The goal of the operator is to determine an appropriate image resolution to achieve the measurement objective while minimizing the time spent flying and processing.
The following sections discuss the details of planning a proper data collection mission. In general, increasing image resolution (flying lower and slower) improves the quality of results at the expense of increased time spent flying and processing data.
High quality raw data is essential for accurate results. For airborne multispectral sensors like your SlantRange system, that means imagery with sharp contrast and sufficient resolution to resolve the plants you are measuring, as well as good supporting data on ambient lighting conditions, and the position and attitude of your aircraft at each time of measurement.
The Remote Pilot has control over four aspects of mission planning.
- Altitude: set flight altitude based on the size of the plants and the type of data you need to collect (Stress vs. Population Analysis). Automated flight calculators are available in Beyond the Cube's Flight Plan app and the SlantRange app in Drone Deploy.
- Speed: set flight speed based on altitude according to the flight calculator, and lighting conditions. Darker, overcast days and morning/evening flights may require slower flight to prevent image blur given the requirement for increased exposure time.
- When to fly: determine when to fly based on plant growth stage and weather conditions.
- Flight direction: set based on field layout and wind conditions. Fly perpendicular to prevailing winds, and along the crop rows when possible.
We often get calls from our customers after setting up their shiny new sensor and they ask questions like, "So what altitude should I fly? And how fast should I fly?" And our answer is always the same, "It depends on what you want to measure."
One of the major advantages of drone-based imagery is the very high resolution you can achieve, which we've leveraged to enable the unique data products we produce in SlantView, such as plant counts, plant size, separation of crops and weeds, and many others. But there are some rather complex relationships and trade-offs between resolution, altitude, speed, and area coverage, such that you may not always want to collect data at the highest resolution achievable with your sensor. And more importantly, some data products don't require the same resolution as others. For example, to count plants you need to fly low enough to achieve the resolution required to separate one plant from its neighbors along a row. And of course, plants have different sizes at different growth stages. Similarly, you don't need as much resolution to measure general stress conditions on a more mature crop.
This is a good point to quickly mention: the confusion that often arises around the word "resolution". It is important to note that resolution is often used to describe the number of pixels in a camera ("I'm using a camera with 12 Megapixel resolution") but is also used to describe the spatial clarity of an image ("My SlantRange sensor provides 4.8 cm resolution at 120 meters altitude.") These are clearly two very different concepts. For the latter case, we prefer the term "Ground Sample Distance" (GSD), which describes the spatial dimension represented by one pixel in an image. But for the purposes of this article, we will stick with "resolution".
When using the calculators in the Flight Plan or DroneDeploy apps, your flight plans will be optimized for the specific conditions of your crops and the specific data products you need to support your decision making. No more guessing. No more trial and error. And no more wasted time on flights that produce bad data. It's like you have one of our remote sensing experts built right into your flight planner!