A dataset for surface defect detection on complex structured parts based on photometric stereo

Abstract Automated Optical Inspection (AOI) technology is crucial for industrial defect detection but struggles with shadows and surface reflectivity, resulting in false positives and missed detections, especially on non-planar parts. To address these issues, a novel defect detection technique based...

Full description

Saved in:
Bibliographic Details
Main Authors: Lin Wu, Yu Ran, Li Yan, Yixing Liu, You Song, Dongran Han
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-04454-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850087752160247808
author Lin Wu
Yu Ran
Li Yan
Yixing Liu
You Song
Dongran Han
author_facet Lin Wu
Yu Ran
Li Yan
Yixing Liu
You Song
Dongran Han
author_sort Lin Wu
collection DOAJ
description Abstract Automated Optical Inspection (AOI) technology is crucial for industrial defect detection but struggles with shadows and surface reflectivity, resulting in false positives and missed detections, especially on non-planar parts. To address these issues, a novel defect detection technique based on deep learning and photometric stereo vision was proposed, along with the creation of the Metal Surface Defect Dataset (MSDD). The proposed Stroboscopic Illuminant Image Acquisition (SIIA) method uses a specially arranged illuminant setup and a Taylor Series Channel Mixer (TSCM) to blend multi-angle illumination images into pseudo-color images. This approach enables end-to-end defect detection using universal object detectors. The method involves mapping color space transformations to spatial domain transformations and utilizing hue randomization for data augmentation. Four object detection methods (FCOS, YOLOv5, YOLOv8, and RT-DETR) were validated on the MSDD, achieving an mAP of 86.1%, surpassing traditional methods. The MSDD includes 138,585 single-channel images and 9,239 mixed images, covering eight defect types. This dataset is essential for automated visual inspection of metal surfaces and is freely accessible for research purposes.
format Article
id doaj-art-549a5dd9d2df4f678a7b72623e9cf1ec
institution DOAJ
issn 2052-4463
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Data
spelling doaj-art-549a5dd9d2df4f678a7b72623e9cf1ec2025-08-20T02:43:10ZengNature PortfolioScientific Data2052-44632025-02-0112111610.1038/s41597-025-04454-6A dataset for surface defect detection on complex structured parts based on photometric stereoLin Wu0Yu Ran1Li Yan2Yixing Liu3You Song4Dongran Han5School of Life Sciences, Beijing University of Chinese MedicineSchool of Life Sciences, Beijing University of Chinese MedicineSchool of Humanities, Beijing University of Chinese MedicineSchool of Management, Beijing University of Chinese MedicineSchool of Software, Beihang UniversitySchool of Life Sciences, Beijing University of Chinese MedicineAbstract Automated Optical Inspection (AOI) technology is crucial for industrial defect detection but struggles with shadows and surface reflectivity, resulting in false positives and missed detections, especially on non-planar parts. To address these issues, a novel defect detection technique based on deep learning and photometric stereo vision was proposed, along with the creation of the Metal Surface Defect Dataset (MSDD). The proposed Stroboscopic Illuminant Image Acquisition (SIIA) method uses a specially arranged illuminant setup and a Taylor Series Channel Mixer (TSCM) to blend multi-angle illumination images into pseudo-color images. This approach enables end-to-end defect detection using universal object detectors. The method involves mapping color space transformations to spatial domain transformations and utilizing hue randomization for data augmentation. Four object detection methods (FCOS, YOLOv5, YOLOv8, and RT-DETR) were validated on the MSDD, achieving an mAP of 86.1%, surpassing traditional methods. The MSDD includes 138,585 single-channel images and 9,239 mixed images, covering eight defect types. This dataset is essential for automated visual inspection of metal surfaces and is freely accessible for research purposes.https://doi.org/10.1038/s41597-025-04454-6
spellingShingle Lin Wu
Yu Ran
Li Yan
Yixing Liu
You Song
Dongran Han
A dataset for surface defect detection on complex structured parts based on photometric stereo
Scientific Data
title A dataset for surface defect detection on complex structured parts based on photometric stereo
title_full A dataset for surface defect detection on complex structured parts based on photometric stereo
title_fullStr A dataset for surface defect detection on complex structured parts based on photometric stereo
title_full_unstemmed A dataset for surface defect detection on complex structured parts based on photometric stereo
title_short A dataset for surface defect detection on complex structured parts based on photometric stereo
title_sort dataset for surface defect detection on complex structured parts based on photometric stereo
url https://doi.org/10.1038/s41597-025-04454-6
work_keys_str_mv AT linwu adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yuran adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT liyan adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yixingliu adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yousong adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT dongranhan adatasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT linwu datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yuran datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT liyan datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yixingliu datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT yousong datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo
AT dongranhan datasetforsurfacedefectdetectiononcomplexstructuredpartsbasedonphotometricstereo