本研究之目的為提升印刷電子圖案邊緣量化的準確性與泛用性。為了達成圖案轉移完整性(pattern transfer completeness, PTC)於邊緣粗糙度(line edge roughness, LER)方面的評估,本文提出了一種新穎的方法,可以量化統計印刷圖案與其相應設計圖案之偏差,並依此開發圖案粗糙度量化演算法。 在本研究中,利用兩種噴墨印刷條件,製作阿基米德、對數與雙曲螺旋的規律但非對稱圖案,並分別展示不同條件下,圖案於演算法結果的LER差異,以顯示本方法在不同圖案與條件下的廣泛支援性與有效性。本演算法使用笛卡爾座標中的參數式來描述圖案與分析,而僅有微小且可忽略的評估誤差,優於既有方法。這項研究也包含數位影像、偵測範圍與其步進距離等不同影響的綜合分析。結果証明,本研究提出之方法不僅正確地反映了PTC,而且還展示了操作靈活性。例如,通過設定更大的偵測範圍或其步進距離,可以提高評估效率。
To evaluate the pattern transfer completeness (PTC) with respect to the line edge roughness (LER), this paper proposes an advanced method which statistically quantifies the deviations of printed patterns from their corresponding designed patterns. Asymmetric patterns of Archimedean, logarithmic, and hyperbolic spirals are demonstrated using inkjet printing under various conditions to show different LERs and the effectiveness of the advanced method. The advanced method analyzes patterns using parametric forms in Cartesian coordinates with negligible evaluation errors which outperform existing methods. This study involved comprehensive analyses of the impacts of digitized images, the detection range, and its movement. The results show that the proposed method not only correctly reflected the PTC, but also exhibited operational flexibilities, e.g. the evaluation efficiency may be enhanced by introducing an enlarged detection range or an increased movement of the detection range.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。