Anomaly detection in radio galaxy data with trainable COSFIRE filters

TL;DR

This paper presents a semi-supervised anomaly detection method for radio galaxy data using trainable COSFIRE filters combined with an unsupervised LOF algorithm, outperforming deep learning autoencoders in accuracy.

Contribution

It introduces a novel, efficient anomaly detection framework leveraging COSFIRE filters and unsupervised learning, reducing reliance on labeled anomalous data.

Findings

Achieved a G-Mean score of 79%, surpassing deep autoencoders' 77%.

Demonstrated effectiveness on a radio galaxy benchmark dataset.

Provides a fast, semi-supervised alternative for anomaly detection in radio astronomy.

Abstract

Detecting anomalies in radio astronomy is challenging due to the vast amounts of data and the rarity of labeled anomalous examples. Addressing this challenge requires efficient methods capable of identifying unusual radio galaxy morphologies without relying on extensive supervision. This work introduces an innovative approach to anomaly detection based on morphological characteristics of the radio sources using trainable COSFIRE (Combination of Shifted Filter Responses) filters as an efficient alternative to complex deep learning methods. The framework integrates COSFIRE descriptors with an unsupervised Local Outlier Factor (LOF) algorithm to identify unusual radio galaxy morphologies. Evaluations on a radio galaxy benchmark data set demonstrate strong performance, with the COSFIRE-based approach achieving a geometric mean (G-Mean) score of 79%, surpassing the 77% achieved by a computationally intensive deep learning autoencoder. By characterizing normal patterns and detecting deviations, this semi-supervised methodology overcomes the need for anomalous examples in the training set, a major limitation of traditional supervised methods. This approach shows promise for next-generation radio telescopes, where fast processing and the ability to discover unknown phenomena are crucial.