Evaluating the Performance of PRISMA Shortwave Infrared Imaging Sensor for Mapping Hydrothermally Altered and Weathered Minerals Using the Machine Learning Paradigm

Abstract

Satellite images provide consistent and frequent information that can be used to estimate mineral resources over a large spatial extent. Advances in spaceborne hyperspectral remote sensing (HRS) and machine learning can help to support various remote-sensing-based applications, including mineral exploration. Leveraging these advances, the present study evaluates recently launched PRISMA spaceborne satellite images to map hydrothermally altered and weathered minerals using various machine-learning-based classification algorithms. The study was performed for the town of Jahazpur in Rajasthan, India (75°06′23.17″E, 25°25′23.37″N). The distribution map for minerals such as kaolinite, talc, and montmorillonite was generated using the spectral angle mapper technique. The resultant mineral distribution map was verified through an intensive field validation survey on surface exposures of the minerals. Furthermore, the obtained pixels of the end-members were used to develop the machine-learning-based classification models. Measures such as accuracy, kappa coefficient, F1 score, precision, recall, and ROC curve were employed to evaluate the performance of developed models. The results show that the stochastic gradient descent and artificial-neural-network-based multilayer perceptron classifiers were more accurate than other algorithms. Results confirm that the PRISMA dataset has enormous potential for mineral mapping in mountainous regions utilizing a machine-learning-based classification framework.

PRISMA Hyperspectral Remote Sensing Data for Mapping Alteration Minerals in Sar-e-Châh-e-Shur Region, Birjand, Iran

“…Hyperspectral remote sensing imagery has an exceptional ability to capture both the visual image and spectral details of targeted objects, simultaneously. The images contain numerous closely arranged and narrow spectral bands in the visible near-infrared (VNIR) to short-wave infrared (SWIR) regions [1,2]. Alteration mineral mapping is a crucial stage for ore mineral exploration and cost-effective management of mineral exploration campaigns.…”

“…2024, 16, 1277 2 of 31 using multispectral imagery. Thus, hyperspectral imagery, encompassing contiguous and comprehensive spectral capabilities, provides a more precise characterization of alteration minerals compared to multispectral remote sensing [2].…”

“…It enables the capture of images with a superior signal-to-noise ratio (SNR) compared to the Hyperion sensor. Essentially, it serves as a satellite-based earth observation mission designed to provide spectroscopic imagery, fostering the development of innovative methods and applications for the management and analysis of ore mineral resources [2,[7][8][9][10]. The coexistence of minerals with similar spectral properties often leads to spectral mixing, resulting in overlapping absorption regions and highly correlated spectra [11,12].…”

“…These challenges, including diverse spectral features and limited ground samples, hinder the function of image processing algorithms and remote sensing sensors. Addressing these issues, the use of PRISMA hyperspectral imagery in mineral exploration across various geographic and geological settings highlights the adaptability and effectiveness of PRISMA imagery in overcoming challenges posed by spectral mixing, light scattering effects, and intricate spectral patterns [2,[7][8][9][10].…”

“…The Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), Independent Component Analysis (ICA), Adaptive Coherence Estimator (ACE), Random Forest (RF), XGboost (XGB), Support Vector Machine (SVM) and many other machine-learningbased classification algorithms were used for processing PRISMA data in mapping alteration minerals, identifying economic mineralization prospects, delineating dolomitization, obtaining high-quality reflectance estimations and achieving high-accuracy lithological mapping [2,[7][8][9][10][14][15][16][17][18][19]. Additionally, novel approaches and algorithms such as the spectral hourglass, iterative informed spectral unmixing technique, fuzzy logic approach, GIS-based algorithm, and informed linear mixing model were implemented to PRISMA data to determine the potential of the dataset for automated alteration mineral identification in diverse environments [20][21][22][23][24][25].…”