Ocean Color Hyperspectral Remote Sensing With High Resolution and Low Latency—The HYPSO-1 CubeSat Mission

Grøtte, Mariusz Eivind; Birkeland, Roger; Honoré‐Livermore, Evelyn; Bakken, Sivert; Garrett, Joseph L.; Prentice, Elizabeth Frances; Sigernes, F.; Orlandić, Milica; Gravdahl, Jan Tommy; Johansen, Tor Arne

doi:10.1109/tgrs.2021.3080175

Cited by 45 publications

(56 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the recent developments in routine low-cost hyperspectral imaging there is the potential to overcome a number of these limitations in being able to acquire frequent, and high resolution, environmental data to improve models of urban biodiversity (Mozgeris et al 2018;Zhang et al 2020;Hartling et al 2021). Such remotely sensed environmental data may be captured using cubesats (Kimm et al 2020;Grøtte et al 2021), small and low-cost satellites, as well as airborne drones (Räsänen et al 2020;Dierssen et al 2021). Of particular use to monitoring highly dynamic and heterogeneous urban environments, such data can be collected at resolutions under 3 m in scale (Salgado-Hernanz et al 2021) and daily in time (Rhodes et al 2022).…”

Section: Future Prospectsmentioning

confidence: 99%

Constructing ecological indices for urban environments using species distribution models

Simons

CALDWELL

et al. 2022

Urban Ecosyst

View full text Add to dashboard Cite

In an increasingly urbanized world, there is a need to study urban areas as their own class of ecosystems as well as assess the impacts of anthropogenic impacts on biodiversity. However, collecting a sufficient number of species observations to estimate patterns of biodiversity in a city can be costly. Here we investigated the use of community science-based data on species occurrences, combined with species distribution models (SDMs), built using MaxEnt and remotely-sensed measures of the environment, to predict the distribution of a number of species across the urban environment of Los Angeles. By selecting species with the most accurate SDMs, and then summarizing these by class, we were able to produce two species richness models (SRMs) to predict biodiversity patterns for species in the class Aves and Magnoliopsida and how they respond to a variety of natural and anthropogenic environmental gradients.We found that species considered native to Los Angeles tend to have significantly more accurate SDMs than their non-native counterparts. For all species considered in this study we found environmental variables describing anthropogenic activities, such as housing density and alterations to land cover, tend to be more influential than natural factors, such as terrain and proximity to freshwater, in shaping SDMs. Using a random forest model we found our SRMs could account for approximately 54% and 62% of the predicted variation in species richness for species in the classes Aves and Magnoliopsida respectively. Using community science-based species occurrences, SRMs can be used to model patterns of urban biodiversity and assess the roles of environmental factors in shaping them.

show abstract

Section: Future Prospectsmentioning

confidence: 99%

Constructing ecological indices for urban environments using species distribution models

Simons

CALDWELL

et al. 2022

Urban Ecosyst

View full text Add to dashboard Cite

show abstract

“…So that we get 128 which contains the sum of the FFT values which is termed the area. The three areas which are the sum of the FFT values are used as features to distinguish between the images extracted from the asphalt road video recordings into good, moderate, lightly damaged, and heavily damaged criteria [28]. After the normalization process, the next process is to classify.…”

Section: Fig2 Noise Reduction Resultsmentioning

confidence: 99%

“…So with this method, it is expected to have a better level of accuracy. The mechanism of the detection process described in this paper is to collect training data with more than 500 examples of high-resolution images which will be processed starting with a resolution of 99 x 99 pixels [28]. Then the training data using the GPU and perform the detection testing process.…”

Section: Road Crack Detection Using Deep Convolutional Neural Networkmentioning

confidence: 99%

Broken Road Detection Methods Comparison: A Literature Survey

Yustiana

Somantri

Gustian

et al. 2022

IJEAT

View full text Add to dashboard Cite

Roads are infrastructure built to facilitate regional development. Good road conditions will certainly provide a sense of comfort for every vehicle that will pass through it. For that, care and attention to road conditions needs to be done. The occurrence of damage to the road will hinder the development process. Currently, detection of damaged roads is still done manually using human resource. It makes the detection process take quite a lot of time to determine how bad the damage is. So there needs a way to help improve time efficiency and accuracy in detecting damaged roads. One of them is by utilizing machine learning technology. In this paper, we will discuss what methodology can be use and their comparisons to be able to use appropriate and effective methodologies to detect cases of damaged roads

show abstract

“…Other challenges when aiming for on-board processing of Imaging spectroscopy are noisy data and no atmospheric correction available at level zero data as a well limited training set. Therefore, we should focus on testing different network structures on simulated and real data similar to the upcoming CHIME mission shortly [114]. Overall, on-board processing of HS imagery is the new area of study that will open many new possibilities in the remote sensing domain.…”

Section: Summary and Discussionmentioning

confidence: 99%

Deep Learning for On-board Processing of Imaging Spectroscopy: A Survey

Ghasemi¹,

Nieke²,

Celesti³

et al. 2023

Preprint

View full text Add to dashboard Cite

Modern hyperspectral imaging technologies generate enormous datasets that could potentially transmit a wealth of information, but such a resource presents numerous difficulties for data analysis and interpretation. Deep learning techniques undoubtedly provide a wide range of potential for solving both traditional imaging tasks and exciting new problems in the spatial-spectral domain. This is true in the primary application area of remote sensing, where hyperspectral technology originated and has made the majority of its progress, but it may be even more true in the vast array of now existing and developing application areas that make use of these imaging technologies. The current review advances on two fronts: on the one hand, it is directed at domain experts who desire an updated overview of how deep learning architectures might work in conjunction with hyperspectral acquisition techniques to address specific tasks in various application sectors. On the other hand, by providing them with a picture of how deep learning technologies are applied to hyperspectral data from (near)real-time perspective. The contributions of this review include the existence of these two points of view and the inclusion of opportunities and important problems associated with the development of future CHIME mission to be launched by European Space Agency (ESA).

show abstract

Ocean Color Hyperspectral Remote Sensing With High Resolution and Low Latency—The HYPSO-1 CubeSat Mission

Cited by 45 publications

References 100 publications

Constructing ecological indices for urban environments using species distribution models

Constructing ecological indices for urban environments using species distribution models

Broken Road Detection Methods Comparison: A Literature Survey

Deep Learning for On-board Processing of Imaging Spectroscopy: A Survey

Contact Info

Product

Resources

About