A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images

Ahmet Topal; Burcu Tunga; Evrim Korkmaz Özay

doi:10.1109/IDAACS68557.2025.11321954

A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images

Ahmet Topal^*
, Burcu Tunga
, Evrim Korkmaz Özay

^*Corresponding author for this work

Department of Mathematics Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The process of anomaly detection focuses on detecting pixels whose spectral signatures differ from the background. This work proposes a High Dimensional Model Representation (HDMR) based Kernel RX anomaly detector to better address the relationships between pixels. To this end, the image generated by HDMR components is given as input to KRX to detect anomalies. In experimental work, our algorithm was compared with well-known traditional anomaly detection algorithms on two hyperspectral images. Results show that proposed algorithm was proven to have stronger performance than other approaches and detected anomalies with fewer false alarms.

Original language	English
Title of host publication	Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems
Subtitle of host publication	Technology and Applications, IDAACS 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	694-699
Number of pages	6
ISBN (Electronic)	9798331580452
DOIs	https://doi.org/10.1109/IDAACS68557.2025.11321954
Publication status	Published - 2025
Event	13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025 - Gliwice, Poland Duration: 4 Sept 2025 → 6 Sept 2025

Publication series

Name	Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS
ISSN (Print)	2770-4262
ISSN (Electronic)	2770-4254

Conference

Conference	13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025
Country/Territory	Poland
City	Gliwice
Period	4/09/25 → 6/09/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Anomaly detection
High Dimensional Model Representation
Hyperspectral images

Access to Document

10.1109/IDAACS68557.2025.11321954

Cite this

Topal, A., Tunga, B., & Özay, E. K. (2025). A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images. In Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025 (pp. 694-699). (Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IDAACS68557.2025.11321954

Topal, Ahmet ; Tunga, Burcu ; Özay, Evrim Korkmaz. / A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images. Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025. Institute of Electrical and Electronics Engineers Inc., 2025. pp. 694-699 (Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS).

@inproceedings{12505098d1644fed97e7689b7b499dce,

title = "A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images",

abstract = "The process of anomaly detection focuses on detecting pixels whose spectral signatures differ from the background. This work proposes a High Dimensional Model Representation (HDMR) based Kernel RX anomaly detector to better address the relationships between pixels. To this end, the image generated by HDMR components is given as input to KRX to detect anomalies. In experimental work, our algorithm was compared with well-known traditional anomaly detection algorithms on two hyperspectral images. Results show that proposed algorithm was proven to have stronger performance than other approaches and detected anomalies with fewer false alarms.",

keywords = "Anomaly detection, High Dimensional Model Representation, Hyperspectral images",

author = "Ahmet Topal and Burcu Tunga and {\"O}zay, \{Evrim Korkmaz\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025 ; Conference date: 04-09-2025 Through 06-09-2025",

year = "2025",

doi = "10.1109/IDAACS68557.2025.11321954",

language = "English",

series = "Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "694--699",

booktitle = "Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems",

address = "United States",

}

Topal, A , Tunga, B & Özay, EK 2025, A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images. in Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025. Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS, Institute of Electrical and Electronics Engineers Inc., pp. 694-699, 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025, Gliwice, Poland, 4/09/25. https://doi.org/10.1109/IDAACS68557.2025.11321954

A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images. / Topal, Ahmet ; Tunga, Burcu; Özay, Evrim Korkmaz.
Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025. Institute of Electrical and Electronics Engineers Inc., 2025. p. 694-699 (Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images

AU - Topal, Ahmet

AU - Tunga, Burcu

AU - Özay, Evrim Korkmaz

PY - 2025

Y1 - 2025

N2 - The process of anomaly detection focuses on detecting pixels whose spectral signatures differ from the background. This work proposes a High Dimensional Model Representation (HDMR) based Kernel RX anomaly detector to better address the relationships between pixels. To this end, the image generated by HDMR components is given as input to KRX to detect anomalies. In experimental work, our algorithm was compared with well-known traditional anomaly detection algorithms on two hyperspectral images. Results show that proposed algorithm was proven to have stronger performance than other approaches and detected anomalies with fewer false alarms.

AB - The process of anomaly detection focuses on detecting pixels whose spectral signatures differ from the background. This work proposes a High Dimensional Model Representation (HDMR) based Kernel RX anomaly detector to better address the relationships between pixels. To this end, the image generated by HDMR components is given as input to KRX to detect anomalies. In experimental work, our algorithm was compared with well-known traditional anomaly detection algorithms on two hyperspectral images. Results show that proposed algorithm was proven to have stronger performance than other approaches and detected anomalies with fewer false alarms.

KW - Anomaly detection

KW - High Dimensional Model Representation

KW - Hyperspectral images

UR - https://www.scopus.com/pages/publications/105032737170

U2 - 10.1109/IDAACS68557.2025.11321954

DO - 10.1109/IDAACS68557.2025.11321954

M3 - Conference contribution

AN - SCOPUS:105032737170

T3 - Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS

SP - 694

EP - 699

BT - Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025

Y2 - 4 September 2025 through 6 September 2025

ER -

Topal A , Tunga B, Özay EK. A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images. In Proceedings of the 13th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2025. Institute of Electrical and Electronics Engineers Inc. 2025. p. 694-699. (Proceedings of the IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS). doi: 10.1109/IDAACS68557.2025.11321954

A Novel HDMR-KRX Algorithm for Anomaly Detection on Hyperspectral Images

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this