Onboard Image Processing for Small Satellites

Arechiga, Austin P.; Michaels, Alan J.; Black, Jonathan

doi:10.1109/naecon.2018.8556744

Cited by 24 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mechanical (power and size) features of the proposed platform satisfy the restrictive CubeSat standard [9], while the electrical feasibility is demonstrated for 2U and 3U CubeSat missions. It should be noted that these results coincide with those recently reported by [2] for a TX2 module of the same manufacturer. the author finds that such module is likewise susceptible to be implemented in small satellites and CubeSats.…”

Section: Discussionsupporting

confidence: 92%

Conceptual low-cost on-board high performance computing in CubeSat nanosatellites for pattern recognition in Earth's remote sensing

Hernández-Gómez¹,

Yañez-Casas²,

Torres-Lara³

et al.

Kalpa Publications in Computing

View full text Add to dashboard Cite

Nowadays, remote sensing data taken from artificial satellites require high space com- munications bandwidths as well as high computational processing burdens due to the vertiginous development and specialisation of on-board payloads specifically designed for remote sensing purposes. Nevertheless, these factors become a severe problem when con- sidering nanosatellites, particularly those based in the CubeSat standard, due to the strong limitations that it imposes in volume, power and mass. Thus, the applications of remote sensing in this class of satellites, widely sought due to their affordable cost and easiness of construction and deployment, are very restricted due to their very limited on-board computer power, notwithstanding their Low Earth Orbits (LEO) which make them ideal for Earth’s remote sensing. In this work we present the feasibility of the integration of an NVIDIA GPU of low mass and power as the on-board computer for 1-3U CubeSats. From the remote sensing point of view, we present nine processing-intensive algorithms very commonly used for the processing of remote sensing data which can be executed on-board on this platform. In this sense, we present the performance of these algorithms on the proposed on-board computer with respect with a typical on-board computer for CubeSats (ARM Cortex-A57 MP Core Processor), showing that they have acceleration factors of average of 14.04× ∼14.72× in average. This study sets the precedent to perform satellite on-board high performance computing so to widen the remote sensing capabilities of CubeSats.

show abstract

Section: Discussionsupporting

confidence: 92%

Conceptual low-cost on-board high performance computing in CubeSat nanosatellites for pattern recognition in Earth's remote sensing

Hernández-Gómez¹,

Yañez-Casas²,

Torres-Lara³

et al.

Kalpa Publications in Computing

View full text Add to dashboard Cite

show abstract

“…Since the launch of the quickbird satellite with high resolution cameras and high capacity on-board processing capabilities, there has been an increase in research and development related to advanced on-board satellite image processing. Numerous works have investigated on-board capabilities of object detection [1], [11], image classification [12], image selection [13], denoising and artefact removal [14], and change detection [15], [16].…”

Section: A On-board Processing Of Satellite Imagesmentioning

confidence: 99%

Towards On-Board Panoptic Segmentation of Multispectral Satellite Images

Fernando¹,

Fookes²,

Gammulle³

et al. 2022

Preprint

View full text Add to dashboard Cite

With tremendous advancements in low-power embedded computing devices and remote sensing instruments, the traditional satellite image processing pipeline which includes an expensive data transfer step prior to processing data on the ground is being replaced by on-board processing of captured data. This paradigm shift enables critical and time-sensitive analytic intelligence to be acquired in a timely manner onboard the satellite itself. However, at present, the on-board processing of multi-spectral satellite images is limited to classification and segmentation tasks. Extending this processing to it's next logical level, in this paper we propose a lightweight pipeline for onboard panoptic segmentation of multi-spectral satellite images. Panoptic segmentation offers major economic and environmental insights, ranging from yield estimation from agricultural lands to intelligence for complex military applications. Nevertheless, the on-board intelligence extraction raises several challenges due to the loss of temporal observations and the need to generate predictions from a single image sample. To address this challenge, we propose a multimodal teacher network based on a cross-modality attention-based fusion strategy to improve the segmentation accuracy by exploiting data from multiple modes. We also propose an online knowledge distillation framework to transfer the knowledge learned by this multi-modal teacher network to a uni-modal student which receives only a single frame input, and is more appropriate for an on-board environment. We benchmark our approach against existing state-ofthe-art panoptic segmentation models using the PASTIS multispectral panoptic segmentation dataset considering an on-board processing setting. Our evaluations demonstrate a substantial 10.7%, 11.9% and 10.6% increase in Segmentation Quality (SQ), Recognition Quality (RQ), and Panoptic Quality (PQ) metrics compared to the existing state-of-the-art model when it is evaluated in an on-board processing setting.

show abstract

“…Large data transfers cause network bottlenecks and increases latency, thus increasing the system response time. For safetycritical industry applications, such as aerospace and defense, latency is a major concern; hence, optimizing it is an important research topic [1][2][3][4].…”

Section: Introductionmentioning

confidence: 99%

“…The number of DPU IPs (1)(2)(3)(4)(5)(6). One can incorporate multiple DPU IPs, allowing for parallel processing of the input samples.…”

mentioning

confidence: 99%

Benchmarking Deep Learning for On-Board Space Applications

Ziaja¹,

Bosowski²,

Myller³

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

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.

show abstract

Onboard Image Processing for Small Satellites

Cited by 24 publications

References 10 publications

Conceptual low-cost on-board high performance computing in CubeSat nanosatellites for pattern recognition in Earth's remote sensing

Conceptual low-cost on-board high performance computing in CubeSat nanosatellites for pattern recognition in Earth's remote sensing

Towards On-Board Panoptic Segmentation of Multispectral Satellite Images

Benchmarking Deep Learning for On-Board Space Applications

Contact Info

Product

Resources

About