This dataset contains the train/validation split for the NEU Steel Surface Detection Dataset. The original dataset can be found here => https://www.kaggle.com/datasets/rdsunday/neu-urface-defect-database The source also contains the description of the dataset and the type of images it contains.

Accompanying paper: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8007367/ Accompanying blog post: https://debuggercafe.com/steel-surface-defect-detection/

In short: The dataset contains close up of images of steel surface defects. There are 6 classes into which the defects can be classified. They are: [ 'crazing', 'inclusion', 'patches', 'pitted_surface', 'rolled-in_scale', 'scratches' ]

You can find more details in the paper.

Total samples: 1800 Training samples: 1700 Validation samples: 100

Annotations are in XML format.

Steel Surface Defect Detection

## Overview

Steel Surface Defect Detection is a dataset for object detection tasks - it contains Objects annotations for 1,799 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [Public Domain license](https://creativecommons.org/licenses/Public Domain).

Leathers (cows) are classified according to their quality at a certain stage in the tannery. The most important factor affecting their quality is surface defects. During the classification stage, the surface defect detection process, which is done by experts with human eyes, is automatically performed with artificial intelligence models trained with these images.

There are a total of 6 types of defects in this dataset: 1) Insect Bite 2) Scratch 3) Hole 4) Stitch Mark 5) Diseased 6) Rupture

Relevant publication: Ataç, H.O., Kayabaşı, A. & Aslan, M.F. The study on multi-defect detection for leather using object detection techniques. Collagen & Leather 6, 37 (2024). https://doi.org/10.1186/s42825-024-00186-2

In automated surface visual inspection, it is often necessary to capture the inspected part under many different illumination conditions to capture all the defects. To address this issue, at CSEM we have acquired a real-world multi-illumination defect segmentation dataset, called CSEM-MISD and we release it for research purposes to benefit the community.

The dataset consists of three different types of metallic parts -- washers, screws, and gears. Parts were captured in a half-spherical light-dome system that filtered out all the ambient light and successively illuminated it from 108 distinct illumination angles. Each 12 illumination angles share the same elevation level and the relative azimuthal difference between the adjacent illumination angles on the same level is 30 degrees. For more details, please read Sections 3 and 4 of our paper.

The washers dataset features 70 defective parts. The gears and screws datasets feature 35 defective, 35 intact and several hundred unannotated parts. Some defects, such as notches and holes, are visible in most images (illuminations) with intensity and texture variations among them, while others, such as scratches, are only visible in a few.

We split the datasets into train and test sets. The train sets contain 32 samples, and the test set 38 samples. Each sample comprises 108 images (each captured under a different illumination angle), an automatically extracted foreground segmentation mask, and a hand-labeled defect segmentation mask.

This dataset is challenging mainly because:

each raw sample consists of 108 gray-scale images of resolution 512×512 and therefore takes 27MB of space;

the metallic surfaces produce many specular reflections that sometimes saturate the camera sensors;

the annotations are not very precise because the exact extent of defect contours is always subjective;

the defects are very sparse also in the spatial dimensions: they cover only about 0.2% of the total image area in gears, 0.8% in screws, and 1.4% in washers; this creates an unbalanced dataset with a highly skewed class representation.

The dataset is organized as follows:

each sample resides in the Test, Train, or Unannotated directory;

each sample has its own directory which contains the individual images, the foreground, and defect segmentation masks;

each image is stored in 8-bit greyscale png format and has a resolution of 512 x 512 pixels;

Image file names are formatted using three string fields separated with the underscore character: prefix_sampleNr_illuminationNr.png, where the prefix is e.g. washer, the sampleNr might be a three-digit number 001, and the illuminationNr is formed of 3 digits, first corresponding to the elevation index (1 - highest angle, 9 - lowest angle), and the additional two corresponding to the azimuth index (01-12).

Each dataset contains light_vectors.csv, which contains the illumination angles (in lexicographic order of the illuminationNr), and light_intensities.csv that contains the numbers corresponding to the light intensity on the scale from 0 to 127. Please, be aware, that the azimuth angles were not calibrated and might be a few degrees misaligned.

We provide data loaders implemented in python at the project's repository.

If you find our dataset useful, please cite our paper:

Honzátko, D., Türetken, E., Bigdeli, S. A., Dunbar, L. A., & Fua, P. (2021). Defect segmentation for multi-illumination quality control systems. Machine vision and Applications.

