TFT液晶顯示面板的Mura瑕疵偵測

Abstract

在本文中，我們提出一個能夠自動偵測LCD面板上四種不均勻(Mura)瑕疵型態的演算法。這四種瑕疵是分別是群聚型瑕疵、垂直塊狀瑕疵、刮痕狀瑕疵、以及漏光瑕疵。要偵測群聚型瑕疵，我們使用了Laplacian of Gaussian (LOG)濾波器。要偵測垂直塊狀型瑕疵，我們檢查原始影像一維投影的曲率變動。要偵測刮痕狀瑕疵，我們設計了一個頻率域上的濾波器來偵測特定的頻率成分。要偵測漏光瑕疵則是利用影像鏡面的方式並且採用偵測與群聚狀瑕疵相同的LOG濾波器方法。這四種不均勻的瑕疵偵測方式被整合成一個有效系統。實驗的結果證明這些演算法的確可以很有效率地偵測出LCD面板上的這四種Mura瑕疵。

In this thesis, we propose an automatic inspection system, which can automatically detect four types of Muras on an LCD panel: Cluster Mural, V-band Mura, Rubbing Mura, and Light Leak Mura. To detect cluster Mura, the Laplacian of Gaussian (LOG) filter is used. To detect v-band Mura, we check the variation tendency of the projected 1-D intensity profile. To detect rubbing Mura, we designed a frequency mask to detect distinct components in the frequency domain. To detect light leak Mura, we apply image mirroring and adopt the same LOG filter used in detecting cluster Muras. All four types of Mura detection are integrated together into an efficient system. Simulation results demonstrate that the proposed automatic Mura detection algorithms can efficiently detect these four types of mura defects on LCD panels.