No abstract

A novel method to classify micro-cracks in Photoluminescence (PL) images of polycrystalline solar cells is proposed. Micro-cracks in PL images are difficult distinguish as they're easily confused with noises that are present which may share the same size and shape features. Instead of relying on shape analysis to classify micro-cracks, the proposed method takes advantage of the patterns that are present at the end points of micro-cracks. Textural features are extracted via grey level co-occurrence matrix at the end points and then used as feature vectors in a SVM classifier. The proposed method is compared against existing shape analysis method and a preliminary experimental result has shown a significant improvement in sensitivity, specificity and accuracy.  Index Terms-solar cell, photoluminescence, micro-crack Teow Wee Teo graduated from University of Bradford, UK with a first-class honours B.Sc. degree from the Department of Computing and Mathematics in 2007 before joining TTVision Technologies Sdn. Bhd. as a R&D Engineer. His research focuses on machine vision applications in the photovoltaic industry with particular interests on machine learning, infrared imaging and luminescence technologies. As present, he is also a Ph.D. candidate under the supervision of Professor Dr Mohd Zaid Bin Abdullah at Universiti Sains Malaysia. Mohd Zaid graduated from Universiti Sains Malaysia (USM) with a B. App. Sc. degree in Electronic in 1986 before joining Hitachi Semiconductor (Malaysia) as a Test Engineer. In 1989 he commenced an M.Sc. in

No abstract

A new approach for high-speed micro-crack detection of solar wafers with variable thickness is proposed. Using a pair of laser displacement sensors, wafer thickness is measured and the lighting intensity is automatically adjusted to compensate for loss in NIR transmission due to varying thickness. In this way, the image contrast is maintained relatively uniform for the entire size of a wafer. An improved version of Niblack segmentation algorithm is developed for this application. Experimental results show the effectiveness of the system when tested with solar wafers with thickness ranging from 125 to 170 µm. Since the inspection is performed on the fly, therefore, a high throughput rate of more than 3600 wafers per hour can easily be obtained. Hence, the proposed system enables rapid in-line monitoring and real-time measurement.

A new approach incorporating adaptive lighting intensity for micro-crack inspection of solar wafers with variable thickness is proposed. Wafer thickness is measured with a pair of laser displacement sensors and the lighting intensity is adaptively adjusted to normalize near infrared (NIR) transmission based on measured thickness. This technique enables the image contrast be maintained at relatively uniform intensity in response to the variation of the solar wafer thickness. An improved version of Niblack segmentation algorithm is developed for this application. Experimental results demonstrate the competitiveness of the proposed system compared with established techniques, and achieves better performance both visually and quantitatively. Meanwhile, the runtime is consistently less than 1 s, corresponding to a throughput rate of approximately 3600 wafers/h. These results suggest that the methods and procedures are suitable for online processing of solar wafers.