Benchmarking Deep Learning for On-Board Space Applications
Benchmarking deep learning algorithms before deploying them in hardware-constrained execution environments, such as imaging satellites, is pivotal in real-life applications. Although a thorough and consistent benchmarking procedure can allow us to estimate the expected operational abilities of the u...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/19/3981 |
| _version_ | 1797515781313921024 |
|---|---|
| author | Maciej Ziaja Piotr Bosowski Michal Myller Grzegorz Gajoch Michal Gumiela Jennifer Protich Katherine Borda Dhivya Jayaraman Renata Dividino Jakub Nalepa |
| author_facet | Maciej Ziaja Piotr Bosowski Michal Myller Grzegorz Gajoch Michal Gumiela Jennifer Protich Katherine Borda Dhivya Jayaraman Renata Dividino Jakub Nalepa |
| author_sort | Maciej Ziaja |
| collection | DOAJ |
| description | Benchmarking deep learning algorithms before deploying them in hardware-constrained execution environments, such as imaging satellites, is pivotal in real-life applications. Although a thorough and consistent benchmarking procedure can allow us to estimate the expected operational abilities of the underlying deep model, this topic remains under-researched. This paper tackles this issue and presents an end-to-end benchmarking approach for quantifying the abilities of deep learning algorithms in virtually any kind of on-board space applications. The experimental validation, performed over several state-of-the-art deep models and benchmark datasets, showed that different deep learning techniques may be effectively benchmarked using the standardized approach, which delivers quantifiable performance measures and is highly configurable. We believe that such benchmarking is crucial in delivering ready-to-use on-board artificial intelligence in emerging space applications and should become a standard tool in the deployment chain. |
| first_indexed | 2024-03-10T06:52:08Z |
| format | Article |
| id | doaj.art-7352bc8b2ac9452ba72ea94d8ebc83ce |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T06:52:08Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-7352bc8b2ac9452ba72ea94d8ebc83ce2023-11-22T16:43:43ZengMDPI AGRemote Sensing2072-42922021-10-011319398110.3390/rs13193981Benchmarking Deep Learning for On-Board Space ApplicationsMaciej Ziaja0Piotr Bosowski1Michal Myller2Grzegorz Gajoch3Michal Gumiela4Jennifer Protich5Katherine Borda6Dhivya Jayaraman7Renata Dividino8Jakub Nalepa9KP Labs, Konarskiego 18C, 44-100 Gliwice, PolandKP Labs, Konarskiego 18C, 44-100 Gliwice, PolandKP Labs, Konarskiego 18C, 44-100 Gliwice, PolandKP Labs, Konarskiego 18C, 44-100 Gliwice, PolandKP Labs, Konarskiego 18C, 44-100 Gliwice, PolandGSTS—Global Spatial Technology Solutions, Dartmouth, NS B2Y 4M9, CanadaGSTS—Global Spatial Technology Solutions, Dartmouth, NS B2Y 4M9, CanadaGSTS—Global Spatial Technology Solutions, Dartmouth, NS B2Y 4M9, CanadaGSTS—Global Spatial Technology Solutions, Dartmouth, NS B2Y 4M9, CanadaKP Labs, Konarskiego 18C, 44-100 Gliwice, PolandBenchmarking deep learning algorithms before deploying them in hardware-constrained execution environments, such as imaging satellites, is pivotal in real-life applications. Although a thorough and consistent benchmarking procedure can allow us to estimate the expected operational abilities of the underlying deep model, this topic remains under-researched. This paper tackles this issue and presents an end-to-end benchmarking approach for quantifying the abilities of deep learning algorithms in virtually any kind of on-board space applications. The experimental validation, performed over several state-of-the-art deep models and benchmark datasets, showed that different deep learning techniques may be effectively benchmarked using the standardized approach, which delivers quantifiable performance measures and is highly configurable. We believe that such benchmarking is crucial in delivering ready-to-use on-board artificial intelligence in emerging space applications and should become a standard tool in the deployment chain.https://www.mdpi.com/2072-4292/13/19/3981on-board processingdeep learningbenchmarkingsegmentationclassificationdetection |
| spellingShingle | Maciej Ziaja Piotr Bosowski Michal Myller Grzegorz Gajoch Michal Gumiela Jennifer Protich Katherine Borda Dhivya Jayaraman Renata Dividino Jakub Nalepa Benchmarking Deep Learning for On-Board Space Applications Remote Sensing on-board processing deep learning benchmarking segmentation classification detection |
| title | Benchmarking Deep Learning for On-Board Space Applications |
| title_full | Benchmarking Deep Learning for On-Board Space Applications |
| title_fullStr | Benchmarking Deep Learning for On-Board Space Applications |
| title_full_unstemmed | Benchmarking Deep Learning for On-Board Space Applications |
| title_short | Benchmarking Deep Learning for On-Board Space Applications |
| title_sort | benchmarking deep learning for on board space applications |
| topic | on-board processing deep learning benchmarking segmentation classification detection |
| url | https://www.mdpi.com/2072-4292/13/19/3981 |
| work_keys_str_mv | AT maciejziaja benchmarkingdeeplearningforonboardspaceapplications AT piotrbosowski benchmarkingdeeplearningforonboardspaceapplications AT michalmyller benchmarkingdeeplearningforonboardspaceapplications AT grzegorzgajoch benchmarkingdeeplearningforonboardspaceapplications AT michalgumiela benchmarkingdeeplearningforonboardspaceapplications AT jenniferprotich benchmarkingdeeplearningforonboardspaceapplications AT katherineborda benchmarkingdeeplearningforonboardspaceapplications AT dhivyajayaraman benchmarkingdeeplearningforonboardspaceapplications AT renatadividino benchmarkingdeeplearningforonboardspaceapplications AT jakubnalepa benchmarkingdeeplearningforonboardspaceapplications |