Overview
In both exploration and AI applications, the quality and standardization of data is the first and most crucial step. Without clean, consistent inputs, even the most advanced tools can produce misleading results.
VRIFY’s Data Augmentation addresses this challenge through a structured workflow that transforms raw geological, geochemical, and geophysical data into continuous, standardized grids (rasters) that feed into DORA. By cleaning, enhancing, and augmenting input datasets, VRIFY’s Modules reveal critical patterns and generate new input layers for better decision-making.
VRIFY's Data Augmentation product is built around a suite of 10+ specialized Modules, each designed to handle different aspects of geological, geochemical, and geophysical data preparation. Every Module plays a distinct role in either enhancing existing information or producing new spatial data layers from known patterns.
Each Module is specifically designed for a particular input type and analytical objective, enabling geoscientific teams to transform raw exploration data into a structured portfolio of decision-ready raster layers. These outputs serve as foundational inputs for DORA’s Prediction Maps and further exploration analyses.
Data Augmentation involves:
Validating, cleaning, and standardizing inconsistent datasets to ensure alignment across geological and geophysical patterns.
Structuring information in a way that enables effective learning, comparison, and modelling across exploration datasets (change of support).
Enhancing existing features through smoothing, interpolation, or other refinement techniques.
Augmenting datasets by generating new layers from known spatial or geoscientific patterns.
Feature Processing Objectives
The main objectives of Data Augmentation are to standardize raw data through the Change of Support process and to derive new information from existing datasets.
By aligning different types of geological, geochemical, and geophysical data onto a common grid format (raster), Data Augmentation creates structured, high-quality inputs for DORA. It also generates new input layers that extend the value of the original data, revealing patterns and relationships critical for exploration analysis.
Change of Support
Change of Support is the process of adjusting datasets to a common spatial scale or resolution. In geostatistics, support refers to the physical area or volume over which each data value is averaged or measured. Because different datasets are collected at different scales, standardizing their support is essential for meaningful comparison and analysis.
These differences in scale exist in part because each geoscientific discipline (geochemistry, geology, and geophysics) has developed its own methods, tools, and spatial conventions for capturing information. Change of support addresses a common challenge in the industry, where data from different specialties is rarely aligned in a way that allows for straightforward comparison.
Change of support is achieved by rasterizing the data, which means converting it into a gridded format where each cell represents a defined spatial unit. Once aligned to a common grid (raster), datasets with originally differentiated supports can be combined, integrated, and used reliably in geospatial analysis.
For example, a soil geochemistry sample typically reflects a very local measurement in point support, an airborne geophysical survey provides averaged values over larger areas in block or cell support, and a satellite image delivers regularly spaced observations in a gridded support that may differ in resolution from other inputs. These differing supports must all be transformed into a common, gridded support (raster) to be meaningfully processed, compared, and interpreted within the DORA system.
Derive New Information
Once data is standardized through the Change of Support process, it can be used with VRIFY’s proprietary AI Modules for enhancement and augmentation. These Modules use machine learning to generate new insights from existing datasets.
Each Module is specifically optimized for different data types and exploration goals, aligning processing with the structure and characteristics of the input data. Together, they support two key outcomes: improving the clarity and interpretability of known datasets (data enhancement), and generating new modelled information (data augmentation) to extend and enrich exploration efforts.
Data Enhancement focuses on clarifying and refining existing information to improve its usefulness, consistency, and interpretability across datasets. Rather than creating entirely new variables or predictions, this process maximizes the value of raw data by improving its quality and coherence.
Data Prediction involves generating new data layers through predictive modelling based on known inputs. This approach extends the dataset by identifying patterns and extrapolating them across space, creating valuable new information in previously unmeasured areas.
The result is a more complete and usable dataset, with newly generated layers that extend well beyond the original inputs, giving exploration teams a richer, higher-quality foundation for analysis than they started with.
Learn More
Still have questions?
Reach out to your dedicated DORA contact or email support@VRIFY.com for more information.