Implications of sensor design for coral reef detection: Upscaling ground hyperspectral imagery in spatial and spectral scales

Caras, Tamir; Hedley, John D.; Karnieli, Arnon

doi:10.1016/j.jag.2017.07.009

Cited by 12 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, there is a pressing need for a rapid and scalable method of assessing coral reefs. It is within this context that close-range [6], or underwater hyperspectral imaging [7] has been developed and deployed for surveying the habitat structure of coral reefs. Hyperspectral imaging has proven to be a powerful remote sensing technique, with many different applications in agriculture [8], forestry [9], urban planning [10], and ecology [11].…”

Section: Discussionmentioning

confidence: 99%

“…Instead, for each transect, we selected and marked regions to generate the largest number of unique class labels. This was achieved through marking only a few ROIs (3)(4)(5)(6)(7)(8)(9)(10) for the most commonly occurring class labels (e.g., "Sediment", "Turf algae", etc.) and then focusing on finding new classes to annotate.…”

Section: Annotation Strategy: Deliberate Bias To Reduce Human Effortmentioning

confidence: 99%

See 1 more Smart Citation

A Trillion Coral Reef Colors: Deeply Annotated Underwater Hyperspectral Images for Automated Classification and Habitat Mapping

Rashid

Chennu

2020

Data

View full text Add to dashboard Cite

This paper describes a large dataset of underwater hyperspectral imagery that can be used by researchers in the domains of computer vision, machine learning, remote sensing, and coral reef ecology. We present the details of underwater data acquisition, processing and curation to create this large dataset of coral reef imagery annotated for habitat mapping. A diver-operated hyperspectral imaging system (HyperDiver) was used to survey 147 transects at 8 coral reef sites around the Caribbean island of Curaçao. The underwater proximal sensing approach produced fine-scale images of the seafloor, with more than 2.2 billion points of detailed optical spectra. Of these, more than 10 million data points have been annotated for habitat descriptors or taxonomic identity with a total of 47 class labels up to genus- and species-levels. In addition to HyperDiver survey data, we also include images and annotations from traditional (color photo) quadrat surveys conducted along 23 of the 147 transects, which enables comparative reef description between two types of reef survey methods. This dataset promises benefits for efforts in classification algorithms, hyperspectral image segmentation and automated habitat mapping.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Annotation Strategy: Deliberate Bias To Reduce Human Effortmentioning

confidence: 99%

A Trillion Coral Reef Colors: Deeply Annotated Underwater Hyperspectral Images for Automated Classification and Habitat Mapping

Rashid

Chennu

2020

Data

View full text Add to dashboard Cite

show abstract

“…For example, LiDAR has proven highly accurate for mapping reef bathymetry. Hyperspectral sensors are currently underutilized, with the potential to provide much more detailed benthic habitat information than traditional multispectral sensors (Caras et al, 2017). Lastly, the majority of the passive remote sensing approaches can be hindered by water turbidity, and other systems such as acoustic sensors can surpass such approaches in certain scenarios, particularly in water more than 50 m in depth (Table S1).…”

Section: Scaling Up Reef Restoration With Remote Sensingmentioning

confidence: 99%

“…As 50% of reef outplanting sites cover an area of just 10-1,000 m 2 (Fabian et al, 2013), spatial resolution finer than most satellite-based sensors is needed to provide detailed site characteristics relevant for outplanting. Airborne hyperspectral remote sensing (also known as imaging spectroscopy) is at the technological forefront of optical sensing, providing the most detailed benthic habitat maps (Goodman et al, 2013;Hedley, 2013;Caras et al, 2017) successfully classifying benthos into bottom-types such as fleshy algae, turf algae, seagrass, coral, and coral rubble (Hochberg et al, 2003). By accessing data over time, changes in benthic maps can inform changes in land use, coastal development and reef connectivity (Mumby et al, 2004;Raitsos et al, 2017) (Table 1; Table S2, Figure 1; Figure S1).…”

Section: Selecting Suitable Technology For Coral Restoration Criteriamentioning

confidence: 99%

Scaling Up Coral Reef Restoration Using Remote Sensing Technology

Foo

Asner

2019

Front. Mar. Sci.

