Papers
More filters
••
TL;DR: This letter shows the first study of Transfer Learning between a simulated data set and a set of real SAR images, and shows that a Convolutional Neural Network pretrained on simulated data has a great advantage over a Convnet trained only on real data, especially when real data are sparse.
Abstract: Data-driven classification algorithms have proved to do well for automatic target recognition (ATR) in synthetic aperture radar (SAR) data. Collecting data sets suitable for these algorithms is a challenge in itself as it is difficult and expensive. Due to the lack of labeled data sets with real SAR images of sufficient size, simulated data play a big role in SAR ATR development, but the transferability of knowledge learned on simulated data to real data remains to be studied further. In this letter, we show the first study of Transfer Learning between a simulated data set and a set of real SAR images. The simulated data set is obtained by adding a simulated object radar reflectivity to a terrain model of individual point scatters, prior to focusing. Our results show that a Convolutional Neural Network (Convnet) pretrained on simulated data has a great advantage over a Convnet trained only on real data, especially when real data are sparse. The advantages of pretraining the models on simulated data show both in terms of faster convergence during the training phase and on the end accuracy when benchmarked on the Moving and Stationary Target Acquisition and Recognition data set. These results encourage SAR ATR development to continue the improvement of simulated data sets of greater size and complex scenarios in order to build robust algorithms for real life SAR ATR applications.
163 citations
••
TL;DR: A new standard on software quality requirements, ISO/IEC 25030, takes a systems perspective and suggests specifying requirements as measures and associated target values.
Abstract: A new standard on software quality requirements, ISO/IEC 25030, takes a systems perspective and suggests specifying requirements as measures and associated target values.
82 citations
••
18 Oct 2010TL;DR: A modeling framework for performing schedulability analysis by using UPPAAL real-time model-checker and a comparison of results with classical response time analysis is compared.
Abstract: We propose a modeling framework for performing schedulability analysis by using UPPAAL real-time model-checker [2]. The framework is inspired by a case study where schedulability analysis of a satellite system is performed. The framework assumes a single CPU hardware where a fixed priority preemptive scheduler is used in a combination with two resource sharing protocols and in addition voluntary task suspension is considered. The contributions include the modeling framework, its application on an industrial case study and a comparison of results with classical response time analysis.
69 citations
••
TL;DR: This paper analyzed SPI research publications with a main emphasis on organizational change using Gareth Morgan's organizational metaphors (1996) as analytical lenses and provided a roadmap to help identify insights and specific articles related to SPI as organizational change.
48 citations
••
02 Jun 2012TL;DR: A way to add the use of formal methods to the agile development process Scrum is described and experiences from using a variant of the strategy in an industrial case are summarised.
Abstract: Formal methods have had a relative low penetration in industry but have the potential for much wider use. The use of agile methods has been highly limited in development of safety-critical systems due to the lack of formal evaluation techniques and rigorous planning. A combination of formal methods and agile development processes can potentially widen the use of formal methods in industry as well as enabling the use of agile methods in development of safety-critical systems. This paper describes a way to add the use of formal methods to the agile development process Scrum. Experiences from using a variant of the strategy in an industrial case are summarised.
35 citations
Authors
Showing all 35 results
Name | H-index | Papers | Citations |
---|---|---|---|
Sune Dueholm Müller | 11 | 35 | 400 |
Morten Ostergaard Pedersen | 10 | 27 | 329 |
K. Hansen | 10 | 19 | 454 |
Sune Wolff | 6 | 13 | 146 |
Samant Khajuria | 4 | 19 | 88 |
Lars Wurtz Jochumsen | 3 | 6 | 25 |
A.C.K. Thomsen | 3 | 6 | 18 |
M. Lokke | 3 | 4 | 11 |
O. Marqversen | 3 | 5 | 16 |
Esben Nielsen | 3 | 6 | 18 |
Rasmus Engholm | 2 | 4 | 106 |
Hans Jensen | 2 | 5 | 29 |
Jan Storbank Pedersen | 2 | 2 | 63 |
Soren Kjarulff Christensen | 1 | 3 | 3 |
Jakob Krogh-Sørensen | 1 | 1 | 3 |