Glossary

Image anomaly detection

What is Image anomaly detection?

Image anomaly detection looks for deviations from the normal state in photos or camera footage – such as scratches, cracks or discolouration on parts in manufacturing, or irregularities in medical images. Often a model first learns from many defect-free examples and then flags anything that differs. This is useful because real defects are rare and can hardly be described completely in advance.

How KI.ordo helps

KI.ordo does not analyse images; the platform works with text only, such as inspection reports or measurements in a table.

Topics

Computer vision
Quality inspection
Manufacturing

Try KI.ordo

Agents, flows and templates in one workspace – free 24-hour demo.

Start demo

← All terms

Related terms

Anomaly detection

Anomaly detection finds data points, events or patterns that differ markedly from the norm. These might be unusually large transactions, sudden jumps in sensor readings, duplicate records or suspicious network activity. It is used in finance, IT security, manufacturing and quality control. Methods range from simple statistical thresholds to learning models; hits are indications that a person should assess.

Data analysis
Quality control
+1

Template

Computer vision

Computer vision is a field of AI in which computers interpret images and video. Typical tasks include detecting and locating objects, reading text in images and measuring components. The technology is used, for example, in quality control in manufacturing, in driver assistance systems, in logistics and in medical imaging.

image analysis
deep learning
+1

Image recognition

Image recognition is a field of computer vision in which software identifies what an image shows – objects, people, text, logos or scenes. It is usually based on neural networks trained on large numbers of labelled images. Applications range from quality control in manufacturing and sorting product photos to supporting medical imaging. With images of people, data protection and potential bias require particular attention.

Computer vision
Image analysis
+1

Semantic segmentation

Semantic segmentation assigns a class to every single pixel in an image, such as road, vehicle, person or sky. The result is a kind of colour-coded map of the image content. Unlike instance segmentation, objects of the same type are not distinguished: three cars are treated together as “car”. It is used, for example, in autonomous driving, in medical imaging to delineate tissue and in the analysis of aerial and satellite images.

computer vision
image analysis
+1