View full text Add to dashboard Cite

Coral reefs face an uncertain future and may not recover naturally from anthropogenic climate change. Coral restoration is needed to rehabilitate degraded reefs and to sustain biodiversity. There is a need for baseline data on global reef distribution, composition, and condition to provide targets for conservation and restoration. Remote sensing can address this issue and is currently underutilized in reef research and restoration. This synthesis integrates current capabilities of remote sensing with key reef restoration criteria, to facilitate coral restoration success. Research into the development of a spectral database for corals, linking habitat type and extent with predator abundance, and identification of species-specific acoustic signatures are needed to advance the use of remote sensing in reef restoration design and monitoring. Reciprocally, reef restoration efforts should innovate at ecosystem, regional, and global levels using remote sensing, to preserve as much of the coral reef biome as possible with continued ocean-climate change.

show abstract

“…Therefore, mean absorptance of the coral and algae mapped classes using WorldView-2 were close to in-situ values (Table 4- 8). A recent study that evaluated the spatial and spectral resolution of hyperspectral imagery for improving coral reef classification concluded that high spatial resolution (< 1 m) can compensate for lower spectral resolution (Caras et al, 2017). However, in this case, the Planet Dove and WorldView-2 sensors have coarser spatial resolution, thus the issues with the sensors' SNR and waveband width were more apparent.…”

Section: Future Applicationsmentioning

confidence: 99%

Mapping coral reef primary production and calcification from a light-use efficiency model with in-situ and remotely sensed data

Perez¹

View full text Add to dashboard Cite

Monitoring coral reef primary and carbonate production is critical to understand changes in ecosystem function over time and between different reef systems. This is essential with increasing anthropogenic stresses from coastal development, overfishing, pollution, and ocean warming and acidification. However, monitoring these processes at spatial scales relevant to on ground management showing details within individual reefs (100's m 2) and across reef systems (1000's km 2) has been difficult due to lack of biogeochemical models that incorporate physical and biological factors for scaling with remote sensing. There is potential to adopt existing models that use absorbed photosynthetically active radiation (APAR) for estimating productivity, such as the light-use efficiency (LUE) model, which has been successfully applied in terrestrial and open ocean systems. The aim of this thesis was to develop methods for applying the LUE model to coral reef environments with field and remotely sensed data collected at multiple spatial scales. The objectives were to: (1) assess variability in community net productivity and calcification rates for algae dominated reef flats and a seagrass community, (2) assess variability of photosynthetic and calcification efficiencies based on APAR for multiple functional-groups: coral, non-calcareous algae, calcareous algae, sediment, and seagrass, and (3) apply the LUE model to map gross productivity and net calcification using multi-spectral imagery. This thesis focused on two reef systems: Heron Reef, southern Great Barrier Reef and Saipan Lagoon, Commonwealth of the Northern Mariana Islands, which have multi-spectral imagery available for model application. The first chapter of this thesis provided a brief overview of productivity and calcification processes in coral reef systems, and current limitations in monitoring these processes with remote sensing. In Chapter 2 (Obj. 1), field measurements of community net productivity and calcification are reported for algae dominated reef flats and a seagrass community. This study provided the first baseline metabolism and calcification rates for Saipan Lagoon. Both the reef flats and seagrass community had high net productivity that was associated with the dominant presence of algae at all sites. In Chapter 3 (Obj. 2), the theoretical basis for the light-use efficiency model is introduced, and how the model used the relationship between irradiance and photosynthesis to estimate productivity. The relationship between productivity, calcification, absorptance, and irradiance I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis and have sought permission from coauthors for any jointly authored works included in the thesis.

show abstract

Implications of sensor design for coral reef detection: Upscaling ground hyperspectral imagery in spatial and spectral scales

Cited by 12 publications

References 34 publications

A Trillion Coral Reef Colors: Deeply Annotated Underwater Hyperspectral Images for Automated Classification and Habitat Mapping

A Trillion Coral Reef Colors: Deeply Annotated Underwater Hyperspectral Images for Automated Classification and Habitat Mapping

Scaling Up Coral Reef Restoration Using Remote Sensing Technology

Mapping coral reef primary production and calcification from a light-use efficiency model with in-situ and remotely sensed data

Contact Info

Product

Resources

About