Codesota · Vision · Industrial7 datasets · 8 models · 43K imagesUpdated 2026-05-04
§ 00 · Industrial anomaly detection

Detecting defects at scale.

From MVTec AD to weld X-rays, this is the register of models powering Industry 4.0 quality control. One-class learning, pixel-level localisation, zero-defect production lines.

The shape of the task is narrow: a fixed set of datasets, AUROC as the scoreboard, three families of approaches. Progress now shows up in tail metrics — PRO, pixel AUROC, and inference speed on factory hardware.

§ 01 · Metrics

What a number actually says.

Image AUROC

Area under the ROC curve

The gold standard for this task. Measures the trade-off between true-positive and false-positive rate across all thresholds. 99%+ is production-ready; below 97% and the operator is in the loop.

Pixel AUROC

Per-pixel localisation

Same curve, evaluated at pixel resolution. Separates models that can draw the defect from models that only flag the image. Drops sharply on small or thin defects.

PRO

Per-region overlap

Averages IoU across connected anomaly regions. More sensitive than pixel AUROC to small defects — a single missed crack hurts PRO much more than it hurts the pixel score.

§ 02 · Leaderboard

Architectures, ranked.

#ModelMVTec ADVisAArchitecture
01
SimpleNet
Research
99.6%95.5%Feature Adapter + Discriminator
02
EfficientAD
Research
99.1%94.8%Student-Teacher + Autoencoder
03
PatchCore
Amazon
99.1%92.1%Memory Bank + k-NN
04
FastFlow
Research
99.4%Normalizing Flow + CNN
05
Reverse Distillation
Research
98.5%Student-Teacher Reverse
06
CFLOW-AD
Research
98.3%Conditional Normalizing Flow
07
DRAEM
Research
98.0%Reconstructive + Discriminative
08
PaDiM
Research
97.9%Multivariate Gaussian + Pretrained CNN

Higher AUROC is better. MVTec AD image-level score drives the ranking; VisA used as a tiebreaker.

§ 03 · Three approaches

The families that dominate.

  1. 01

    Memory Bank

    Store embeddings of normal samples, flag nearest-neighbour distance. No training after feature extraction.

    PatchCore · PaDiM · SPADE
  2. 02

    Normalising Flows

    Learn the distribution of normal features; anomalies have low likelihood. Precise localisation.

    FastFlow · CFLOW-AD · CS-Flow
  3. 03

    Student–Teacher

    Student network mimics teacher on normal data; discrepancy is the anomaly score. 600+ FPS on factory hardware.

    EfficientAD · Reverse Distillation
§ 04 · Weld & steel

Pipeline integrity. Structural safety.

Surface inspection and X-ray radiography are where anomaly detection earns its keep: defects invisible to a human inspector, at line speed, with an auditable score per image.

Common defect types
  • · Porosity
  • · Slag inclusion
  • · Lack of fusion
  • · Cracks
  • · Rolled-in scale
  • · Pitted surface
Weld X-ray detection87.3% mAP

YOLOv8 fine-tuned · 30+ FPS

Severstal steel defect91.2% Dice

Kaggle competition winning solution

NEU steel surface78.4% mAP

6 defect types · hot-rolled steel

§ 05 · Datasets

The register.

MVTec AD

2019

5,354 high-resolution images across 15 object and texture categories. The gold standard for industrial anomaly detection with pixel-level annotations.

Images
5,354
Metric
auroc

MVTec 3D-AD

2021

4,147 high-resolution 3D point cloud scans and RGB images across 10 categories. First comprehensive 3D anomaly detection benchmark.

Images
4,147
Metric
auroc

VisA

2022

10,821 high-resolution images across 12 objects with complex structures. More challenging than MVTec with realistic industrial scenarios.

Images
10,821
Metric
auroc

KolektorSDD2

2021

3,335 images of electrical commutators with surface defects. Real industrial dataset with challenging small defects.

Images
3,335
Metric
auroc

Weld Defect X-Ray

2021

Radiographic images of welds with annotations for porosity, slag inclusion, lack of fusion, cracks. Critical for pipeline and structural inspection.

Images
4,500
Metric
map

Severstal Steel Defect

2019

12,568 steel sheet images with 4 defect classes from Kaggle competition. Real industrial data from major steel producer.

Images
12,568
Metric
dice

NEU-DET

2013

1,800 grayscale images of hot-rolled steel strip with 6 defect types: rolled-in scale, patches, crazing, pitted surface, inclusion, scratches.

Images
1,800
Metric
map

Run your own benchmarks.

We’re looking for GPU compute partners to run comprehensive anomaly-detection benchmarks on industrial datasets — including datasets we can’t publish publicly.

Partner with us Back to Computer Vision