TerraEye

In areas covered by vegetation or transformed by human activity, direct exposure of soil and rocks is often obscured, limiting the potential for detailed mineral mapping. To overcome this challenge, we have developed advanced methodologies for generating Bare Earth Composites (BECs) using satellite data from Sentinel-2, ASTER, or a combination of both sensors (resulting in a 21-band composite). These methodologies allow us to find uncovered pixels and reveal the underlying mineral features, thus increasing the area available for mapping. Our approach leverages the spectral and spatial capabilities of Sentinel-2 and ASTER data to extract surface information, which is otherwise masked by vegetation. This results in high-quality composites that provide insights into the mineral structure with a spatial resolution of 10–30 meters. By incorporating this methodology, we estimate that the total area available for accurate mapping could increase by as much as 20%. This substantial improvemen...

A reliable, accurate tool is the key to good results. We use several tools based on various approaches and techniques in satellite data analysis. One of the most fundamental techniques is geological spectral indices. What are geological spectral indices and their use cases? Spectral indices allow the user to extract and highlight the targeted and most desirable information about the area of interest. It can be either information on mineralogy, alteration mineralogy, or lithology. For example, when you need to check whether the area has any clay minerals present, a spectral index can determine the answer. Another use case example is to analyze what the surface of the outcrop of a specific rock type is in a geologically complex location. In this case, spectral indices built for highlighting different lithotypes will be successful. How is a spectral index built? Every mineral has its own, individual spectral signature filled with spectral features, a reflectance spectrum, which refl...

 In standard satellite images, the Earth is shown in natural colors, similar to how we see it with the naked eye. To broaden that perspective, in TerraEye we have implemented False Color Composites (FCC). FCC images utilize different spectral bands that are invisible to human vision, highlighting geological features based on how materials reflect light at various wavelengths. The technique works by assigning different spectral bands to red, green, and blue (RGB) channels giving the new colored composites of the analysed region. FCC images highlight surface features related to mineralization that may otherwise go undetected in standard imagery. This process allows for the identification of different rock types and mineralization zones based on their spectral characteristics. When enhanced with AI-powered algorithms, spectral indices, and lineament maps, FCC analysis provides highly precise insights that streamline the mineral discovery process. Interpreting FCC for Mineral Explora...

TerraEye introduces a lineament detection algorithm based on Digital Elevation Model (DEM) data. This approach allows for the identification of surface expressions of linear structures, which can be part of local geological formations significant for mineralization. As such, lineament data serves as an important input in the analysis of areas for potential mineral resources. The detection process is carried out using the Canny edge detection algorithm (Mohammadpour et al., 2020), which has proven to be effective for this purpose. In this version, the results are generated from a DEM derived from the global Copernicus GLO-30 dataset. The Canny algorithm operates in several stages: first, it filters noise to eliminate irrelevant details, then identifies and selects pixels where DEM values change sharply. From these selected pixels, lines are created, and in the final step, thresholding allows for the extraction of strong and clear edges or more subtle ones depending on the settings. The ...

We have developed AI Clusters a tool designed to enhance lithological mapping by applying machine learning techniques, specifically K-means clustering , to multi- and hyperspectral data . This process enables the classification of geological structures based on their spectral characteristics . Initially, a Principal Component Analysis (PCA) is applied to reduce the dimensionality of the data and focus on the most significant features. By combining PCA and K-means clustering , AI Clusters simplifies the analysis and allows for clearer identification of geological and lithological patterns. Once the clustering is complete, the results can be compared with RGB images, RGB Composites and mineral mapping outputs to provide a more comprehensive understanding of geological formations. This method can be used within a single geological structure to differentiate mineral compositions, offering valuable insights for mineral exploration and geological analysis. How AI Clusters Tool Enhance...

  In the complex field of mineral exploration, the ability to analyze vast amounts of satellite data accurately and efficiently is a key competitive advantage. TerraEye, a cloud-based application, provides advanced tools to process and interpret satellite data, particularly from Sentinel-2 and EMIT sensors. This blog post offers a detailed guide on how to navigate TerraEye, focusing on data processing and how to interpret the results of automated analyses. Additionally, we'll cover how to order additional datasets and request further processing of your Area of Interest (AOI) directly within the app. How to Get Started with TerraEye Getting started with TerraEye is straightforward: Sign Up: Create an account on the TerraEye website by entering your details and agreeing to the terms of service. Log In: Use your credentials to access the TerraEye dashboard. Explore Demo Sites: Familiarize yourself with TerraEye’s features by exploring the demo sites available on your dashboard. Cre...