Anomaly detection with low magnetic flux: A fluxgate sensor network application Yavuz Ege a,⇑ , Mustafa Çoramık a , Murat Kabadayı a , Hakan Çıtak b , Osman Kalender c , Emrah Yürüklü c , Unal Kurt d , Sedat Nazlıbilek e a Balıkesir University, Necatibey Faculty of Education, Department of Physics, 10100 Balıkesir, Turkey b Balıkesir University, Balikesir Vocational High School, 10100 Balıkesir, Turkey c Bursa Orhangazi University, Department of Electrical-Electronics Engineering, 16350 Bursa, Turkey d Amasya University, Department of Electrical-Electronics Engineering, 05100 Amasya, Turkey e Atılım University, Faculty of Engineering, Department of Mechatronics Engineering, 06830 Ankara, Turkey article info Article history: Received 26 February 2015 Received in revised form 12 October 2015 Accepted 4 December 2015 Available online 12 December 2015 Keywords: Remote sensing Remote detection Magnetic anomaly Fluxgate sensor Magnetic materials abstract Recent studies on remote detection methods were mostly for improving variables like sensing distance, sensitivity and power consumption. Especially using anisotropic magneto-resistive sensors with low power consumption and high sensitivity for detecting subsurface magnetic materials became very popular in last decades. In our study, for detecting subsurface materials, we have used fluxgate sensor network for having even higher sensitivity and also minimizing the power consumption by detecting the changing rates of horizontal component of earth’s magnetic flux which is assumed to be very low. We have constituted a magnetic measurement system which comprises a detector system, which has a mechanism enables sensors to move in 3-D space, a data acquisition module for processing and sending all sensor information, and a computer for running the mag- netic flux data evaluation and recording software. Using this system, tests are carried out to detect anomalies on horizontal component of earth’s magnetic flux which is created by different subsurface materials with known magnetic, chemical and geometric proper- ties. The harmonics of horizontal component of earth’s magnetic flux in scanned area are analyzed by the help of DSP Lock-In amplifier and the amplitudes of high variation har- monics are shown as computer graphics. Using the graphic information, the upside surface geometry of subsurface material is defined. For identifying the magnetic anomalies, we have used the scale-invariant feature transform (SIFT)–binary robust invariant scalable keypoints (BRISKs) as keypoint and descriptor. We used an algorithm for matching the newly scanned image to the closest image in database which is constituted of mines and possible other metal objects like cans, etc. Results show that, if the proposed detection sys- tem is used instead of metal detectors which cannot distinguish mines from other metal materials and alert for every type of metal with different geometries, it can be said that miss alarm count, work force and time can be decreased dramatically. In this paper, mostly the setup of the system is described and in Appendix A some experimental outputs of the system for different geometries of metal samples are given. And also for comparing the results of the proposed system, additional experiments are carried out with a different type of sensor chip, namely KMZ51, and also given in Appendix A. Ó 2015 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.measurement.2015.12.004 0263-2241/Ó 2015 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. Tel.: +90 266 241 27 62x143; fax: +90 266 2495005. E-mail addresses: yavuzege@gmail.com (Y. Ege), mustafacoramik@hotmail.com (M. Çoramık), hcitak@balikesir.edu.tr (H. Çıtak), osman.kalender@bou. edu.tr (O. Kalender), emrah.yuruklu@bou.edu.tr (E. Yürüklü), unal.kurt@amasya.edu.tr (U. Kurt), snazlibilek@atilim.edu.tr (S. Nazlıbilek). Measurement 81 (2016) 43–56 Contents lists available at ScienceDirect Measurement journal homepage: www.elsevier.com/locate/measurement