Dr. Pedram Ghamisi
Department of Exploration

Phone: 0351 260 - 4405

Machine Learning

Machine learning bridges the gap between geology, artificial intelligence, and remote sensing (Earth observation). For that, the group designs, develops, and implements different frameworks where a variety of datasets, e.g. multi- and hyper-spectral imagery or laboratory measurements are used to map mineralogical features and identify areas with high economic potential in more complex and remote areas. Different methods allow to develop efficient, robust, and reproducible algorithms to analyse datasets captured by different platforms in order to assist the geologists and improve exploration and characterization of mineral deposits.

Current Research

  • Fusion of laboratory analysis and hyperspectral data for the classification and quantification of mineral phases in drill core samples
  • Mineral mapping based on hyperspectral data
  • Satellite, airborne, and unmanned aerial systems (UAS) image analysis
  • Assessment of dynamic of changes caused by mining activities using time-series of data
  • Multi-sensor data fusion for accurate classification and change detection
  • Advanced machine learning and deep learning frameworks for raw material detection and classification
  • Unmixing, feature selection, extraction, and dimensionality reduction

Applied Methods

  • Mathematical optimization approaches
    • such as sparse, low-rank, or total variation
  • Machine learning methods
    • such as deep learning, ensemble learning, composite kernel classifiers, unsupervised learning, semi-supervised learning, and active learning
Multisource and Multitemporal Data Fusion ©Copyright: Dr. Ghamisi, Pedram
(a) The multiscale nature of diverse data sets captured by multisensor data (spaceborne, airborne, and UAV sensors) in Namibia [14]. (b) The tradeoff between spectral and spatial resolutions. (c) Elevation information obtained by lidar sensors from the University of Houston. (d) A time-series data analysis for assessing the dynamic of changes using red-green-blue (RGB) and urban images captured from 2001 to 2006 in Dubai, United Arab Emirates. (Ghamisi et al., 2019)

Selected Publications

  • P. Ghamisi, B. Rasti, N. Yokoya, Q. Wang, B. Hofle, L. Bruzzone, F. Bovolo, M. Chi, K. Anders, R. Gloaguen, P. M. Atkinson, J. A. Benediktsson (2019)
    Multisource and Multitemporal Data Fusion in Remote Sensing: A Comprehensive Review of the State of the Art. in IEEE Geoscience and Remote Sensing Magazine, vol. 7, no. 1, pp. 6-39
    DOI-Link: 10.1109/MGRS.2018.2890023
  • I. Cecilia Contreras Acosta, M. Khodadadzadeh, L. Tusa, P. Ghamisi and R. Gloaguen (2019)
    A Machine Learning Framework for Drill-Core Mineral Mapping Using Hyperspectral and High-Resolution Mineralogical Data Fusion. in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
    DOI-Link: 10.1109/JSTARS.2019.2924292
  • L. Tusa, L. Andreani, M. Khodadadzadeh, C. Contreras, P. Ivascanu, R. Gloaguen and J. Gutzmer (2019)
    Mineral Mapping and Vein Detection in Hyperspectral Drill-Core Scans: Application to Porphyry-Type Mineralization. Minerals, vol. 9, no. 2
    DOI-Link: 10.3390/min9020122