Automatic detection of multiple UXO-like targets using magnetic anomaly inversion and self-adaptive fuzzy c-means clustering
| Autor: | Guo-Quan Ren, Hongbo Fan, Yingtang Zhang, Gang Yin, Zhining Li |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Magnetometer
business.industry Computer science Centroid Geology Self adaptive Pattern recognition Inversion (meteorology) 02 engineering and technology Geophysics 010502 geochemistry & geophysics 01 natural sciences Fuzzy logic law.invention law Magnet 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Magnetic anomaly business Cluster analysis 0105 earth and related environmental sciences |
| Zdroj: | Exploration Geophysics. 48:67-75 |
| ISSN: | 1834-7533 0812-3985 |
| DOI: | 10.1071/eg14126 |
| Popis: | We have developed a method for automatically detecting UXO-like targets based on magnetic anomaly inversion and self-adaptive fuzzy c-means clustering. Magnetic anomaly inversion methods are used to estimate the initial locations of multiple UXO-like sources. Although these initial locations have some errors with respect to the real positions, they form dense clouds around the actual positions of the magnetic sources. Then we use the self-adaptive fuzzy c-means clustering algorithm to cluster these initial locations. The estimated number of cluster centroids represents the number of targets and the cluster centroids are regarded as the locations of magnetic targets. Effectiveness of the method has been demonstrated using synthetic datasets. Computational results show that the proposed method can be applied to the case of several UXO-like targets that are randomly scattered within in a confined, shallow subsurface, volume. A field test was carried out to test the validity of the proposed method and the experimental results show that the prearranged magnets can be detected unambiguously and located precisely. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|