At MicaSense we strive to help growers, land managers, and researchers use drone-based multispectral imagery to make informed decisions. Our company is comprised of remote sensing experts, engineers, and GIS professionals passionate about developing high quality sensors that are dependable in the field, produce repeatable data, and are compatible with as many software platforms as possible. Our users are from many industries such as farming, forestry, and environmental land management. To learn more about how data from our sensors can be applied, visit our blog at micasense.com/blog.
There are five basic steps involved in the process:
- Choose the sensor and drone that best fits your needs
- Integrate the sensor onto your aircraft
- Collect overlapping images of the area of interest
- Process the data into a reflectance map
- Analyze the data and make decisions
Choosing a Sensor
We currently offer three sensors to choose from, each with different imaging capabilities. All our sensors integrate with most UAV platforms and come with a DLS 2 light sensor, Calibrated Reflectance Panel, and necessary cabling.
RedEdge-MX - The industry standard five band sensor.
The RedEdge-MX is a rugged, built to last, professional multispectral sensor equipped with the spectral bands required for basic vegetative health indexes and additional bands needed for advanced analytics.
Altum- A revolutionary 3-in-1 solution for unparalleled sensing.
Altum is the ultimate solution for accuracy, flexibility, and power, capturing synchronized thermal, multispectral, and high resolution imagery and producing aligned outputs for advanced analytics.
Dual Camera- Double the spectral resolution with half the hassle.
The RedEdge-MX Dual Camera Imaging System is a synchronized 10-band solution for advanced remote sensing and agricultural research. This solution seamlessly integrates two five band cameras, the RedEdge-MX and the new RedEdge-MX Blue, capturing 10-band imagery in one flight.
For the newest sensor information, visit micasense.com
Choosing a Drone
The first step in choosing an aircraft is deciding between a fixed wing (resembles an airplane) and multirotor (helicopter with multiple propellers). Fixed wing drones generally have a longer flight time, can fly faster, and can cover more area in a single flight. Multirotor drones are usually easier to fly, don’t need as much space to take off and land, and are less expensive.
In general, we recommend using an aircraft that can support the weight of the camera and any additional payload that it needs to carry to support your operations. Ideally the aircraft should have enough power to fly an entire mission with the full payload, but multiple flight missions (where batteries are exchanged) are also common. Both fixed-wing and multirotor options work well for this type of data collection. Another factor to consider is how easy it is to integrate the camera onto the aircraft - we discuss some options for this in the following section.
If you have questions on the best drone for use, please see our Drone Solutions Page where you can find fixed wing drones that have MicaSense integrations or drone resellers who can advise you on the best multirotor options for you.
Integrating the camera
Use a pre-made integration kit
The simplest way to integrate a camera is to use an integration kit or purchase a fixed wing drone that already has the sensor integrated. We offer integration kits for the Matrice and Inspire lines of DJI drones. Many of our partners also offer integration kits for specific drones or will perform the integration for you.
Do it yourself
If you plan to do the integration yourself, please use the appropriate integration guide which can be found on our knowledge base. The most important things to consider are:
- Powering the camera
- Placement of the light sensor (DLS), which should be the topmost item on the aircraft.
- Providing GPS to the camera (all our camera kits include a GPS unit embedded in the DLS 2, but you can integrate your own GPS should you wish to use a PPK or RTK solution instead).
MicaSense offers a free iOS application called Atlas Flight which automatically works with DJI drones and MicaSense sensors. If using a third-party flight planner, ensure that it supports the camera parameters or you will not collect data with the expected overlap (which is one of the keys to a successful mission).
It is not necessary or common for a third-party flight planner to trigger the camera. That is ok! The camera will actually trigger itself if you set it into overlap/GPS mode. Please see MicaSense Sensor: Automatic Triggering Options for more details. If you are flying a DJI M200 or M300 series drone you can purchase a sensor with a PSDK skyport which will allow you to have the drone trigger your sensor while flying with one of the DJI flight apps.
The goal of processing is to stitch the overlapping images together to form what is known as a reflectance map, which is a mosaic of the area of interest where each of pixels in the image represents the actual reflectance of the imaged object. The overall process is generally called photogrammetry.
Data captured with MicaSense sensors can be processed with a variety of photogrammetry applications. Please see our Software Solutions page for a list of software solutions that currently process data captured with our sensors or see our Image Processing Workflow
Do it yourself
For users and partners that would like to implement their own raw image processing workflow, we have published a series of tutorials and software on the use of MicaSense imagery and metadata for achieving the best results. You can find the tutorials and code in Github here: MicaSense Image Processing. This allows you to calibrate and align your data for individual images to be stacked. If you do want an orthomosaic of your entire flight you will still need a program that stitches your images together.
A reflectance map can be analyzed using a number of different indices, composites, and other insightful formulas.
Processed MicaSense sensor data can also be analyzed in a variety of applications, including open-source applications like QGIS, which can generate NDVI and other agricultural indices and color composites. With software like this, you will have to know which formulas to use and how to set up the view in order to find the best way to analyze the data. We've written a brief introduction to how to use the "raster calculator" in QGIS to create useful analytics.
If you are not interested in processing and doing the analysis yourself, a number of companies provide analysis services. Many of our partner companies provide this information but generally they are focused on a specific industry or group of crops.
Understanding the data
Each crop and situation is different. For best results, we recommend sharing the data and consulting with a local agronomist, forester, or conservationist who is familiar with the vegetation, weather patterns, and growing practices of the region. Many of the analytics companies will assist you in understanding the data and giving you advice on how to proceed with it.
For some basic analysis tips and general ideas on how others have used this type of data, please see our blog or publications.. For example, you can see how researchers were able to analyse forest health using multispectral data together with satellite data.