Image Processing Techniques for Correlation of Photoluminescence Images of As-Cut Wafers with Final Cell IV Parameters
| Autor: | True, B., Stavrides, A., Latchford, I. |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2011 |
| Předmět: | |
| DOI: | 10.4229/26theupvsec2011-2cv.2.38 |
| Popis: | 26th European Photovoltaic Solar Energy Conference and Exhibition; 2095-2098 We present an image processing algorithm for photoluminescence (PL) images of multi-crystalline silicon wafers. The algorithm takes advantage of existing machine vision techniques to generate a metric which predicts final cell Voc times Isc and efficiency based on the PL image of the as-cut wafer from which the cell is made. The algorithm is based on image thresholding and basic statistical analysis of the PL intensity. It can be applied to a high resolution (more than 1 MP) photoluminescence image in less than 1 second, so it can be used on inline systems acquiring images at speeds of 3600 wafers per hour. A correlation coefficient approaching 0.9 with mean absolute error of 1% or less is routinely obtained between predicted efficiency and measured efficiency using this algorithm and Intevac’s NanoVista® PL tool. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|