塑料齿轮表面缺陷图像数据集

本数据集为塑料齿轮表面缺陷数据集，包含上端面（顶针面）、下端面（水口面）和侧面（齿面）三种视角下的塑料齿轮全方位缺陷图像，共计有1165张图片，16种不同类型的表面缺陷。
数据集的原始版权归**北京工业大学精密测控技术与仪器实验室（石照耀教授团队）**所有，数据集的应用和研究成果已在以下公开发表的论文中进行阐述：
[1] Shi Z Y, Fang Y M, Song H X. Intelligent Inspection Method and System of Plastic Gear Surface Defects Based on Adaptive Sample Weighting Deep Learning Model[J]. Sensors, 2024, 24(14): 4660.
特别感谢东莞星火齿轮有限公司和深圳兆威机电股份有限公司为本数据集的搜集工作提供的平台支持。
欢迎学术界和工业界的研究人员就数据集的使用与作者进行讨论，作者联系方式为：[e-mail]，方一鸣博士。
数据集使用条款：
此数据集的使用仅限于非商业学术研究，若需用于商业目的或其他目的，请联系数据集所有者获取许可。
数据集可以参考以下推荐格式进行引用：Zhaoyao Shi and Yiming Fang, *Plastic Gear Surface Defect Dataset*, version 1.0, Beijing University of Technology, 2025. Available: [DOI 或网址].
数据集缺陷标签说明见附件“说明.pdf”
版本号：v1.0
发布日期：[2025.7.16]

This dataset is a plastic gear surface defect dataset, which contains comprehensive defect images of plastic gears from three perspectives: upper end face (ejector pin side), lower end face (water gap side), and side surface (tooth surface). A total of 1165 images are included, covering 16 distinct types of surface defects.

The original copyright of this dataset is held by the Precision Measurement and Control Technology and Instrument Laboratory of Beijing University of Technology (Team of Professor Zhaoyao Shi). The applications and research outcomes of this dataset have been detailed in the following published paper:
[1] Shi Z Y, Fang Y M, Song H X. Intelligent Inspection Method and System of Plastic Gear Surface Defects Based on Adaptive Sample Weighting Deep Learning Model[J]. Sensors, 2024, 24(14): 4660.

We hereby express our sincere thanks to Dongguan Spark Gear Co., Ltd. and Shenzhen Zhaowei Electromechanical Co., Ltd. for providing platform support for the collection of this dataset.

Researchers from both academia and industry are welcome to contact the authors regarding the utilization of this dataset. Contact information: Dr. Yiming Fang, [e-mail].

Dataset Usage Terms:
Usage of this dataset is restricted to non-commercial academic research. For commercial or other intended uses, please contact the dataset owner to obtain prior permission.

The dataset may be cited in the following recommended format: Zhaoyao Shi and Yiming Fang, *Plastic Gear Surface Defect Dataset*, version 1.0, Beijing University of Technology, 2025. Available: [DOI or URL].

Please refer to the attachment "Description.pdf" for the defect label instructions of this dataset.

Version: v1.0
Release Date: [July 16, 2025]