Severstal Steel Defect Detection

## Overview

Severstal Steel Defect Detection is a dataset for object detection tasks - it contains Steel Defects annotations for 6,666 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).

The authors of the Severstal: Steel Defect Detection dataset acknowledge that steel holds a paramount position as one of the most vital building materials in modern construction. Its resilience against both natural elements and wear caused by human activities has rendered it indispensable worldwide. In the pursuit of enhancing the efficiency of steel production, the Severstal competition aims to play a pivotal role in the detection of defects within the steel production process.

Metal Surface Defect

## Overview

Metal Surface Defect is a dataset for object detection tasks - it contains Scratches annotations for 238 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).

patches (pa)

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).

Steel Surface Defect

## Overview

Steel Surface Defect is a dataset for object detection tasks - it contains Burn 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 [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Metal Defect Detection

## Overview

Metal Defect Detection is a dataset for object detection tasks - it contains Defects annotations for 888 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 Info

This dataset is made from the original data in this competition. Every element is cut into smaller chunks of 256x256 dimensions.

Files

• .7z archive with the train images
• "train_sliced_DEFECTS.csv" contains only the images with defects and their according masks
• "train_sliced_ALL.csv" contains all the images (containing defects + no defects)

I kept the structure of the original data, so you can acquire all needed information to work with the current dataset from here:

Soon I will upload the sliced test images.

Dataset

This dataset was created by Ammar Alhaj Ali

Contents

welds (Wl)

The global market for metal surface defect detectors is experiencing robust growth, driven by increasing automation in manufacturing, stringent quality control standards across diverse industries, and the rising adoption of advanced imaging technologies. The market is projected to reach a value of $2.5 billion in 2025, exhibiting a Compound Annual Growth Rate (CAGR) of 8% from 2025 to 2033. This growth is fueled by the expanding applications of metal surface defect detection across automotive, aerospace, electronics, and construction sectors. The demand for improved product quality and reduced manufacturing defects is a primary driver. Furthermore, the integration of artificial intelligence (AI) and machine learning (ML) algorithms is enhancing the accuracy and efficiency of defect detection systems, further bolstering market expansion. Technological advancements in sensor technologies (e.g., advanced vision systems, laser scanners) and the development of more sophisticated software for analyzing detected defects are also contributing to the overall market growth. However, high initial investment costs associated with implementing advanced defect detection systems and the need for skilled personnel to operate and maintain these systems present certain challenges. Nevertheless, the long-term benefits of improved product quality, reduced waste, and enhanced production efficiency are outweighing these restraints. The market is segmented by technology type (e.g., optical inspection, eddy current testing, ultrasonic testing), application (e.g., automotive parts, aerospace components, electronics manufacturing), and geography. Key players in the market include ZEISS Industrial Metrology, AMETEK Surface Vision, MABRI.VISION, and Lumina Instruments, among others, constantly innovating and expanding their product portfolios to cater to evolving industry needs. The competitive landscape is characterized by ongoing product development, strategic partnerships, and mergers & acquisitions, all aiming to capture a larger share of this expanding market.

Steel Defect Detection Final

## Overview

Steel Defect Detection Final is a dataset for object detection tasks - it contains STEEL DEFECT annotations for 1,769 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).

Yearly citation counts for the publication titled "Metal Surface Defect Detection Based on a Transformer with Multi-Scale Mask Feature Fusion".

Different network models for metal surface defect detection performance.

Steel Sheet Defect Detection

## Overview

Steel Sheet Defect Detection is a dataset for object detection tasks - it contains Defects annotations for 3,374 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).

Computational Complexity and Model Size of StarNet with Modules.

Performance comparison of the model with other semantic segmentation models.

Nowadays, industrial electronic products are integrated into all aspects of life, with PCB quality playing a decisive role in their performance. Ensuring PCB factory quality is thus crucial. Common PCB defects serve as key references for evaluating quality. To address low detection accuracy and the bulky size of existing models, we propose an improved PCB-YOLO model based on YOLOv8n.To reduce model size, we introduce a novel CRSCC module combining SCConv convolution and C2f, enhancing PCB defect detection accuracy and significantly reducing model parameters. For feature fusion, we propose the FFCA attention module, designed to handle PCB surface defect characteristics by fusing multi-scale local features. This improves spatial dependency capture, detail attention, feature resolution, and detection accuracy. Additionally, the WIPIoU loss function is developed to calculate IoU using auxiliary boundaries and address low-quality data, improving small-target recognition and accelerating convergence. Experimental results demonstrate significant improvements in PCB defect detection, with mAP50 increasing by 5.7%, and reductions of 13.3% and 14.8% in model parameters and computational complexity, respectively. Compared to mainstream models, PCB-YOLO achieves the best overall performance. The model’s effectiveness and generalization are further validated on the NEU-DET steel surface defect dataset, achieving excellent results. The PCB-YOLO model offers a practical, efficient solution for PCB and steel defect detection, with broad application prospects.

The global steel coil defect detection robot market is experiencing robust growth, driven by the increasing demand for high-quality steel products and the rising adoption of automation in the steel industry. The market's expansion is fueled by several key factors: the need for improved product quality to meet stringent industry standards, the escalating labor costs associated with manual inspection, and the desire for enhanced production efficiency and reduced waste. Technological advancements in robotics, computer vision, and artificial intelligence are significantly contributing to the market's growth, enabling more accurate and faster defect detection compared to traditional methods. Furthermore, the growing trend towards Industry 4.0 and smart manufacturing is accelerating the adoption of these advanced robotic systems. While the initial investment cost can be substantial, the long-term return on investment is significant, driven by reduced production losses and improved overall product quality. Major players such as Steel Warehouse, Bühler AG, ISRA Vision, AMETEK Surface Vision, and Shanghai Hengrui Measurement and Control Technology are actively shaping the market landscape through continuous innovation and strategic partnerships. The market segmentation reveals a diverse landscape, with different types of robots catering to specific needs, and various applications within steel production. Regional variations in adoption rates exist, largely influenced by factors such as technological maturity, industrialization levels, and government regulations. While market restraints include the high initial investment costs and the need for skilled personnel to operate and maintain the systems, the overall growth trajectory remains positive. The forecast period (2025-2033) anticipates a consistent increase in market size, reflecting the ongoing demand for automation and quality control solutions in the steel industry. Future trends point towards the development of more sophisticated AI-powered robots with enhanced capabilities for defect classification and predictive maintenance, thereby optimizing production efficiency and minimizing downtime.

Yolo V7 Metal Defects

## Overview

Yolo V7 Metal Defects is a dataset for instance segmentation tasks - it contains Metal Defect 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 [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).