透過您的圖書館登入
IP:3.88.185.100
  • 學位論文

觸控面板之自動化光學檢測系統的研製

Design of Automated Optical Inspection System of Touch Panels

指導教授 : 林宏達
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


近年智慧型手機及平板電腦的使用日漸普及,使觸控面板之需求量大增,而常用的電容式觸控面板(Capactive touch panel, CTP)其表面具有規則性線路所形成的結構性紋路,這也增加面板之表面瑕疵檢驗的困難度。觸控面板上的表面瑕疵可分為線型瑕疵及面型瑕疵,其中面型瑕疵包含髒汙、水痕、氣泡等涵蓋範圍較大的瑕疵,此類瑕疵具有低對比、明暗度漸層變化、不規則狀且不具方向性等特性,而線型瑕疵大多為刮傷造成,且具有方向性,在產線中皆有可能產生面型及線型瑕疵,若要同時以同一程序找出兩種不同類型之瑕疵,則有一定的難度。 因此本研究主要針對觸控面板之表面瑕疵提出一自動化瑕疵檢驗方法,利用曲波轉換 (Curvelet Transform, CT)特性搭配多角度濾波方式進行觸控面版之背景紋路的衰減,並將濾波後曲波域影像進行重建至空間域中,以管制界線法進行二值化閥值的決定,並將影像分割為瑕疵及背景。本研究初步針對60(20張線型瑕疵、20張面型瑕疵、20張正常影像)張小樣本影像進行參數訓練,訓練後參數針對大樣本154張影像進行瑕疵檢測,其中線型及面型瑕疵各59張,而正常影像36張,實驗結果顯示可有效判斷瑕疵是否存在與其發生的位置,其瑕疵檢出率(1-β)可達93.33%,瑕疵誤判率(α)為1.26%,正確判斷率(CR)為98.69%。

並列摘要


Capacitive touch panels (CTP) with advantages of water-proof, stain-proof, scratch-proof, fast response, are widely used in various electronic products. The surface of CTPs are multi-layer structured and are classified as structural textures. It is a difficult inspection task when defects embedded in surface of CTPs with structural textures. The surface defects are usually classified into linear and area types of defects. The area type includes dirt, water marks, bubbles and other defects with larger scope. Such defects have low contrast, brightness with gradient changes, irregular and non-directional shapes. The linear type defect is commonly caused by scratch and has directionality. Both of the defect types may appear simultaneously on product surfaces in the CTP production line. In this study, we propose a novel approach to detect the two types of defects on CTP surfaces. The proposed method applies the Curvelet Transform (CT) with multi-angle filtering to remove the structural textures of background and delete the angle direction of background texture. Then the filtered image is inversely transformed back to spatial domain. In the reconstructed image, the background texture is attenuated and the defects are enhanced. Finally, a thresholding value is determined by statistical control limits and the restored image can be easily segmented to into two categories namely dark defects, and white background. Experimental results show that the proposed method achieves a high 92.26% flaw detection rate (1-β), a low 1.32% false alarm rate(α), a high 98.61% correct classification rate (CR).

參考文獻


[59] 黃智政,「應用區塊離散餘弦轉換搭配灰預測模式檢測車用鏡面玻璃表面瑕疵」,碩士論文,朝陽科技大學工業工程所,台中(2009)。
[61] 蔡環樺,「觸控面板之自動化表面瑕疵檢測」,碩士論文,朝陽科技大學工業工程與管理所,台中(2012)。
[57] 陳韋吉,「透明玻璃之自動化可視瑕疵檢測」,碩士論文,朝陽科技大學工業工程所,台中(2011)。
[55] 陳興倫,「使用小波包分解與部分最小平方法於光學元件之可視瑕疵檢測」,碩士論文,朝陽科技大學工業工程與管理所,台中(2007)。
[53] 唐珮玲,「應用HHT於曲面光學元件之可視瑕疵檢測」,碩士論文,朝陽科技大學工業工程與管理所,台中(2012)。

被引用紀錄


張育凡(2015)。應用機器視覺系統檢測高滲透壓刀輪切割 TFT-LCD 玻璃後斷面之研究〔碩士論文,國立中央大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0031-0412201512043287
林宥凱(2015)。車用後視鏡之輪廓瑕疵檢測〔碩士論文,朝陽科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0078-2502201617130116

延伸閱讀