意義 ......................................................................1 基于數(shù)字圖像處理的檢測(cè)系統(tǒng) ..........................................................................2 數(shù)字圖像檢測(cè)技術(shù)的國(guó)內(nèi)外研究現(xiàn)狀 ..............................................................3 本文的主要研究?jī)?nèi)容及采用的方法 ..................................................................4 2 圖像檢測(cè)的實(shí)現(xiàn) ......................................................................................... 5 數(shù)字圖像處理的概念 ..........................................................................................5 數(shù)字圖像 ....................................................................................... 5 數(shù)字圖像處理的三個(gè)層次 ........................................................... 6 數(shù)字圖像處理的 特點(diǎn) .................................................................... 7 本課題邊緣檢測(cè)基本步驟 ..................................................................................8 感興趣區(qū)域的分割 ....................................................................... 8 圖像增強(qiáng) ......................................................................................11 邊緣檢測(cè) ............................................................................ 17 定位測(cè)量 ..................................................................................... 22 單位標(biāo)定 ..................................................................................... 25 本章小結(jié) ............................................................................................................26 3 CCD 圖像的讀取與標(biāo)定 .......................................................................... 27 攝像機(jī)模型 ........................................................................................................27 基于 OPENCV 的標(biāo)定 ........................................................................................30 讀取標(biāo)定用的一組圖像 ............................................................. 30 對(duì)讀入的圖像進(jìn)行角點(diǎn)檢測(cè) ..................................................... 32 格式轉(zhuǎn)換 ..................................................................................... 40 計(jì)算內(nèi)參矩陣 ............................................................................. 40 計(jì)算外參數(shù)矩陣 ......................................................................... 41 分析實(shí)驗(yàn)結(jié)果 ............................................................................. 41 本章小結(jié) ............................................................................................................41 4 算法實(shí)現(xiàn)及軟件開發(fā) ............................................................................... 43 用到的重要算法 ................................................................................................43 中值濾波 ..................................................................................... 43 SOBEL 圖像增強(qiáng) ........................................................................ 43 閾值分割 ..................................................................................... 44 輪廓跟蹤 ..................................................................................... 45 Canny 算法步驟 .......................................................................... 46 Hough 變換 ............................................................................... 49 OPENCV 庫(kù)的運(yùn)用 ...........................................................................................49 OpenCV 模塊 ............................................................................... 50 OpenCV 的特點(diǎn) ........................................................................... 50 OpenCV 的應(yīng)用 ........................................................................... 51 程序代碼 ............................................................................................................53 Window ..................................................................... 53 基于 OpenCV 的 Canny 算子實(shí)現(xiàn) ............................................. 54 軟件實(shí)現(xiàn) ............................................................................................................56 本章小結(jié) ............................................................................................................61 5 總結(jié)與展望 ............................................................................................... 62 本次畢設(shè)的成果 .................................................................................................62 畢設(shè)存在的問題 ................................................................................................62 感想與體會(huì) ........................................................................................................63 致 謝 ............................................................................................................. 64 參考文獻(xiàn) ............