SSGD: A smartphone screen glass dataset for defect detection

TL;DR

This paper introduces a new dataset of 2504 high-resolution images of touch screen glass with seven defect types, enabling development and benchmarking of defect detection methods for quality control.

Contribution

The paper provides the first publicly available dataset for touch screen glass defect detection and benchmarks CNN and Transformer models on this dataset.

Findings

CNN and Transformer models face challenges detecting defects in high-resolution images.

The dataset facilitates future research in automatic defect detection for touch screens.

Benchmark results highlight the difficulty of defect detection in complex scenarios.

Abstract

Interactive devices with touch screen have become commonly used in various aspects of daily life, which raises the demand for high production quality of touch screen glass. While it is desirable to develop effective defect detection technologies to optimize the automatic touch screen production lines, the development of these technologies suffers from the lack of publicly available datasets. To address this issue, we in this paper propose a dedicated touch screen glass defect dataset which includes seven types of defects and consists of 2504 images captured in various scenarios.All data are captured with professional acquisition equipment on the fixed workstation. Additionally, we benchmark the CNN- and Transformer-based object detection frameworks on the proposed dataset to demonstrate the challenges of defect detection on high-resolution images. Dataset and related code will be available at https://github.com/Yangr116/SSGDataset.