透過您的圖書館登入
IP:3.15.147.53
  • 學位論文

非接觸式三次元量測量測能力評估

Evaluation of the Measurement Capability for a 3D Non-Contact Measurement System

指導教授 : 江行全
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


本研究之目的在建立一套非接觸式三次元量測能力評估方法,以三次元座 標量測儀(CMM)做為校正雷射三角測頭加CCD相機三次元量測系統的工具 。非接觸式量測系統是由雷射三角測頭及CCD相機所組成,此系統以夾具 固定在CMM之接觸式探頭上,我們以經由拉丁方格法設計的工件模組為被 測物,調整相機的光軸和雷射射出光軸均與塊規表面垂直,並且控制光源 與背景顏色等環境因素,利用影像處理技術,獲得被測物的輪廓及量測目 標點之座標,結合由雷射探頭所獲得之Z軸座標,完成3D之座標量測,再 透過座標轉換與三次元量測儀的數據做單點比較,計算3D空間之距離差距 ;而尺寸之量測誤差則是以量測值與CMM測得的尺寸比較長度的差異。單 點比較時的量測能力指標,是以各點的單軸誤差計算誤差統計量, C語言 撰寫的程式以此進行統計分析,做變異數分析和檢定,決定顯著因子和後 續實驗,以此計算非接觸式量測系統之精確度( accuracy )和重複性( repeatability ),並繪製各軸差異圖與統計圖表;尺寸比較時,則以尺 寸的差距,計算精確度和重複性,繪製尺寸差異散佈圖。由本研究的實驗 結果顯示,在單點比較的評估程序時,應將評估的對象,限定在正在使用 中的拉丁方格和轉換矩陣上;而尺寸的誤差不是尺寸長度的多項式函數。 在本研究所做過的實驗中,單點比較時,以整體(包括全部的區塊)的精確 度和重複性來看,實驗四以多重迴歸法解轉換矩陣,測試原有的拉丁方格 的情況下,accuracy為0.028557mm,repeatability為0.027767mm,可代 表此非接觸量測系統,單點比較時整體的量測能力指標;而在尺寸比較時 ,實驗四的第三個拉丁方格,其尺寸差距的 accuracy 為 -0.011864mm, repeatability為0.039242mm,可代表非接觸量測尺寸的量測能力指標。 @ 最後對本系統的誤差做探討,並儘可能的降低誤差,使系統的量測能 力在特定的量測條件下能達到最高。

並列摘要


The objective of this research was to develop an evaluation methodology for the 3D measuring capability of non-contact measurement systems. We use three dimension coodinate measuring machine(CMM) as a reference for non-contact measurement system. The non-contact measurement system which consists of a laser probe and a CCD camera, is fixed on the CMM using a specfic fixture,and the measurement module is a set of workpiece arranged as a Latin-square. Some operational conditions of the experiment are as follows : 1.)To adjust the optic axis of camera and laser emitter to be vertical with the surface of the measurement module. 2.)To control source of illumination,color of background to get stable image. 3.)To acquire calibration data before starting experiment. The 3D coordinates of the target was obtained from image processing technology and distance measured by the laser probe. These coordinates,through coordinate transformation, can be compared with the coordinates measured by CMM. A statistical analysis program was used to analyze the experiment data. The program consists of two parts, the ANOVA and the relative efficiency. The measurement capability was represented in terms of accuracy and repeatability, by each axis separately or collectively based on the analysis results. Finally, some error sources were discussed and future research were suggested.

延伸閱讀