Unmanned Aerial Vehicles (UAVs) are a type of remote sensing platform that can collect data and images about the Earth's surface. While they differ from traditional satellite-based remote sensing systems, UAVs share many similarities with them.

UAVs, also known as drones, are remotely piloted or autonomous aircraft that fly at various altitudes, subject to each country's laws and regulations. 

Drone cameras capture visible light, known as RedGreenBlue (RGB), within the 400-700 nm wavelength range. The choice of camera type depends on the mission, with options ranging from monochrome (grayscale) and RGB (three bands) to multispectral (five bands) and hyperspectral (many bands). Multispectral and hyperspectral cameras are particularly valuable in forestry and agriculture, capturing images beyond visible light to assess vegetation health and other environmental factors. Hyperspectral cameras, though powerful, are costly and offer diminishing returns with extended wavelengths.

Thermal cameras, or infrared (IR) cameras, detect temperature differences in their field of view by capturing infrared radiation emitted by objects. These cameras are essential in forestry for monitoring tree health and environmental conditions.

LiDAR (Light Detection and Ranging) is a remote sensing method that uses pulsed lasers to measure distances to objects, creating high-resolution 3D models of environments. By calculating the time it takes for a laser pulse to return from an object, LiDAR generates accurate 3D point clouds, which are used for mapping, geology, and urban planning applications.

Radar, specifically Ground Penetrating Radar (GPR), is used in forestry to detect subsurface features like rocks, soil composition, and water presence. GPR emits high-frequency radio waves into the ground, which reflect off subsurface objects and create detailed 2D or 3D images. Although GPR technology for UAVs is still emerging and costly, it holds promise for future forestry applications.

Photogrammetry is the process of creating 3D models or maps from overlapping images taken from different angles and perspectives. It involves using cameras to capture images of an area from multiple viewpoints, which are then stitched together to form a digital mosaic. The resulting 3D model or map can be used to create detailed topographic representations, monitor changes over time, and analyze spatial relationships between objects. 

A UAV-supported Assessment purpose is to use the drone photogrammetry analysis to support achieve the objectives of the evaluation. The UAV-supported Assessment consists of seven stages as described below.

This step involves identifying the specific location or region within the certificate holder that requires monitoring or mapping, such as areas with high conservation values, recent disturbance, or potential logging sites. The initial inspection stage is crucial as it allows for a thorough assessment of the terrain, including topography, vegetation density, and any obstacles that may impact UAV flight planning. This information is vital in determining the most effective flight route, ensuring safe and efficient data collection, and minimizing any potential environmental impacts. 

Since most of the UAV flights are expected to take place in different countries and areas that are not immediately accessible, it is worth looking at public sources of information to perform the initial inspection. Obtaining local flying permits and informing all involved parties of the drone flight is of utmost importance at this stage. 

Another key aspect of this stage is to find a purpose for each flight during the mission planning to precisely define the objectives of the flight survey. The mission objectives set at this stage will define many factors in the later stages.

Mission planning is the second stage of the UAV-supported Assessment and photogrammetry analysis, where the focus is on meticulously planning the flight path to ensure accurate and efficient data collection. This involves identifying the specific areas or features to be surveyed and creating a detailed route that takes into account factors like weather conditions, topography, and obstacles like trees or power lines. Proper mission planning is crucial for achieving high-quality results, as it allows the UAV pilot to anticipate and mitigate potential issues before takeoff, ensuring that all necessary data is collected without compromising safety or efficiency. 

The flight and data collection parameters need to be studied and set at this stage. This includes mission type (waypoints, continuous flight, etc.), mission bounds, and camera/sensor settings. A pertinent factor for data collection is the percent of forward and side overlap between each pair of captured images. Typically having a very high percentage of overlap between images allows us to process the data at a higher resolution, but the drawback is the exponential cost of processing power and time. It is therefore advised to have a minimum of 70-85% front overlap and 60-70% side overlap.

An assessment agenda must incorporate drone flights in a clear daily mission plan. A suggestion is also to create classifications depending on the main topic to be analyzed during the drone flights.

The on-site preparations and pre-flight checklist are crucial steps in performing a successful UAV survey. During this stage, the operator must ensure that all necessary equipment is present and functioning properly, including the drone itself, batteries, propellers, and any additional sensors or cameras being used. The operator must also conduct a thorough pre-flight inspection to verify that the drone's GPS, compass, and other navigation systems are accurate and calibrated correctly. Additionally, the checklist should include checks for weather conditions, ensuring that wind speed and direction are within safe operating limits, as well as verifying the presence of any obstacles or hazards in the flight area. The importance of this stage cannot be overstated, as a thorough pre-flight check can help prevent accidents or damage to equipment, and ensure that the data collected is accurate and reliable.

The fourth stage involves performing the flight survey as planned in the earlier stages. During this stage, the UAV is flown over the designated area, capturing high-resolution images or collecting data using various sensors such as cameras, LiDAR, or multispectral sensors. The flight plan was carefully planned in stage 2 to ensure thorough coverage of the area, taking into account factors like weather conditions, vegetation density, and terrain features. The importance of this stage lies in its ability to provide a comprehensive and accurate dataset for later processing and analysis. A well-executed flight survey ensures that all areas of interest are captured, reducing the risk of missed data or gaps in coverage. 

