DPSB model-based clustering algorithm for mineral mapping in hyperspectral imagery

The Dirichlet Process based on Stick-Breaking (DPSB) model-based clustering algorithm was instigated to Hyperion remote sensing imagery to map alteration minerals related to porphyry copper deposit. The DPSB clustering results were evaluated using Normalized Information Distance (NID), Normalized Mu...

Full description

Saved in:
Bibliographic Details
Published in:IEEE International Geoscience and Remote Sensing Symposium proceedings pp. 5470 - 5472
Main Authors: Yousefi, Mastoureh, Tabatabaei, Seyed Hassan, Pour, Amin Beiranvand, Pradhan, Biswajeet
Format: Conference Proceeding
Language:English
Published: IEEE 17.07.2022
Subjects:
Alteration minerals
Clustering algorithms
Copper
Data models
DPSB
Hyperion
Mineral exploration
Mineralization
Remote Sensing
Rocks
Spectroradiometers
Static VAr compensators
ISSN:2153-7003
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Dirichlet Process based on Stick-Breaking (DPSB) model-based clustering algorithm was instigated to Hyperion remote sensing imagery to map alteration minerals related to porphyry copper deposit. The DPSB clustering results were evaluated using Normalized Information Distance (NID), Normalized Mutual Information (NMI) and Modified adjusted Rand index (MARI) methods. Comparing the DPSB results with the geological map of the study area reveals the MARI of 0.2253, NMI of 0.1799 and MARI of 0.2253, respectively. To verify the extent of propylitic, argillic, advanced argillic, propylitic-argillic and sericite alteration zones associated with porphyry copper deposit, a total number of 24 rock samples were measured by a SVC XHR-1024i field portable spectroradiometer. Results derived from the DPSB model-based clustering algorithm show high accuracy in distinguishing propylitic, argillic, advanced argillic, propylitic-argillic and sericite alteration zones. It is advocated that the DPSB model-based clustering algorithm can be broadly implemented to hyperspectral and multispectral remote sensing data for detecting alteration zones associated with porphyry mineralization systems in other metallogenic provinces around the world.
ISSN:2153-7003
DOI:10.1109/IGARSS46834.2022.9884227