Science Goes Bush: An Overview Of Hyperspectral Remote Sensing
Unmanned aerial vehicles (UAVs) have rapidly become a viable alternative to satellites and manned aircraft for conducting hyperspectral remote sensing.
Likewise, hyperspectral (HS) sensing instruments have evolved along with UAVs; both are getting smaller, lighter, and easier to use.
How Hyperspectral Imaging Works
HS sensors are attached to UAVs in order to capture slices of incoming scenes (by way of a physical slit) and they present each slice as discrete wavelength components on a focal plane array (FPA).
Dispersing the image slices into discrete wavelengths is achieved with a diffraction grating.
In addition to HS sensors, ancillary instruments such as light detection and ranging (LiDAR) and GPS can also be used to ensure complete and accurate imagery collection.
A planimetrically correct image is obtained by removing the effects of perspective (tilt) and relief (terrain) with the use of these instruments.
These portable commercial, off-the-shelf (COTS) aerial drone solutions can map terrain features of potential conflict areas and classify major material classes very easily.
The visible spectrum of electromagnetic radiation is what the human eye is capable of seeing and identifying. This ranges from about 380 nanometers to around 700 nanometers.
Ultraviolet sits below this spectrum, while infrared lies above this “visible” portion of the spectrum. Bees can see UV, while snakes see infrared. This is where HS sensors can help us.
They collect, in the case of hyperspectral, a full spectrum of image data for every pixel within the field of view.
Why Is Hyperspectral Imaging Important?
In precision agriculture, there are a variety of vegetative indices (VIs) that are dependent on seeing into infrared ranges, where fluorescence spectral signatures of chlorophyll are detectable.
It is possible to see and quantify chlorophyll fluorescence to reveal information about crop stress and vigor that cannot be seen by the naked eye.
A wide variety of vegetative indices use hyperspectral data to answer the question: Do my crops have diseases I cannot see? Do my soils have sufficient nutrients? Am I at risk for invasive species?
The Right Hyperspectral Tool For The Job
Headwall’s Nano-Hyperspec® is a completely integrated hyperspectral sensor designed for the VNIR (400-1000nm) spectral range.
A completely integrated lightweight (<0.52kg) VNIR hyperspectral sensor for small UAV applications that includes on-board data-processing/storage and GPS/IMU. Today’s UAVs are exceptionally small and light and they demand payloads to match.
A key advantage of Nano-Hyperspec is that it also includes 480GB of on-board data collection/storage, plus attached GPS/IMU functionality. This allows the payload bay of the UAV to be optimised for other needs such as video or thermal imaging. Weight and space is saved, making for a more efficient airborne solution.
Nano-Hyperspec is perfect for today’s new breed of hand-launched UAVs and drones, representing a mix of fixed-wing and multi-rotor models to meet practically any deployment scenario.
- VNIR spectral range (400-1000nm)
- 270 spectral bands, 640 spatial bands
- Maximum frame rate: 350Hz
- 480 GB internal storage
- Direct-attached GPS / IMU capabilities
- Light, small, robust
People assume they just need to buy a sensor and a UAV and put them together.
However, battery life matters, balance issues occur, and the relationship between ground speed and frame rate can make things difficult.
PAS can recommend the right kind of UAV, take its size/weight/power into consideration, integrate spectral sensors with other instruments such as LiDAR, and provide complete solutions. Contact us for more information today.
Out In The Field: Hyperspectral Remote Sensing
Hyperspectral remote sensing has many applications for the agriculture industry.
From crop disease detection to infrastructure inspection; environmental monitoring to pollution analysis; high spectral resolution for field remote sensing analysis helps with:
Using a combination of field and satellite remote sensing methods, oceanographers, limnologists and other environmental and marine researchers can perform thorough assessments of vast waterways, glaciers and wetlands.
Headwall’s Nano-Hyperspec® is about the size and weight of a Rubik’s Cube, so it represents the smallest true hyperspectral airborne sensor on the market. It holds a solid-state drive that eliminates the need to have a supplemental data computer that would consume weight and space. The GPS-IMU manages all the positioning functions as well as any roll-pitch-yaw anomalies that otherwise would impact data quality. Headwall offers a range of GPS-IMU instruments, with the most precise of them shown here.
Headwall does considerable work to assure the quality of hyperspectral data coming from its sensors. This can include stabilising gimbals (the Ronin gimbal shown here) as well as powerful orthorectification done during post-processing. Headwall manages this with its own powerful suite of applications software.
The thermal camera is small and light, which means the whole instrument payload can be easily managed by the DJI Matrice 600 Pro, which is one of the most popular enterprise-level UAVs on the market today.
Agricultural Remote Sensing
For responsible analysis and management of some of the Earth’s most precious resources, PAS has the instrument to suit your needs.
Forestry / Plant Remote Sensing
Forests and ecological resources are important assets that need to be properly managed. Accurate analysis of forest health, population, growth areas and damage is possible through sophisticated remote-sensing technologies available from PAS.
The Importance Of Good Software
The development of hyperspectral software to help with management along the way and during post-processing is just as important as reducing the size and weight of the hyperspectral sensor.
Nano-Hyperspec can be controlled using Headwall’s airborne Hyperspec III software while in the air.
During post-processing, the software also can draw in GPS data to enable orthorectification. It has a polygon tool to determine the “start-stop” coordinates for the sensors.
By using LiDAR instruments to estimate the height of objects and artifacts beneath the UAV flight path, the hyperspectral data can be represented accurately.
In order to manage UAV flights, sensor operations, and so on, a software platform with a simple interface but the ability to manage gigabytes worth of hyperspectral data is helpful.
The basic Headwall hyperspectral software performs these important tasks, but Headwall has enhanced the software to handle GPS data and orthorectification during post-processing.
Nano-Hyperspec is available in two options:
- OEM-programmable sensor configuration
- High-performance sensor configuration for end-users
PAS can guide you to the solution you require, providing sales and product support, training and upgrades as you implement leading Headwall hyperspectral technology. Speak to PAS for expert guidance on hyperspectral imaging options from Headwall Photonics.