Defect Detection And Classification Of Silicon Solar Wafer Featuring Nir Imaging And Improved Niblack Segmentation

Mahdavipour, Zeinab (2016) Defect Detection And Classification Of Silicon Solar Wafer Featuring Nir Imaging And Improved Niblack Segmentation. PhD thesis, Universiti Sains Malaysia.

[img]
Preview
PDF - Submitted Version
Download (726kB) | Preview

Abstract

Menghasilkan tenaga yang boleh diperbaharui berkuantiti tinggi memerlukan kecekapan yang tinggi dalam fabrikasi produk wafer silikon, yang juga merupakan komponen asas panel solar. Oleh yang demikian, pemeriksaan kualiti yang tinggi untuk wafer solar semasa proses pengeluaran sangat penting. Dalam tesis ini, sistem pengesanan kecacatan yang cekap dan automatik menggunakan strategi pengelasan dan kelompok termaju telah dicadangkan. Dalam kajian ini, satu skema mesin penglihatan untuk mengesan keretakan mikro dan kecacatan-kecacatan yang lain dalam pembuatan polihabluran dan mono kristal wafer solar dicadangkan dan dibangunkan. Pemeriksaan retak mikro sangat mencabar kerana kecacatan ini sangat kecil dan tidak boleh dilihat dengan mata kasar. Kewujudan struktur heterogenus yang lain dalam wafer solar seperti bahan-bahan kasar dan kawasan gelap menjadikan pemeriksaan lebih mencabar. Dalam tesis ini, sebuah inspektor retak mikro yang mengandungi pencahayaan inframerah yang dekat dan algoritma segmentasi Niblack yang diperbaharui telah dicadangkan. Keputusan emperikal dan visual menunjukkan ketepatan dan prestasi yang lebih baik dari segi angka merit Pratt dan kaedah penilaian yang lain berbanding dengan formula pengambangan Niblack yang sedia ada. Keputusan angka merit (FOM), ketepatan (ACC), pekali kesamaan dadu (DSC) dan sensitiviti yang masing-masingnya sentiasa lebih tinggi daripada 0.871, 99.35 %, 99.68 %, dan 99.75 % bagi imej-imej dalam kajian ini. Sementara itu, satu set deskriptor bersepadanan dengan penerangan ciri-ciri bentuk Fourier eliptik, diekstrak bagi setiap kecacatan yang telah dikesan, dan dinilai bagi setiap kluster bagi tujuan pengelompokan dan pengelasan. Pengelasan menggabungkan analisis ciri keamatan kecacatan, penggunaan tanpa pengawasan kelompok purata-k dan pelbagai kelas algoritma SVM. Kaedah-kaedah ini telah digunakan untuk pengesanan, pengelompokan dan klasifikasi imej wafer solar polihabluran, bersepadanan dengan kecacatan seperti keretakan mikro, kekotoran, dan cap jari. Keputusan kajian menunjukkan bahawa kaedah purata-k dan penklasifikasi SVM mampu mengelompok dengan tepat kecacatan-kecacatan tersebut dengan ketepatan, indeks Rand, dan Bayang indeks dengan nilai purata masing-masing sebanyak 99.8 %, 99.788 %, dan 98.43 %. ________________________________________________________________________________________________________________________ Producing a high yield of renewable energy requires a high efficiency in product fabrication of silicon wafers, which is the basic building component of solar panels. For this reason, the high quality inspection of solar wafers during the procedures of production is very important. In this thesis, an automatic and efficient defect detection system, utilising advanced classification and clustering strategies are proposed. In this study a machine vision scheme for detecting micro-cracks and other defects in polycrystalline and monocrystalline solar wafer manufacturing is proposed and developed. Micro-crack inspection is very challenging, because this type of defect is very small and completely invisible to the naked eye. The presence of other heterogeneous structures in solar wafers like grainy materials and dark regions further complicates the problem. In this study an efficient micro-crack inspector comprising near infrared illumination and an improved Niblack segmentation algorithm is proposed. Empirical and visual results demonstrate that the proposed solutions are competitive when compared to existing Niblack thresholding formulas and other standard methods, and achieve better precision and performance in terms of Pratt’s figure of merit and other evaluation methods. Result in a figure of merit (FOM), accuracy (ACC), dice similarity coefficient (DSC), and sensitivity were consistently higher than 0.871, 99.35 %, 99.68 %, and 99.75 %, respectively, for all images tested in this study. Meanwhile, a set of descriptors corresponding to Elliptic Fourier Features shape description is extracted for each defect and is evaluated for each cluster to use for clustering and classification part. The classification combines the analysis of defect intensity features, the application of unsupervised k-mean clustering and multi-class SVM algorithms. The methods have been applied for detecting, clustering and classification polycrystalline solar wafer images, corresponding to defects such as micro cracks, stain, and fingerprints. Results indicate that the k-mean and SVM classifier can accurately cluster the defects with accuracy, Rand index, and Silhouette index averaging at 99.8 %, 99.788 %, and 98.43 %, respectively.

Item Type: Thesis (PhD)
Additional Information: Full text is available at http://irplus.eng.usm.my:8080/ir_plus/institutionalPublicationPublicView.action?institutionalItemId=2799
Subjects: T Technology
T Technology > TK Electrical Engineering. Electronics. Nuclear Engineering > TK7800-8360 Electronics
Divisions: UNSPECIFIED
Depositing User: Mr Mohd Jasnizam Mohd Salleh
Date Deposited: 12 Jul 2018 08:21
Last Modified: 15 Aug 2018 03:35
URI: http://eprints.usm.my/id/eprint/41026

Actions (login required)

View Item View Item
Share