查詢結果分析
來源資料
相關文獻
- 兩種晶圓缺陷辨識技術之比較
- Quantitative Comparison between Artificial Neural Networks and Bilinear Interpolation to Predict a Real Robot's Sonar Sensor Readings
- An Implementation of Distributed Framework of Artificial Neural Network for Big Data Analysis
- Automatic Classification System for Identifying Spatial Pattern
- 晶圓針測之測試條件設定與誤宰診斷
- 運用人工智慧發掘具訴訟風險專利
- 深度學習在Smart Beta交易策略之應用
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
頁籤選單縮合
題 名 | 兩種晶圓缺陷辨識技術之比較=Comparison of Two Wafer-Scale Defect Cluster Identifiers |
---|---|
作 者 | 黃振榮; | 書刊名 | 臺東師院學報 |
卷 期 | 13(上) 民91.06 |
頁 次 | 頁1-22 |
分類號 | 448.552 |
關鍵詞 | 晶圓針測; 缺陷偵測; 中間值過濾器; 近鄰分羣; 類神經網路; 多層感知器; Wafer probing; Defect detection; Median filter; Nearest neighbor clustering; Neural networks; Multilayer perceptro; |
語 文 | 中文(Chinese) |
中文摘要 | 在電氣測試階段,晶圓上的每顆晶粒階須經過檢測以篩選出合格之晶粒。然而此 類之檢測儀器仍無法偵測出晶圓上的所有缺陷,諸如刮痕或污點等等等。為確切偵測出此類 檢測儀器無法辨識之缺陷,晶圓測試公司必須安排五到十?員工預人工檢測錯誤方式一一檢 視每個晶圓並確切標示出缺陷的所有晶粒。為了節省人事支出並提高工作效率,我們特別提 出兩個晶圓缺陷檢測自動化程式並加以比較之。其中一個程式是採用影像處理技術中的中間 值過濾器而另一個則應用了類神經網路中的多層感知器技術。實驗結果顯示,我們提出的兩 個作法中,其中以影像處理技術為主的程式可偵測出絕大多數的晶圓缺陷並標示其內含之晶 粒,另外一個以類神經網路技術為主的程式則可偵測出實驗樣本的每個缺陷。雖然這兩種技 術會將百分之一的合格晶粒歸類為缺陷晶粒,但對晶圓測式公司於成本考量上是可以接受 的。目前該公司正使用以影響處理技術為主的程式來檢測晶圓缺陷,並節省了每月十萬美元 之支出。 |
英文摘要 | During an electrical testing stage, each die on a wafer must be tested to determine whether it functions as it was originally designed. In the case of a clustered defect on the wafer, such as scratches, stains, or localized filed patterns, the tester may not detect all of the defective dies in the flawed area. TO avoid the defective dies proceeding to final assembly, a testing factory must assign five to ten workers to check the wafers and hand mark the defective dies in the flawed region or close to the flawed region. This work compares two automatic wafer-scale defect cluster identifiers. One employs median filtering technique, and the other uses a multilayer perceptron to detect the defect cluster and mark all the defective dies. The experimental results illustrate that the multilayer perceptron approach is very effective in detect identification and achieves better performance than the median filter technique. Currently, the testing factory uses the median filter approach to identify the wafer defects automatically and saves up to $100,000 each month. |
本系統中英文摘要資訊取自各篇刊載內容。