Surface Defect Detection: dataset & papers

🐎📈 Constantly summarizing open source data sets in the field of surface defect research is very important. Important critical papers from year 2017 have been collected and compiled, which can be viewed in the :open_file_folder: [Papers] folder. 🐋

https://pic3.zhimg.com/80/v2-f2f84bb695855bae7a0093cd9022f3c2_720w.jpg">

At present, surface defect equipment based on machine vision has widely replaced artificial visual inspection in various industrial fields, including 3C, automobiles, home appliances, machinery manufacturing, semiconductors and electronics, chemical, pharmaceutical, aerospace, light industry and other industries. Traditional surface defect detection methods based on machine vision often use conventional image processing algorithms or artificially designed features plus classifiers. Generally speaking, imaging schemes are usually designed by using the different properties of the inspected surface or defects. A reasonable imaging scheme helps to obtain images with uniform illumination and clearly reflect the surface defects of the object. In recent years, many defect detection methods based on deep learning have also been widely used in various industrial scenarios.

Compared with the clear classification, detection and segmentation tasks in computer vision, the requirements for defect detection are very general. In fact, its requirements can be divided into three different levels: "what is the defect" (classification), "where is the defect" (positioning) And "How many defects are" (split).

1. Key Issues in Surface Defect Detection

1）Small Sample Problem

The current deep learning methods are widely used in various computer vision tasks, and surface defect detection is generally regarded as its specific application in the industrial field. In traditional understanding, the reason why deep learning methods cannot be directly applied to surface defect detection is because in a real industrial environment, there are too few industrial defect samples that can be provided.

Compared with the more than 14 million sample data in the ImageNet dataset, the most critical problem faced in surface defect detection is small sample problem. In many real industrial scenarios, there are even only a few or dozens of defective images. In fact, for the small sample problem which is one of the key problems in industrial surface defect detection, there are currently 4 different solutions:

- Data Amplification and Generation

The most commonly used defect image expansion method is to use multiple image processing operations such as mirroring, rotation, translation, distortion, filtering, and contrast adjustment on the original defect samples to obtain more samples. Another more common method is data synthesis, where individual defects are often fused and superimposed on normal (non-defective) samples to form defective samples.

- Network Pre-training and Transfer Learning

Generally speaking, using small samples to train deep learning networks can easily lead to overfitting, so methods based on pre-training networks or transfer learning are currently one of the most commonly used methods for samples.

- Reasonable Network Structure Design

The need for samples can also be greatly reduced by designing a reasonable network structure. Based on the compressed sampling theorem to compress and expand small sample data, we use CNN to directly classify the compressed sampling data features. Compared with the original image input, compressing the input can greatly reduce the network's demand for samples. In addition, the surface defect detection method based on the twin network can also be regarded as a special network design, which can greatly reduce the sample requirement.

- Unsupervised or Semi-supervised Method

In the unsupervised model, only normal samples are used for training, so there is no need for defective samples. The semi-supervised method can use unlabeled samples to solve the network training problem in the case of small samples.

2）Real-time Problem

<...

m-abbasi-m/Car-Gear-Surface-Defect-Detection dataset hosted on Hugging Face and contributed by the HF Datasets community

C10-DET represents a dataset designed for the detection of defects on large-scale metallic surfaces. It poses significant challenges in terms of the variety of defect categories, the number of images, and data volume. The presence of surface defects on metallic materials can have detrimental effects on the quality of industrial products, making efficient detection of metallic defects essential to meet the quality standards set by various industries. Consequently, there has been a growing interest in the field of metallic surface defect detection, leading to substantial advancements in quality control for industrial applications. However, the task of identifying metallic surface defects is inherently complex, primarily due to environmental factors like lighting, light reflections, and the unique properties of metal materials. These factors significantly increase the intricacy of surface defect detection. Note, that while the original paper stated that the dataset contained 3,570 grayscale images, the current version offers 2,300 images.

Weld Surface Defect Detection

## Overview

Weld Surface Defect Detection is a dataset for object detection tasks - it contains Classes annotations for 7,165 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Context

Surface defect inspection and detection within narrow, hollow cylindrical surfaces i.e., pipes and barrels can enormously impact the structural integrity of industrial products. Defect detection can play a major role in building inspection, finding minor defects to prolong the product's life.

Content

The dataset contains images of inside cylindrical surfaces. All images are in a single folder. The image folder contains 1,071 channel 3 images. The annotations for each image are recorded in seperate XML files. for five types of defects i.e., dirt, rusting, pitting, chipping, and thermal cracking.

The dataset contains more than 43 000 labeled wood surface defects and covers overall ten types of the most common defects, including live knots, dead knots, knots with crack, cracks, resins, marrows, quartzity, missing knots, blue stain, and overgrown. Each image in the dataset is provided with a semantic map and a bounding box label that allows performing semantic segmentation as well as localization tasks. All data were collected directly from a wood production line during the manufacturing process.