This stage has the same importance as the pre-flight checks discussed earlier. The objective of the post-flight checklist is to ensure mission objectives are met and aid in maintaining the UAV to ensure its longevity. 

Stage 6 of the UAV-supported Assessment and photogrammetry analysis involves data processing and generating photogrammetric products. This critical step takes the raw images collected from the flight mission and transforms them into valuable information. Through sophisticated software, the images are tied together to form various photogrammetric products, such as orthorectified images, orthomosaics, digital surface models, digital terrain models, and point clouds. These products provide accurate and detailed information on the areas’ structure, composition, and spatial relationships. The importance of this stage lies in its ability to unlock the full potential of the UAV data, allowing it to gain insight.

An orthomosaic is a type of photogrammetric product that combines multiple overlapping photographs taken from different angles into a single, seamless, and accurate mosaic image. This process involves geo-referencing the images to match the real-world coordinates of the area being surveyed, which enables the creation of a detailed and precise representation of the environment. Applications of othomosaics are often used to analyze tree coverage, species, activities, and several other visual inspections. 

A Digital Elevation Model (DEM) is a 3D representation of the Earth's surface, typically created from remote sensing data. DEMs are used to capture the topographic features of an area, including hills, valleys, and other terrain elements. By interpolating between point measurements, DEMs can create a continuous surface model that accurately represents the elevation values of the study area. DEMs can be used to analyze forest slope, aspect, and relief, which are crucial factors in understanding forest ecology and habitat quality. Additionally, DEMs can help identify areas prone to natural hazards like landslides or floods, enabling proactive measures for risk reduction and mitigation.

Vegetation indices (VIs) are calculated from multispectral images by combining different bands of data to quantify various aspects of vegetation health and density. These indices provide valuable information about the biological activity and physical characteristics of plants, such as photosynthetic activity, water content, and chlorophyll concentration. Common VIs include the Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), and Enhanced Vegetation Index (EVI). These indices are calculated by subtracting the reflectance values of certain bands from others, or by using mathematical formulas that incorporate multiple bands. The resulting values can range from -1 to 1, with higher values indicating greater vegetation density and health. 

The last stage of the UAV-supported Assessment and photogrammetry analysis involves an in-depth analysis of the photogrammetric products, which includes processing and interpreting the generated products. This stage requires specialized expert knowledge to extract meaningful information from the 3D point clouds, orthomosaics, and DEMs generated during the survey. In addition, specialized machine learning and artificial intelligence models can be trained and deployed to meet the missions’ objectives. For example, if a mission’s objective is to detect Palm trees and count them, a supervised machine-learning algorithm can be developed and trained on the Palm trees' data. Once a model is trained, it can be used multiple times in many different forest areas.

Many possible analysis objectives can be calculated, derived, and visualized. The choice of analysis is typically based on the mission goals. Land cover classification, plantation detection, plant health assessment, and water quality are some of the analysis objectives that can be achieved.

Incorporating drones into a business requires planning and structured implementation to optimize efficiency, data collection, and operational safety. 

Research and Drone Selection

Start by researching drone models that suit your company’s needs, such as aerial photography, surveying, or inspections. Key considerations include drone size, battery life, payload capacity, flight speed, and range. Ensure that the model aligns with your budget, accounting for extra equipment like batteries or cameras. Visiting trade shows or reaching out to different manufacturers is recommended to get an overview of different drone models.

Organizational Preparation

Before purchasing, assess the organization’s readiness:

• Regulatory Compliance: Understand and adhere to local drone laws, registering drones with national agencies if required (such as in the EU).

• Insurance: Liability insurance is essential to cover potential accidents or damages, which is legally required in many regions.

• Data Handling: Ensure your company has the infrastructure to manage the large data files drones will produce, like cloud storage or data servers.

It is recommended to plan a drone workshop to define the company’s use cases and specific needs.

Personnel and Responsibility Assignment

Designate team members to manage drone operations:

• Drone Operators: Pilots should be certified, with proper training on flight safety and regulations. In the EU, certifications like A1/A3 and A2 are required for different drone sizes and use cases.

• Data Analysts: Develop a team for analyzing the data collected by the drones to derive actionable insights.

Drone Purchase and Registration

Once everything is in place:

• Choose a drone model based on operational needs and budget.

• Obtain liability insurance to protect against accidents.

• Register the drones with the appropriate aviation authorities, following local laws.

Training and Certification

Locate certified training programs for drone operators, focusing on both flight skills and data analysis. Ensure pilots have the necessary certifications, such as the EU’s A1/A3 or A2, depending on the type and size of the drone, and receive a technical introduction to the drone model if needed. Emphasize post-flight data analysis training for efficient information processing.

Operational Roll-Out and Monitoring

With all elements in place, launch your drone operations. Regularly review performance, flight safety, and data collection practices. Conduct periodic audits to ensure continued efficiency, safety, and regulatory compliance.