Computer Vision Toolbox Automated Visual Inspection Library

Identify anomalies or defects in images to assist and improve quality assurance processes.


Updated 13 Sep 2023

The Computer Vision Toolbox™ Automated Visual Inspection Library offers functions that enable you to train, calibrate, and evaluate anomaly detection networks.
The library enables:
Training and evaluating state of the art anomaly detectors including PatchCore, FCDD, and FastFlow. All detectors support standalone deployment with MATLAB Coder, GPU Coder, and MATLAB Compiler.
Calibrating trained networks by setting the anomaly thresholds for max number of false positive and negatives that are acceptable.
Evaluating the trained networks quantitively using validation metrics and qualitatively by visualizing anomaly heatmaps.
Evaluating an anomaly detector’s performance across a test set using an easy to use user interface (UI)
Labeling, training and calibration ground truth data using the Image Labeler app by marking defective areas with a segmentation mask.
For more information, please visit the automated visual inspectiondocumentation page which shows you how to get started with anomaly detection using deep learning. The documentation also features dedicated examples such as the Pill QC example which uses the support package in an end-to-end detection workflow.
MATLAB Release Compatibility
Created with R2022b
Compatible with R2022b to R2023b
Platform Compatibility
Windows macOS (Apple silicon) macOS (Intel) Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!