Surface Defect Detection

## Overview

Surface Defect Detection is a dataset for object detection tasks - it contains Scratches Patches Inclusions annotations for 60 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Dataset from http://faculty.neu.edu.cn/songkechen/zh_CN/zhym/263269/list/index.htm

NEU surface defect database

Kechen Song and Yunhui Yan

Description:

In the Northeastern University (NEU) surface defect database, six kinds of typical surface defects of the hot-rolled steel strip are collected, i.e., rolled-in scale (RS), patches (Pa), crazing (Cr), pitted surface (PS), inclusion (In) and scratches (Sc). The database includes 1,800 grayscale images: 300 samples each of six different kinds of typical surface defects.

Fig. 1 shows the sample images of six kinds of typical surface defects, the original resolution of each image is 200×200 pixels. From Fig. 1, we can clearly observe that the intra-class defects existing large differences in appearance, for instance, the scratches (the last column) may be horizontal scratch, vertical scratch, and slanting scratch, etc. Meanwhile the inter-class defects have similar aspects, e.g., rolled-in scale, crazing, and pitted surface. In addition, due to the influence of the illumination and material changes, the grayscale of the intra-class defect images is varied. In short, the NEU surface defect database includes two difficult challenges, i.e., the intra-class defects existing large differences in appearance while the inter-class defects have similar aspects, the defect images suffer from the influence of illumination and material changes. http://faculty.neu.edu.cn/_tsf/00/09/FzeYrqRveYRn.jpg" alt="http://faculty.neu.edu.cn/_tsf/00/09/FzeYrqRveYRn.jpg">

For defect detection task, we provided annotations which indicate the class and location of a defect in each image. We have carefully clicked annotations of each target in these images. Fig. 2 shows some examples of detection results on NEU-DET. For each defect, the yellow box is the bounding box indicating its location and the green label is the class score. http://faculty.neu.edu.cn/_tsf/00/09/ZjUbqmjAJvYz.png" alt="http://faculty.neu.edu.cn/_tsf/00/09/ZjUbqmjAJvYz.png">

To encourage research in computer vision for the railway, we present Rail-5k: a real-world image dataset for object detection of defects and accessories on the rail, along with methods for shooting, fine-frained category definition, and instance-level annotation. We collected 5,000 high-quality RGB images from high-speed railway and subway across China, where each image with resolution as high as 0.03mm per pixel. We annotate 1100 images with 13 types of defects and accessories that are the most important to rail maintenance such as rail surface, wheel-rail contact band, crack, spalling, corrugation, fastening, screw. The dataset is superior to existing datasets in image quantity, resolution, annotation quality, dense and small objects. It also contains real-world corrupted images with dark, overexposure, blur, other tools, different lens distance, category transition, different screws, which are infeasible for non-experts to annotate and recognize. As a pilot study of rail defect detection, we perform comprehensive experiments using SOTA models. Our experiments demonstrate several challenges Rail-5k posed to both computer vision and railway engineering. Future versions of this dataset will include even more images, segmentation annotations as well as more channels.

Metallic Surface Defect Detection

## Overview

Metallic Surface Defect Detection is a dataset for object detection tasks - it contains Defects annotations for 2,295 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Surface Defect Detection AI Market Outlook

As per our latest research, the global Surface Defect Detection AI market size reached USD 2.47 billion in 2024, reflecting robust adoption across multiple industries. The market is projected to grow at a CAGR of 22.8% from 2025 to 2033, reaching a forecasted value of USD 19.12 billion by 2033. This remarkable expansion is driven by the increasing need for automated quality assurance, the proliferation of smart manufacturing initiatives, and the rapid advancements in artificial intelligence and computer vision technologies.

The exponential growth of the Surface Defect Detection AI market can largely be attributed to the surging demand for automation in manufacturing processes. As industries strive to enhance productivity and minimize human error, AI-powered defect detection systems have emerged as a critical solution. These systems not only enable real-time inspection and identification of surface anomalies but also significantly reduce operational costs by minimizing the need for manual inspection. Furthermore, the integration of machine learning algorithms allows these systems to continuously improve their accuracy and adapt to new types of defects, which is particularly valuable in industries with high variability in product surfaces, such as automotive, electronics, and metals.

