We develop and use unique algorithms and machine learning solutions to better process data

DATA PROCESSING

Field data processing occupies a special place in our approach. Through many years of experience in studying Heliometric data, we independently developed an integrated approach to data processing and created unique algorithms that guarantee us the best result. With our AI-based solution, we are able to train the system to recognise incorrect data, highlight anomalies, and integrate a large number of factors that affect helium values. Thanks to our technology, we are confident that the final helium data is the closest to the true value.

Factors we take into account

Field data management

Over 20 different measurement parameters are recorded during the survey
All data is encrypted and stored on a secure medium
The data is duplicated by the system to prevent loss
The data is sent via satellite to the survey coordination center for further express processing and quality control

Field data processing and mapping

Initial processing is carried out for the purpose of express analysis of data quality and coordination of field work
The field data processing takes place at the survey coordination center and includes:

-checking the operation of the field system

- checking the integrity and quality of the entire spectrum of recorded data

- quality control of measurements

- verification of the accuracy of the coordinates of measurement points

- preparation and filtering of field data

- mapping of primary Heliometric anomalies

Express analysis makes it possible to compile matrices of average values of Heliometric data in the surface layer of air and in soil gas, on the basis of which maps of the distribution of anomalous Helium content are built
Final processing of data begins after the collection of a sufficient amount of field material (after 15-20 field survey days)
When processing the measurement results, different types of fluctuations, weather conditions and temperature are excluded; soil type and its’ quality, the possible influence of geodynamic processes, oil pollution of the environment are taken into account, which makes it possible to identify the trend of the measured value
The duration of the final data processing varies from 20 to 30 days, depending on the amount of information collected
For processing field data, proprietary software and a set of algorithms using machine learning elements are used to recognize the values, characteristics and types of Helium flow in each measurement, while using many field data analysis parameters
Particular attention is paid to the stage of reconciliation of the volume of data with the existing database of previous works, depending on the characteristic of geological conditions
As a result of passing through all stages of processing, for each sampling point, a value is assigned that is the only correct one, taking into account all the specified parameters
The Heliometric field parameters are normalized to the standard deviation of the measured parameter
The obtained values of the Helium content in the surface air and in the soil gas (real Heliometric values) make it possible to construct maps of the distribution of the anomalous Helium content with reference to coordinates.
Prior to the involvement of geological and geophysical data, Heliometric anomalies are ranked only by the magnitude of the anomaly (intensity) and area (linear dimensions).
Heliometric field parameters for mapping take into account abundances, anomalies and variations