Ms. Kristin Hammarstrøm Løkken Profile

Kristin Hammarstrøm Løkken

Senior scientist at Norwegian Defence Research Establishment

SPIE Involvement:

Author

Publications (6)

Proceedings Article | 20 September 2020 Paper

Investigating robustness of adversarial camouflage (AC) for naval vessels

Kristin Hammarstrøm Løkken, Lars Aurdal, Alvin Brattli, Hans Christian Palm

Proceedings Volume 11543, 115430G (2020) https://doi.org/10.1117/12.2573676

KEYWORDS: Camouflage, Gallium nitride, Neural networks, Visualization

Read Abstract +

The use of camouflage is widespread in the biological domain, and has also been used extensively by armed forces around the world in order to make visual detection and classification of objects of military interest more difficult. The recent advent of ever more autonomous military agents raises the questions of whether camouflage can have a similar effect on autonomous agents as it has on human agents, and if so, what kind of camouflage will be effective against such adversaries. In previous works, we have shown that image classifiers based on deep neural networks can be confused by patterns generated by generative adversarial networks (GANs). Specifically, we trained a classifier to distinguish between two ship types, military and civilian. We then used a GAN to generate patterns that, when overlaid on parts of military vessels (frigates), made the classifier confuse the modified frigates with civilian vessels. We termed such patterns "adversarial camouflage" (AC) since these patterns effectively camouflage the frigates with respect to the classifier. The type of adversarial attack described in our previous work is a so-called white box attack. This term describes adversarial attacks that are devised given full knowledge of the classifier under attack. This is as opposed to black box attacks, which describe attacks on unknown classifiers. In our context, the ultimate goal is to design a GAN that is capable of black box attacks, in other words: a GAN that will generate AC that has effect across a wide range of neural network classifiers. In the current work, we study techniques to improve the robustness of our GAN-based approach by investigating whether a GAN can be trained to fool a selection of several neural network-based classifiers, or reduce the confidence of the classifications to a degree which makes them unreliable. Our results indicate that it is indeed possible to weaken a wider range of neural network classifiers by training the generator on several classifiers.

Proceedings Article | 24 April 2020 Paper

Robustness of adversarial camouflage (AC) for naval vessels

Kristin Hammarstrøm Løkken, Alvin Brattli, Hans Christian Palm, Lars Aurdal, Runhild Aae Klausen

Proceedings Volume 11394, 113940W (2020) https://doi.org/10.1117/12.2558506

KEYWORDS: Camouflage, Neural networks, Image filtering, Evolutionary algorithms, Optical sensors

Read Abstract +

Proceedings Article | 19 September 2019 Presentation + Paper

Adversarial camouflage for naval vessels

Lars Aurdal, Kristin Hammarstrøm Løkken, Runhild Aae Klausen, Alvin Brattli, Hans Christian Palm

Proceedings Volume 11169, 111690K (2019) https://doi.org/10.1117/12.2532756

KEYWORDS: Camouflage, Neural networks, Image analysis, Detection and tracking algorithms, Sensors, Network architectures, Evolutionary algorithms, Visualization, Artificial intelligence

Read Abstract +

Proceedings Article | 9 October 2018 Paper

Supporting artificial intelligence with artificial images

Lars Aurdal, Alvin Brattli, Eirik Glimsdal, Runhild Aae Klausen, Kristin Hammarstrøm Løkken, Hans Christian Palm

Proceedings Volume 10794, 107940M (2018) https://doi.org/10.1117/12.2324969

KEYWORDS: Infrared imaging, Long wavelength infrared, Neural networks, Image processing, Image segmentation, Infrared signatures, Network architectures

Read Abstract +

Proceedings Article | 26 April 2018 Paper

Detection metrics and ship [D]RI

Arthur van Rheenen, Lars Heen, Eirik Madsen, Erik Brendhagen, Kristin Løkken, Bernt Almklov, Eirik Glimsdal

Proceedings Volume 10625, 106250V (2018) https://doi.org/10.1117/12.2304583

KEYWORDS: Long wavelength infrared, Mid-IR, Spatial frequencies, Target recognition, Minimum resolvable temperature difference, Target detection, Sensors, Cameras, Targeting Task Performance metric, Spatial resolution

Read Abstract +

Proceedings Article | 3 June 2013 Paper

An ATR architecture for algorithm development and testing

Gøril Breivik, Kristin Løkken, Alvin Brattli, Hans Palm, Trym Haavardsholm

Proceedings Volume 8744, 874402 (2013) https://doi.org/10.1117/12.2015369

KEYWORDS: Image segmentation, Cameras, Automatic target recognition, Algorithm development, Image processing algorithms and systems, Image processing, Infrared radiation, Detection and tracking algorithms, Long wavelength infrared, Infrared cameras

Read Abstract +

Showing 5 of 6 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years