透過您的圖書館登入
IP:3.239.96.229
  • 學位論文

均值移動分割於多晶矽太陽能面板晶粒分析

Grain Analysis of Multi-crystalline Solar wafer Using Modified Mean-shift Segmentation

指導教授 : 范書愷

摘要


在這個科技高度發展的時代,人們對於能源的依賴程度越來越高,能源分配的問題影響的不只在經濟發展上,也嚴重地影響環境與生態。因此,尋找替代能源成為重要課題,太陽能成為很重要的發電來源;由於多晶矽太陽能電池製造成本較為經濟而成為市場上的主流。而一般機器視覺應用於多晶矽太陽能面板的研究是針對太陽能晶片的表面瑕疵進行檢測。由於在多晶矽面板中晶粒尺寸愈大、晶界越少,其品質與光電轉換效率愈佳,因此,本論文目的是提出一個簡單的非破壞性的測量,利用機器視覺方法對多晶矽太陽能面板進行影像處理,分割出不同的晶粒並計算出晶粒平均面積跟數量,提出一個能夠當作多晶矽太陽能面板的品質的衡量指標; Mean-shift 方法具有分群之能力,常應用在影像平滑、影像分割、物件追蹤上。進行影像平滑及分割收斂快速且具有消除雜訊並同時保留邊緣的優點。本研究針對多晶矽太陽能面板影像進行區域分割,先進行強化影像以避免低對比區塊無法分割出不同區塊的問題,再以Mean-shift平滑以及Mean-shift分割進行區塊分割,參數設定上使用一個融合分割區域內相似度和鄰近區塊間相異度的指標來決定。區域分割後將得出之結果進行區域面積以及數量的計算,得出平均晶粒面積作為多晶矽太陽能面板量測之指標。實驗結果顯示透過Mean-shift平滑以及分割可得到晶格分割的目的,並且能以分割後計算得到的平均面積當作多晶矽面板的品質量測指標。

並列摘要


In the era of high-technology, the degree of dependence on energy source became higher, and energy allocation problem influences not only in economic development, but also seriously affects the environment and ecology. Solar power is an attractive alternative source of energy. Multi-crystalline solar cell was popular in the market share because of the cheaper manufacturing cost. Multi-crystalline solar wafers have the characteristic that larger average grain size surface, less grain boundaries brings higher quality and conversion efficiency. In the past researches, machine vision was applied to defect detection in solar wafer surface. In this thesis, we would like to propose a new method to assess the wafer quality by using the image processing techniques that can segment different grains in the solar wafer to calculate the number of grain and the average grain size as the evaluation indicators. Mean-shift technique has been an attractive alternative for noise removal, region segmentation and object tracking in image processing. To accurately segment the contrasting regions, we perform the image enhancement first. Since mean-shift smoothing and segmentation is able to converge quickly and preserve edges and boundaries, so we use mean-shift to segment the solar cell image for obtaining the different grains. An experimental procedure is conducted to find the best-practice parameter for mean-shift. Subsequently, the obtained segmentation results are employed to calculate the grains average size as the evaluation metric for solar wafer. The computational results show that the average grain size obtained by means of mean-shift segmentation can be expressed as the quality indicator for solar wafer. The grain number computed by using mean-shift is also approximately close to the ground-true result.

參考文獻


[4] M. Z. Jacobson," Review of solutions to global warming, air pollution, and energy security," Energy and Environment Science, vol. 2, 2008, pp. 148-173
[7] M. Pilla, F. Galmiche and X. Maldague, "Thermographic inspection of cracked solar cells," Proceedings of SPIE, Seattle, WA, USA, 2002, pp. 699-703.
[8] Z. Fu, Y. Zhao, Y. Liu, Q. Cao, M. Chen, J. Zhang, J. Lee, (2004) "Solar cell crack inspection by image processing," in: Int’l. Conf. on Business of Electronic Product Reliability and Liability, Shanghai, China, 2004, pp. 77-80.
[9] D. M. Tsai, C. C. Chang and S. M. Chao, "Micro-crack inspection in heterogeneously textured solar wafers using anisotropic diffusion," Image and Vision Computing, vol. 28, 2010, pp. 491-501.
[10] P. Salembier and L. Garrido, "Binary partition tree as an efficient representation for image processing, segmentation, and information retrieval," IEEE transaction on image processing, vol.9, no. 4, 2000, pp.561-576.

延伸閱讀