|
|
|
| A study of landmine target recognition based on Mahalanobis distance template feature |
| Cheng-Hao WANG1,2, Dan-Dan CHENG1 |
1. China Research Institute of Radiowave Propagation, Qingdao 266107, China
2. Science and Technology on Near Surface Detection Laboratory, Wuxi 214035, China |
|
|
|
|
Abstract Mine detection by ground penetrating radar is an important application direction, and its detection effect on non-metallic mines or mines with low metal content is remarkable. In this paper, aimed at tackling the problem that the target feature extraction is difficult when the ground penetrating radar detects the mine, the authors propose the SVM recognition algorithm based on the Mahalanobis distance template feature and give the recognition result. This method can effectively extract the target characteristics of mines, and is helpful to data interpretation of ground penetrating radar and recognition and location of mine targets.
|
|
Received: 11 October 2018
Published: 15 August 2019
|
|
|
|
|
|
|
radar B-scan image
|
|
Algorithm flowchart
|
|
Target and background image
|
|
Mahalanobis distance template eigenvector
|
|
Euclidean distance template eigenvector
|
|
Positive samples
|
|
Negative samples
|
|
Principle of SVM
|
|
Identification result
|
| [1] |
Konduri R K, Solomon G Z, Dejong K , et al. Genetic optimization of the HSTAMIDS landmine detection algorithm[J]. Proceedings of SPIE — The International Society for Optical Engineering, 2004,5415:883-891.
|
| [2] |
Ho K C, Gader P D, Wilson J N . Improving landmine detection using frequency domain features from ground penetrating radar[C]// Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International. IEEE, 2004:1617-1620 vol.3.
|
| [3] |
Daniels D J, Curtis P, Lockwood O . Classification of landmines using GPR[C]// Radar Conference, 2008. RADAR '08. IEEE. IEEE, 2008: 1-6.
|
| [4] |
Steinway W J, Reidy D M . Clutter removal processing for improved mine detection using a frequency-stepped GPR[C]// Detection and Remediation Technologies for Mines and Minelike Targets IX. International Society for Optics and Photonics, 2004: 896-904.
|
| [5] |
Bartosz E E, Dejong K, Duvoisin H A , et al. Nonlinear processing of radar data for landmine detection[J]. Proceedings of SPIE — The International Society for Optical Engineering, 2004.
|
| [6] |
Sezgin M . Simultaneous buried object detection and imaging technique utilizing fuzzy weighted background calculation and target energy moments on ground penetrating radar data. EURASIP Journal on Advances in Signal Processing, 2011(55):1-12.
|
| [7] |
Ho K C, Harris S, Zare A , et al. Anomaly detection of subsurface objects using handheld ground penetrating radar. Proc. SPIE Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XX, 94541B, 2015.
|
| [8] |
曾昭发, 刘四新, 冯晅 . 探地雷达原理与应用[M]. 北京: 电子工业出版社, 2010.
|
| [8] |
Zeng Z F, Liu S X, Feng X. Ground penetrating radar theory and applications[M]. Publishing House of Electronics Industry, 2010.
|
| [9] |
Chang C C, Lin C J . LIBSVM: a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011,2(3):27-65.
|
| [10] |
汪西莉, 焦李成 . 一种基于马氏距离的支持向量快速提取算法[J]. 西安电子科技大学学报:自然科学版, 2004,31(4).
|
| [10] |
Wang X L, Jiao L C . A fast algorithm for extracting the support vector on the Mahalanobis distance[J]. Journal of Xidian University, 2004,31(4).
|
|
|
|
