Miniature Hyperspectral Sensors for Imaging on Small UAVs

Small hyperspectral imaging sensors yield critical information not available with digital video or thermal cameras.

As unmanned vehicles get smaller and smaller, operational expectations and mission objectives demand a much broader array of platform capabilities — a necessity even with reduced payloads requiring less weight, smaller size, and less power consumption. An interesting dichotomy exists between the need for increasing amounts of valuable information and the operational ability to capture and synthesize this information in real time.

Figure 1. Hyperspectral Image taken from UAV with targets highlighted. (Headwall Photonics, Inc.)

A key example is seen when one takes stock of the trends driving the development and deployment of small unmanned aerial vehicles (UAVs) and small unmanned ground vehicles (SUGVs). With obvious advantages, smaller unmanned vehicles — whether airborne systems or ground-based robots — provide substantially greater operational benefits in theater with lower risk for the warfighter. As a result, military and defense agencies are looking to increase the capabilities of unmanned vehicles by incorporating very small hyperspectral imaging sensors. These hyperspectral sensors yield critical information not available with digital video or thermal cameras concerning objects on the ground in the context of a three-dimensional view comprising both spatial position and chemical composition.

Figure 2. Example of a miniature Hyperspectral Sensor. (Headwall Photonics, Inc.)

Hyperspectral imaging represents a key advancement in sensor technology that is increasingly utilized as an essential payload component. Based on reflectance spectroscopy, these microhyperspectral sensors capture all spectral information within the field of view and store this information as a hyperspectral datacube. These sensors are used to assess objects within the field of view based on unique chemical “fingerprints.” For example, camouflage netting can be easily discerned from surrounding vegetation as a result of its uniquely different chemical spectrum. Or, for any point on the ground or for any object in the scene, the complete chemical spectra can be generated and interrogated against known threats or assets. Depending upon mission objectives, the hyperspectral datacube can be processed for many applications including target identification and tracking, surveillance and reconnaissance, spectral tagging for special operations, border patrol and interdiction, and more.

Hyperspectral imaging is a proven technique that was first implemented in the 1980s for remote sensing of the Earth. In the mid 1990s, military and defense requirements spawned a new set of tasks that were well suited to the capabilities of these imaging spectrometers. With years of implementation experience, hyperspectral instruments have been designed to provide high-resolution imagery with no image distortion (typically known as keystone or smile), which is a key concern for imaging in harsh or hostile environments. This imaging performance is a result of a concentric imaging design that utilizes a high-efficiency, holographic diffraction grating as the dispersive optic in the system. Standard spectral ranges consist of the visible-near-infrared range (VNIR 400 to 1,000 nm), near-infrared (NIR 900 to 1,700nm), and shortwave infrared range (SWIR 1,000 nm to 2,500 nm).

Through careful design, small hyperspectral sensors are optimized to provide for a wide field of view (FOV) and enhanced instantaneous field of view (IFOV), which is an indicator of spectral and spatial resolving power of the sensor.

With UAVs such as the U.S. Army’s Raven (hand-launched, 4.2 pounds, 5-foot wingspan) being deployed in ever increasing numbers, sensor designers and optical engineers will continue to be challenged to develop sensing instrumentation that not only provides critical information but does so in smaller and smaller package sizes.

This article was written by David Bannon, co-founder of Headwall Photonics, Fitchburg, MA. For more information, contact Mr. Bannon at This email address is being protected from spambots. You need JavaScript enabled to view it., or visit http://info.hotims.com/10970-523 .