A computer vision approach for monitoring the spatial and temporal shrimp distribution at the LoVe observatory

Osterloff, Jonas; Nilssen, Ingunn; Nattkemper, Tim Wilhelm

doi:10.1016/j.mio.2016.03.002

Cited by 26 publications

(18 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This technique was tested at a high frequency (i.e., 30 min) and over a long–lasting period of time (i.e., one year). We found that it is a reliable benchmark for scaling to other still video sources, either in deeper continental margin areas–such as in the LoVE observatory 35 and the NEPTUNE network 36 )–or on board of mobile platforms–such as ARGO floats, ROVs, AUVs 33 , 37 , 38 and crawlers 39 , 40 if the hypothesis that the not relevant information (e.g. background, bio-fouling over the camera) changes more slowly, along the time, than the relevant subjects (e.g.…”

Section: Discussionmentioning

confidence: 85%

Tracking Fish Abundance by Underwater Image Recognition

Marini

Fanelli

Sbragaglia

et al. 2018

Sci Rep

133

101

View full text Add to dashboard Cite

Marine cabled video-observatories allow the non-destructive sampling of species at frequencies and durations that have never been attained before. Nevertheless, the lack of appropriate methods to automatically process video imagery limits this technology for the purposes of ecosystem monitoring. Automation is a prerequisite to deal with the huge quantities of video footage captured by cameras, which can then transform these devices into true autonomous sensors. In this study, we have developed a novel methodology that is based on genetic programming for content-based image analysis. Our aim was to capture the temporal dynamics of fish abundance. We processed more than 20,000 images that were acquired in a challenging real-world coastal scenario at the OBSEA-EMSO testing-site. The images were collected at 30-min. frequency, continuously for two years, over day and night. The highly variable environmental conditions allowed us to test the effectiveness of our approach under changing light radiation, water turbidity, background confusion, and bio-fouling growth on the camera housing. The automated recognition results were highly correlated with the manual counts and they were highly reliable when used to track fish variations at different hourly, daily, and monthly time scales. In addition, our methodology could be easily transferred to other cabled video-observatories.

show abstract

Section: Discussionmentioning

confidence: 85%

Tracking Fish Abundance by Underwater Image Recognition

Marini

Fanelli

Sbragaglia

et al. 2018

Sci Rep

133

101

View full text Add to dashboard Cite

show abstract

“…The image patches are anyhow by orders of magnitude smaller than the entire images in FIA and a subset of them can be displayed on a screen in a rapid serial visual presentation (i.e., showing one patch at a time and with dynamic updates like in a slideshow) or in a rich gridded visual presentation (i.e., several patches in parallel and static) (see Figure 2 on the right). SPC has seen rare applications in marine image annotation yet [except in case of posterior inspections of annotations obtained by a computer vision system (Schoening et al, 2012b;Osterloff et al, 2016;Schoening et al, under review)] since it needs a sophisticated data base model to represent the annotations ROIs and classifications. Annotation projects carried out with the so called random points approach can be considered a SPC task, since the observer inspects a given number of randomly set points in an image and classifies the image content at those image points to one category.…”

Section: A Taxonomy For Labeling Tasksmentioning

confidence: 99%

“…This call has been answered by promising achievements in algorithmic annotation, such as automated classification of seabeds and habitats (Pican et al, 1998;Pizarro et al, 2009), coral segmentation (Purser et al, 2009;Tusa et al, 2014), and classification (Beijbom et al, 2012(Beijbom et al, , 2015, detection, and spatial analysis of shrimp populations (Purser et al, 2013;Osterloff et al, 2016), detection and classification of mega fauna in benthic images (Schoening et al, 2012b;Schoening et al, under review) or the computation of image footprint size with automatically detected laser points (Schoening et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

RecoMIA—Recommendations for Marine Image Annotation: Lessons Learned and Future Directions

2016

Self Cite

View full text Add to dashboard Cite

Marine imaging is transforming into a sensor technology applied for high throughput sampling. In the context of habitat mapping, imaging establishes thereby an important bridge technology regarding the spatial resolution and information content between physical sampling gear (e.g., box corer, multi corer) on the one end and hydro-acoustic sensors on the other end of the spectrum of sampling methods. In contrast to other scientific imaging domains, such as digital pathology, there are no protocols and reports available that guide users (often referred to as observers) in the non-trivial process of assigning semantic categories to whole images, regions, or objects of interest (OOI), which is referred to as annotation. These protocols are crucial to facilitate image analysis as a robust scientific method. In this article we will review the past observations in manual Marine Image Annotations (MIA) and provide (a) a guideline for collecting manual annotations, (b) definitions for annotation quality, and (c) a statistical framework to analyze the performance of human expert annotations and to compare those to computational approaches.

show abstract

“…No detailed analysis has been performed on the community composition or the reef mounds in the Hola trough, but still images show that the dominant reef-building coral species is Lophelia pertusa (Osterloff et al 2016b). This scleractinian coral is the dominant reef builder in European waters (Roberts et al 2006), although Madrepora oculata can also occur as smaller parts of the reefs on the Norwegian shelf (Järnegren and Kutti 2014).…”

Section: Location Of the Love Observatorymentioning

confidence: 99%

Cabled ocean observatory data reveal food supply mechanisms to a cold-water coral reef

Engeland

Godø

Duineveld

et al. 2019

Progress in Oceanography

View full text Add to dashboard Cite

We investigated food supply mechanisms to a cold-water coral (CWC) reef at 260 m depth on the Norwegian continental shelf using data from a cabled ocean observatory equipped with Acoustic Doppler Current Profilers (ADCPs), an echosounder, and sensors for chlorophyll, turbidity and hydrography in the benthic boundary layer (BBL). Tidal currents of up to tens of cm s-1 dominated BBL hydrodynamics while residual currents were weak (~10 cm s-1), emphasizing a supply and high retention of locally produced phytodetritus within the trough. A direct connection between the reefs and surface organic matter (OM) was established by turbulent mixing and passive particle settling, but relative contributions varied seasonally. Fresh OM from a spring-bloom was quickly mixed into the BBL, but temperature stratification in summer reduced the surface-to-bottom connectivity and reduced the phytodetritus supply. A qualitative comparison among acoustic backscatter in the ADCPs (600 kHz, 190 kHz) and echosounder (70 kHz) suggests that vertically migrating zooplankton may present an alternative food source in summer. Nocturnal feeding by zooplankton in the upper water column sustains downward OM transport independent from water column mixing and may dominate as food supply pathway over sedimentation of the phytodetritus, especially during stratified conditions. In addition, it could present a concentrating mechanism for nutritional components as compensation for the deteriorating phytodetritus quality. Overall, the observed patterns suggest seasonal changes in the food supply pathways to the reef communities. The moderating role of temperature stratification in phytodetritus transport suggests stronger dependence of the cold-water corals on zooplankton for their dietary requirements with increased stratification under future climate scenarios. This study demonstrates the added value of permanent ocean observatories to research based on dedicated campaigns and regular monitoring.

show abstract

A computer vision approach for monitoring the spatial and temporal shrimp distribution at the LoVe observatory

Cited by 26 publications

References 19 publications

Tracking Fish Abundance by Underwater Image Recognition

Tracking Fish Abundance by Underwater Image Recognition

RecoMIA—Recommendations for Marine Image Annotation: Lessons Learned and Future Directions

Cabled ocean observatory data reveal food supply mechanisms to a cold-water coral reef

Contact Info

Product

Resources

About