Reducing Anomaly Detection in Images to Detection in Noise
This approach provides an unsupervised method for anomaly detection in images, addressing a domain-specific problem with potential applications in fields like surveillance or medical imaging.
The paper tackles the challenge of detecting anomalies in arbitrary background images by reducing it to detecting anomalies in residual images dominated by noise, thereby replacing complex background modeling with simpler noise modeling and enabling rigorous threshold calculation using a contrario detection theory.
Anomaly detectors address the difficult problem of detecting automatically exceptions in an arbitrary background image. Detection methods have been proposed by the thousands because each problem requires a different background model. By analyzing the existing approaches, we show that the problem can be reduced to detecting anomalies in residual images (extracted from the target image) in which noise and anomalies prevail. Hence, the general and impossible background modeling problem is replaced by simpler noise modeling, and allows the calculation of rigorous thresholds based on the a contrario detection theory. Our approach is therefore unsupervised and works on arbitrary images.