The types of remote sensing can be classified based on various criteria that reflect their applications and technologies. One of the main ways to classify them is by the ranges of electromagnetic radiation, such as visible light, infrared and ultraviolet radiation, microwaves, and radio waves. Additionally, remote sensing types can be divided into active and passive, depending on whether the system uses its own radiation source or records reflected radiation. There is also a classification based on the type of platform on which the sensing sensors are mounted, such as satellites, unmanned aerial vehicles (drones), or balloons. Each of these types has its own features and provides valuable information for a wide range of applications, from monitoring climate changes and the environment to urban development planning and agriculture.

• Features: Measures reflected or emitted electromagnetic radiation in discrete wavelengths across a wide spectrum with small bands.
• Advantages: Allows obtaining spectral characteristics of objects over a wide spectrum range, provides high capability for distinguishing surface types and materials.

There are also other categories of remote sensing.

In passive sensing, measurements are made using the radiation emitted from Earth's surface objects, thus not requiring the use of a radiation source in the necessary part of the spectrum.

In active sensing, the system uses its own radiation source to illuminate the Earth's surface and then records and analyzes the signal returned from the objects.

For example, RADAR and LiDAR sensors are typical active remote sensing instruments that measure the time delay between emission and return to determine the location, direction, and speed of an object. The collected remote sensing data is then processed and analyzed using remote sensing equipment and computer software (the most sophisticated solutions are analytical products presented in near real-time), which are available in various applications, primarily in geographic information systems (GIS).

• Specialized devices that allow for aerial photography without a person on board.
• Application: Used in monitoring fields, forests, and forest-industrial areas for safe and efficient problem recognition. Offers the highest spatial resolution.

For example, the summarized characteristics of the most in-demand remote sensing data sources on the market today are shown in Table below.

Globally, remote sensing is needed to explain and make decisions in two areas of human life - changes in the environment and infrastructure for any human activity.

Remote sensing has over 250 thematic directions and every year it grows with new directions as an innovative field of research.

Remote sensing monitoring is needed for detecting anomalies on land, in water, in the air, in cities, on roads, during natural disasters, or due to human factors, whether limited in size or extensive. Spatial factors over time play a role everywhere. Fixing anomalies and analyzing their appearance is what remote sensing does.

Goal: To make effective decisions resulting from remote sensing analysis that contribute to improving and ensuring the safety of life and society, preserving and restoring nature, and preventing infrastructure destruction.

Food Security – undoubtedly the primary and crucial task of remote sensing monitoring is food security and sufficiency for survival in the environment:

• Tracking the development of agricultural crops from sowing to harvest based on LULC remote sensing technology. Mapping indices, field condition indicators, assessing weeds, drought, salinity, pest infestations. Precision farming.
• Monitoring the cleanliness of water resources. Registering water pollution.
• Searching for new water sources in adverse areas.

Tasks:
Environmental monitoring of oil pollution and creation of a thematic map of pollution in licensed areas for a specific period.
Using satellite images with high spatial resolution (1.5 m), the following pollution areas need to be identified:

• Areas littered with industrial and consumer waste;
• Zones of accumulation of high-viscosity bituminous oil, liquid oil;
• Polluted areas where self-recovery is highly likely to occur within 1-2 years; Polluted areas where due to high salinity reclamation in the next season is impossible;
• Areas of continuous dead forest mass on pollution and littered sites.

Solution:
To decipher oil pollution in licensed areas (LUs), satellite images TripleSat and SPOT were used, which have a resolution of 1 m and 1.5 m in the panchromatic channel and 4 m and 6 m in the multispectral channels.
The following processing of the obtained images was carried out:

• Orthotransformation – to increase the accuracy of image georeferencing;
• Atmospheric correction – to eliminate the negative impact of the atmosphere on the decoding results;
• Pan sharpening – to enhance the spatial accuracy of the image.

Test sites were selected for decoding to choose the most optimal and accurate method. Various classification methods with training (Spectral Angle Mapper, Maximal Likelihood, etc.), spectral indices (NDVI, etc.), and color synthesis options were tested.

The most accurate method is visual expert decoding by specialists in remote sensing of the Earth (RSE), since using classifications reveals a large number of false objects that are not of practical interest. Various combinations of spectral channels in combination with the NDVI index were used for the decoding process.

Result:
The result of the work was a thematic map of various oil pollution in the form of a vector layer for loading into any geographic information system.

Solution:
1. 115 radar images (from the last 5 years) from the Sentinel-1A/B satellite were analyzed. A number of images were excluded from the full set of available images due to natural and climatic factors at the time of capture that prevented reliable measurements (presence of snow cover, thunderstorms, ionosphere activity). 37 images were suitable for further use;
2. The digital model of the field area was built based on the interferometric processing of a pair of radar images from the Sentinel-1B satellite.
3. Interferometric processing of the obtained series of radar images was performed using the persistent scatterers (PS) method;

4. Using the results of the interferometric processing, a map of average annual displacement rates was created;
5. Vertical displacements measured by high-precision leveling were compared with line-of-sight (LOS) displacements obtained by radar interferometry;
6. After obtaining vertical ground surface displacements, deformation calculations were conducted;
7. After calculating quantitative assessments of surface deformation parameters, potential geodynamic risks were evaluated;

8. Overall conclusions were drawn and relevant recommendations were provided;
9. The relationship between hydrocarbon production and ground subsidence was analyzed.

Tasks:
Obtaining optical satellite imagery and UAV imagery for further use in pipeline route patrolling tasks.

Solution:
1. Ordered archival imagery for the work areas from "Kanopus-V" and "Resurs-P" satellite images. Evaluated spectral separability of classes to identify mining objects. Results showed which objects can be successfully classified in the images and which may cause confusion, requiring more careful selection of standards and visual control by specialists during interpretation.
2. Updated and supplemented data from a previously created GIS.
3. In the NextGIS QGIS software, data was updated using new information. The resulting GIS map-scheme contained geoinformation layers that reflected preliminary information about all "mining objects," quarry areas, information on minerals, presence of dumps, etc.
4. Interpreted images and classified data.
5. Selected 40 images from 65 archival data, prioritized images without snow or clouds, summer-autumn period. Interpreted images visually, then classified and combined data into one layer.
6. Analyzed the obtained objects in conjunction with existing subsurface fund data and prepared for field studies. In addition to data integration, a special "confidence" scale was developed:

• 25% – quarry presence is unlikely;
• 50% – quarry presence is moderately likely;
• 75% – quarry presence is highly likely;
• 100% – the object is a quarry.

7. Data was evaluated accordingly and all other relevant attribute data was added (quarry condition, SWD presence, extraction area, type of mineral, satellite image used for interpretation, year of shooting).
8. Several road graphs were created and routes were planned for visiting objects of interest in preparation for field work.
9. The final stage of the project involved field surveys of objects of interest. As a result of the field work, 43 potential mining objects were surveyed, 21 of which were found to be quarries.

Tasks:
Conduct monitoring of urban changes using ultra-high spatial resolution satellite imagery.

Result:
A complete set of urban change monitoring results. The report includes cartographic materials highlighting changes, analytical data, and recommendations for further actions. The customer can use this data for informed decision-making and effective urban management.