Before the hyperspectral imager can be used, it must be calibrated. The calibration provides calibration coefficients that are applied to the data before further analysis. Spectral calibration maps from spectral pixels to actual wavelengths, which is needed to recognise spectral signatures in the data. Radiometric calibration converts from sensor counts to spectral radiance, which is used to know the intensity of the light observed.

One of the on-board tasks is going to be target detection, but in order to be able to detect features or targets in the acquired images, it is beneficial to have images with a spatial resolution as high as possible. If the resolution is low, the task of differentiating between objects becomes hard. Higher spatial resolution can be achieved by doing on-board processing of images. One of the possible methods is deconvolution algorithms, which are based on an inverse convolution.

The image to the left is an image as observed, it is blurry and has low resolution. The image to the right is the same image after deconvolution.

 

Written by Karine Avagian

Yesterday we had a great workshop with Idletechs' Petter and Harald. Idletechs are working together with HYPSO on the on-board processing. As you can see, the whiteboard is an essential tool when building satellites!

We also have a gigantic one in our lab that is in frequent use.

The wealth of spatio-spectral data obtained using modern hyperspectral imager has brought in new challenges in the analysis and extraction of useful information from hyperspectral datasets. That is why we are developing an on-board processing system which will reduce the data downlinked to our ground stations. A significant part of the on-board pipeline is target detection. The objective of target detection algorithms is to find an object of interest in the hyperspectral image. For example, a target detection system can determine the algae species based on its spectral signature and many specific plant pigments as well as certain organic and inorganic compounds.

Detecting targets with different dimensionality reduction techniques.

Written by: Đorđije Bošković

One of the main motivations behind many space technology projects and applications, is the desire to learn more about the Earth and to use this knowledge to improve the society and improve our lives. This is especially true for many Earth observation missions, such as our HYPSO.

EU and ESA teamed up in 2016, and engaged in common projects for space outreach. One of their goals is to better educate everyone on how you employ space technology in your daily life, even if you are aware of it or not. Two infographics showing this are shown here.

The same motivation is shared by many other Norwegian activities. For example, to identify possible areas for technology transfer, both between space and "Earth", and from current industry to enhance space projects.

Some of these activities are presented at a Tekna breakfast meeting at Work Work the 23rd of January. Contributors from the Norwegian Space Agency, NTNU and CiRIS will give you a short introduction to some of the exciting projects currently in motion. Sign up here!

https://www.tekna.no/kurs/frokost-i-rommet-37547/

 

Written by Roger Birkeland

In addition to the previously mentioned HSI (HyperSpectral Imaging) mission, we have another mission for the Software Defined Radio (SDR) in the pipeline.

The longer-term goal of the SDR mission is to provide Arctic researchers with easier and faster access to scientific data. We want to design a flexible communications system that includes SmallSats and different kinds of sensor equipment. To make it flexible we will use an SDR (Software Defined Radio) so that configuration can be changed easily.

Written by Gara Quintana Diaz.

For the HSI camera to capture something interesting, HYPSO's attitude will be controlled such that the camera points towards parts of the oceans we want to observe. Magnetorquers and Reaction Wheels are used to avoid the satellite spinning from perturbations or simply pointing in the wrong direction. By modeling the physics of the system, it is possible to simulate how the satellite would move during pointing or a slew maneuver under actuation. 

The goal is to set a reference attitude and angular rate such that the camera would slew across a target area to achieve sufficient overlapping pixels to be fused in Super-Resolution algorithms and enhance the image resolution.

Written by Bjørn Kristiansen

We have returned form Porto! It was great to meet up with the Underwater Systems and Technology Laboratory at  U.Porto, and to get some insight into how their successful organization is doing things. (https://lsts.fe.up.pt/)

Due to some troublesome weather and a series of unfortunate events we were not able to do ALL the things we had planned, but we still managed to get some nice pictures and a lot of experience for future field trials of the sensor.

Written by: Sivert Bakken

Each image that the HYPSO mission captures must be downlinked to one of our ground stations. In doing so, the image file passes through multiple subsystems and is transferred over several communication links, including a radio link which is somewhat undependable by nature. At any point in this network, packets may be corrupted or lost, and the connection to the satellite may suddenly be disrupted.

Current work is being done to make sure that our protocols are able to safely transfer images despite the many challenges that are met.

 

By Magne Hov

Orbit NTNU is a student organization founded in the spring of 2018. Their goal is to develop, test and launch CubeSats and act as an incubator for space interest among the students at NTNU. The organization is based on the students’ interest in space technology and wish for making use of theory in practical applications. Thus, with its 40 members from different engineering fields of study Orbit brings together the separate disciplines in challenging and extremely rewarding projects.

Trick Question! You don't, not for sure.

But to get a close estimate it is important to do tests.
Next week we'll be flying a prototype of the instrument intended to go to space.

It will be flown in Porto as there is still a chance of algal blooms there!
Wish us luck!

Written by: Sivert Bakken

To ensure the reliability of our imaging processing pipeline, it is necessary to test each constituent module. Because we have not collected satellite-based data with the v6 hyperspectral imager yet (see previous post), we are testing some of the modules on images created from a simulated push-broom technique. 

One of the properties that the image registration and motion blur correction modules must account for is uncertainty in the attitude (how the satellite is positioned) of the satellite. Below are three simulated images[1] that show the types of distortions that occur in  pushbroom  imaging when the satellite experiences periodic angular motion in the pitch, roll and yaw directions. The distinct boundaries in the original image of the coffee cup help to accentuate the distortions, which are more difficult to notice in images without distinct boundaries, such as the ocean. In the future, we will test the algorithms on data simulated from the slewing maneuver and on images with fewer sharp boundaries, such as the open ocean. Stay tuned to hear more about our image processing techniques!

[1] coffee cup from scikit-image: http://scikit-image.org/docs/dev/api/skimage.data.html#skimage.data.coffee
"CC0 by the photographer (Rachel Michetti)"

 

Written by Joseph Garrett.

Today, the V6 hyperspectral imager (a launch prototype) had its first flight aboard the NTNU UAVlab's octocopter at the Udduvoll airfield near Trondheim. A pattern was flown over both land and river, flying 2 m/s at 100m altitude in overcast conditions.

Lots of data to analyze now as we work out the details for our next flight tests in Porto this November, stay tuned!

 

Special thanks to our pilot, Pål Kvaløy, and João Fortuna. 

Written by Elizabeth Prentice.

A couple of weeks ago, Endre Dåvøy from Orbit was kind enough to make us a new badge! 

We've already started using it in templates and such, and look forward to including it on more material.

 

 

On of the darkest days of the year; not an ideal day for oceanographic observations in the Norwegian waters, two applications securing funding and more activity were approved.

The Norwegian Research Council gave the green light for our application for a project named MASSIVE - Mission-oriented autonomous systems with small satellites for maritime sensing, surveillance and communication (news-link). This opens up the posibilities for more PhD-scholarships, satellite hardware and more.

In addition, the Norwegian Space Centre approved our application for a project called Construction and testing of a prototype of a hyper spectral camera for small satellites (news-link). 

In traditional Christmas style - we have a surprise for December 1st:

Our new webpage!

We will add more content in the coming weeks, and hope you will give us some feedback on what you want to see.

The HSI mission team will also meet up for dinner after work, to get to know each other better and have some fun outside of campus.