應用機器視覺於微型照相模組之瑕疵檢測

Automated Inspection of Compact Camera Module Lens Using Computer Vision

王詩婷(Shih-Ting Wang)

指導教授 ： 田方治

國立臺北科技大學/管理學院/工業工程與管理系碩士班/碩士(2013年)

摘要

隨著電子科技業與光學產業的蓬勃發展，相機鏡頭廣泛的應用於手機鏡頭、掌上型電腦、個人筆記型電腦等。相機鏡頭是由多片鏡片組合而成，在其生產過程當中容易產生細微瑕疵，包含刮傷、毛絲、鏡片髒等。多數企業對於相機鏡頭瑕疵檢測方式仍以傳統人工目視方式檢測，透過多次移動、不同方向與重複性檢測，才能檢測一顆CCM鏡頭。由於CCM鏡頭大多數的瑕疵都非常小且多樣性，使人眼無法勝認太細微的瑕疵缺陷，並且在檢測人員主觀性的判斷性，不易控制產品品質的檢測標準。
因此本研究應用機器視覺模仿人類視覺檢測之光學系統，結合工業相機、光業鏡頭、光源設備與XYZ軸控、程式介面開發與影像處理軟體建構一套CCM鏡頭表面瑕疵檢測系統。

關鍵字

影像處理 ； 機器視覺 ； 手機鏡頭 ； 均值漂移平均

並列摘要

With the flourishing electronic industry and optical industry, compact camera module (CCM) lenses are widely applied in lenses of mobile phones and personal computers. CCM lenses consist of multiple lenses and manufacturing may create defects. Most enterprises still use the human eye to inspect for defects, through repeatedly moving different directions and reiterative inspection. However, inspectors can’t control for slight defects because many CCM lens defects are tiny and diversified.
Consequently, the purpose of this study is to develop a CCM lens defect inspection system to replace the human eye; this automated inspection will combine CCD, lens, lighting, XYZ control table, programming interface development and image processing.

並列關鍵字

Image Processing ； Machine Vision ； CCM Lens ； Mean Shift

參考文獻

[21]大立光電股份有限公司財經百科網頁：

[1]林義祥，機器視覺應用於照相手機鏡片模組邊緣瑕疵檢測，碩士論文，國立交通大學工業工程與管理學系研究所，新竹，2012。

[20]Hong-Dar Lin, Wan-Ting Lin, and Huan-Hua Tsai, Automated Industrial Inspection of Optical Lenses Using Computer Vision, Chaoyang University of Technology, Taiwan, 2011.

[11]Dongping Ming, Tianyu Ci, Hongyue Cai, Longxiang Li, Cheng Qiao and Jinyang Du, "Semivariogram-Based Spatial Bandwidth Selection for Remote Sensing Image Segmentation With Mean-Shift Algorithm," Geoscience and Remote Sensing Letters, 2012, pp. 813-817.

[13]Fukunaga and Hosterler, "The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition," Information Theory, 1975, pp. 32-40.