Robust Dislocation Defects Region Segmentation for Polysilicon Wafer Image With Random Texture Background

Haiyong Chen,Jiali Liu,Shuang Wang,Kun Liu

doi:10.1109/access.2019.2942218

Haiyong Chen, Jiali Liu + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2942218

Copy DOI

Abstract

The detection of dislocation defects in polysilicon wafers helps to improve the power generation efficiency and service life of solar cells. However, dislocation defect detection is challenging due to similarity of morphology and intensity between the non-uniform random texture background and dislocation defect regions. In this paper, a novel and robust Multi-scale Feature Saliency Map (MFSM) is proposed to segment dislocation defects accurately. In order to highlight the dislocation area and weaken the background, we employ the Parameter-optimized Atmospheric Scattering Model (PASM) to enhance image contrast and preserve dislocation defect region information. Then, the multi-scale gradient feature is employed to obtain the multi-scale feature saliency map including all possible contours from the enhanced image. Furthermore, the watershed transform is employed to remove pseudo-defective regions arcs in MFSM. Finally, super hierarchical region tree is used to rank the likelihood of dislocation contours to obtain accurate dislocation area. The experimental results show that the proposed method can effectively segment dislocation defects and have good adaptability and robustness to complex background.

Highlights

With the depletion of fossil energy, solar energy is playing an increasingly important role
We propose a dislocation segmentation algorithm based on the Multi-scale Feature Saliency Map (MFSM) to solve the morphological similarity between non-uniform texture background and dislocation defects
WORK In this paper, a robust Multi-scale Feature Saliency Map combined with Parameter-optimized Atmospheric Scattering model (PASM-MFSM) is proposed to segment the dislocation regions of the polysilicon wafer images

Summary

INTRODUCTION

With the depletion of fossil energy, solar energy is playing an increasingly important role. By studying the microscopic structure of polysilicon, it is found that the dislocations are caused by local irregular arrangement of silicon atoms, which seriously affect the material properties [2], [4], [5] In this case, we need to select good quality polysilicon wafers for processing into solar cells, so it is necessary to evaluate the dislocation density of polysilicon wafers. A Parameter-optimized Atmospheric Scattering Model (PASM) is employed to enhance the dislocation defect image contrast and a novel and robust Multi-scale Feature Saliency Map (MFSM) is proposed to segment the enhanced dislocation defects accurately. The interaction between near arcs in multi-scale feature saliency map introduces pseudo-defective regions arcs To solve this problem, we employ watershed transform to connect the regions both containing the arc.

RELATED WORK

BRIGHTEST LIGHT AREA ESTIMATION

TRANSMISSION VALUE ESTIMATION

DISLOCATION REGION SEGMENTATION

EXPERIMENTAL RESULTS AND DISCUSSION

CONCLUSION AND FUTURE WORK