Another key factor fueling the market's growth is the increasing emphasis on product quality and regulatory compliance across sectors like pharmaceuticals, food and beverage, and packaging. With stricter quality standards and rising consumer expectations, organizations are increasingly investing in advanced AI-driven inspection technologies to ensure defect-free products. The rapid digital transformation in manufacturing, coupled with the adoption of Industry 4.0 principles, has further accelerated the deployment of surface defect detection AI solutions. These technologies not only improve yield and reduce waste but also provide actionable insights through data analytics, enabling manufacturers to optimize their production processes and maintain a competitive edge.

The Surface Defect Detection AI market is also witnessing significant advancements in hardware and software components, making these solutions more accessible and scalable for enterprises of all sizes. The emergence of cloud-based deployment models has democratized access to sophisticated AI algorithms, allowing even small and medium enterprises to leverage cutting-edge defect detection capabilities without heavy upfront investments in infrastructure. Additionally, the ongoing development of edge computing and high-resolution imaging technologies is enhancing the speed and accuracy of defect detection, further driving market adoption. These technological innovations, combined with the growing need for operational efficiency, are expected to sustain the market's momentum throughout the forecast period.

Regionally, Asia Pacific remains the dominant market, accounting for the largest share of global revenue in 2024. The region's leadership is underpinned by the rapid industrialization in countries such as China, Japan, and South Korea, where manufacturers are increasingly adopting AI-driven quality assurance solutions to maintain global competitiveness. North America and Europe also represent significant markets, driven by strong regulatory frameworks, high levels of automation, and a focus on innovative manufacturing practices. Meanwhile, emerging economies in Latin America and the Middle East & Africa are gradually embracing these technologies, supported by growing investments in industrial automation and digital transformation initiatives.

Component Analysis

The Surface Defect Detection AI market is segmented by component into software, hardware, and services, each playing a pivotal role in the deployment and effectiveness of defect detection solutions. The software segment encompasses AI-powered algorithms, computer vision modules, and analytics platforms that form the core of defect detection systems. These software solutions a

Metal Surface Defect Detection 1

## Overview

Metal Surface Defect Detection 1 is a dataset for object detection tasks - it contains Surface Defects Hj0i QnpH Surface Defects 93Ko annotations for 11,048 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

A surface defect detection dataset created by Northeastern University (NEU) that covers six different types of surface defects.

These are trained Faster RCNN weights for Steel Surface Defect Detection dataset. Dataset link => https://www.kaggle.com/datasets/sovitrath/neu-steel-surface-defect-detect-trainvalid-split Blog post link => Steel Surface Defect Detection using Object Detection

Contains the following trained weights: * Faster RCNN ResNet50 FPN V2 * Faster RCNN ViTDet * Faster RCNN MobileNetV3 Large FPN

PCB Defect is a dataset tailored for an object detection task, encompassing 1386 images annotated with 2953 labeled objects across six distinct classes: open_circuit, short, spurious_copper, and other nuanced defects such as missing_hole, mouse_bite, and spur. Specifically curated for Tiny Defect Detection (TDD), the dataset is instrumental for advancing quality control measures in the production of printed circuit boards (PCBs), which is a fundamental and crucial aspect of manufacturing processes in the electronics industry.

Solar cells are playing a significant role in aerospace equipment. In view of the surface defect characteristics in the manufacturing process of solar cells, the common surface defects are divided into three categories, which include difficult-detecting defects (mismatch), general defects (bubble, glass-crack and cell-crack) and easy-detecting defects (glass-upside-down). Corresponding to different types of defects, the deep learning model with different optimization methods and a classification detection method based on multi-models fusion are proposed in the paper. In the proposed model, in order to solve the mismatch problem between the default anchor boxes size of YOLOv5s model and the extreme scale of the battery mismatch defect label boxes, the K-means algorithm was adopted to re-cluster the dedicated anchor boxes for the mismatch defect label boxes. In order to improve the comprehensive detection accuracy of YOLOv5s model for the general defects, the YOLOv5s model was also improved by the methods of image preprocessing, anchor box improving and detection head replacing. In order to ensure the recognition accuracy and improve the detection speed for easy-detecting defects, the lightweight classification network MobileNetV2 was also used to classify the cells with glass-upside-down defects. The experimental results show that the proposed optimization model and classification detection method can significantly improve the defect detection precision. Respectively, the detection precision for mismatch, bubble, glass-crack and cell-crack defects are up to 95.64%, 91.8%, 93.1% and 98.0%. By using lightweight model to train the glass-upside-down defect dataset, the average classification accuracy reaches 100% and the detection speed reaches 13.29 frames per second. The comparison experiments show that the proposed model has a great improvement in detection accuracy compared with the original model, and the defect detection speed of lightweight classification network is improved more obviously, which confirms the effectiveness of the proposed optimization model and the multi-defect classification detection method for solar cells defect detection.