Plankton Image Classification Based on Multiple Segmentations

Hirata, Nina S. T.; Fernandez, Mariela Atausinchi; Lopes, Rubens M.

doi:10.1109/cvaui.2016.022

Cited by 8 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Underwater computer vision tasks include the detection, tracking, and recognition of fish [41], [101]- [103], other sea animals [42], [104], and plankton [105]; the classification of coral reefs [106] and the positioning of vehicles [107]. These tasks enable marine species discovery [108], coral preservation [109]- [111], ocean exploration with unmanned vehicles [107] and archaeological excavation [112], [113],.…”

Section: Computer Vision Tasksmentioning

confidence: 99%

Underwater image filtering: methods, datasets and evaluation

Li,

Mazzon,

Cavallaro

2020

Preprint

View full text Add to dashboard Cite

Underwater images are degraded by the selective attenuation of light that distorts colours and reduces contrast. The degradation extent depends on the water type, the distance between an object and the camera, and the depth under the water surface the object is at. Underwater image filtering aims to restore or to enhance the appearance of objects captured in an underwater image. Restoration methods compensate for the actual degradation, whereas enhancement methods improve either the perceived image quality or the performance of computer vision algorithms. The growing interest in underwater image filtering methods-including learning-based approaches used for both restoration and enhancement-and the associated challenges call for a comprehensive review of the state of the art. In this paper, we review the design principles of filtering methods and revisit the oceanology background that is fundamental to identify the degradation causes. We discuss image formation models and the results of restoration methods in various water types. Furthermore, we present taskdependent enhancement methods and categorise datasets for training neural networks and for method evaluation. Finally, we discuss evaluation strategies, including subjective tests and quality assessment measures. We complement this survey with a platform (https://puiqe.eecs.qmul.ac.uk/), which hosts state-of-the-art underwater filtering methods and facilitates comparisons.

show abstract

Section: Computer Vision Tasksmentioning

confidence: 99%

Underwater image filtering: methods, datasets and evaluation

Li,

Mazzon,

Cavallaro

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Bi et al [6] used a combination of MSER and Sauvola thresholding to extract large ROIs (larger than 5,000 pixels, or approximately 0.5 mm 2 ) and small ROIs, respectively, and obtained good segmentation results for high-turbidity underwater plankton images. Hirata et al [34] used dynamic thresholds, the algorithm of Otsu, and the watershed algorithm together to segment targets and demonstrated that images of different plankton require different segmentation methods. Despite the considerable research efforts on target segmentation for underwater plankton images, there is currently no universally applicable adaptive segmentation method.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Downscaling Segmentation for Noisy, Low-Contrast in Situ Underwater Plankton Images

Cheng

2020

IEEE Access

View full text Add to dashboard Cite

Finding and segmenting objects in noisy low-contrast in situ underwater plankton images is challenging because of the difficulty of separating potential plankton objects from the complex background and numerous and diverse other particles. In the present study, a dynamic downscaling model was developed to rapidly extract complete and clean regions of interest (ROIs) from images with highly variable content and quality. The original image was downscaled, and dynamic segmentation was performed in a scale pyramid space to ensure the integrity of weak targets based on local two-dimensional (2D) entropy parameters. Subsequently, a series of local thresholds and clustering gradients was examined iteratively for ROI selection. The performance of the local 2D entropy parameters relative to water turbidity (in a scattering medium and with high background noise) and image size was examined. To suppress the background and increase the sharpness of potential targets, a sharpness descriptor and gradient clustering were employed. The method was compared with the currently commonly used local threshold-based Sauvola segmentation using the same set of images. The results showed that the proposed method improves the ROI extraction accuracy and reduces oversegmentation for in situ underwater plankton images. It was concluded that the proposed method is a fast and robust segmentation technique and could facilitate the deployment of in situ plankton imaging systems for process-based research and routine plankton monitoring. INDEX TERMS dynamic downscaling, gradient clustering, in situ underwater plankton image, region of interest, scale pyramid space, segmentation.

show abstract

“…In the past decade, imaging has become an operational solution to acquire meso-and macrozooplankton data at fine spatial scale and/or high frequency (e.g. Romagnan et al, 2015) by the development of numerous dedicated instruments (Davis et al, 2005;Benfield et al, 2007;Gorsky et al, 2010;Picheral et al 2010;Cowen and Guigand, 2008), and dedicated analytical methods (Failletaz et al, 2016;Hirata et al, 2016;Gonzalez et al, 2016). Compared to other methods to analyse zooplankton such as microscopic examination or optical methods (e.g.…”

Section: Introductionmentioning

confidence: 99%

The ZooCAM, a new in-flow imaging system for fast onboard counting, sizing and classification of fish eggs and metazooplankton

Colas

Tardivel

Perchoc

et al. 2018

Progress in Oceanography

View full text Add to dashboard Cite

In this paper we present the ZooCAM, a system designed to digitize and analyse on board large volume samples of preserved and living metazooplankton (i.e. multicellular zooplankton) and fish eggs > 300 µm ESD. The ZooCAM has been specifically designed to overcome the difficulties to analyse zooplankton and fish eggs in the framework of the PELGAS survey, and provide high frequency data. The ZooCAM fish eggs counts were comparable to those done with a dissecting microscope. The ZooCAM enabled the accurate prediction and fast on board validation of staged anchovy and sardine eggs in almost real time after collection. A comparison with the ZooScan, on a more complex zooplanktonic community, provided encouraging results on the agreement between the 2 instruments. ZooCAM and ZooScan enabled the identification of similar communities and produced similar total zooplankton abundances, size distributions, and size spectra slopes, when tested on the same samples. However these results need to be further refined due to the small number of samples used to compare the two instruments. The main ZooCAM drawback resides in a slight but sensible underestimation of abundances and sizes, and therefore individual and community biovolumes. The ZooCAM have been successfully deployed over 4 years, on numerous surveys without suffering any major failure. When used in line with the CUFES it provided high resolution maps of staged fish eggs and zooplanktonic functional groups. Hence the ZooCAM is an appropriate tool for the development of on board, high frequency, high spatial coverage zooplanktonic and ecosystemic studies. Highlights ► The ZooCAM is a new inflow imaging system for fast onboard counting, sizing and classification of fish eggs and metazooplankton. ► The ZooCAM has been used on-board for 4 years without suffering any failure and enabled the analysis ∼10,000 samples so far (CUFES, WP2 and other plankton nets). ► The ZooCAM provided staged fish eggs counts that were equivalent to those traditionally done by microscopic examination. These data were used for 2016 and 2017 anchovy and sardine stocks estimations in the Bay of Biscay (DEPM method) ► The ZooCAM enable the real time, on-board, imaging of complex communities zooplankton samples and provide results comparable to on-land, well established counterpart, the ZooScan. ► The ZooCAM is a valuable tool to improve zooplanktonic studies, in term of spatial coverage and temporal frequency, to perform ecosystemic studies.

show abstract

Plankton Image Classification Based on Multiple Segmentations

Cited by 8 publications

References 10 publications

Underwater image filtering: methods, datasets and evaluation

Underwater image filtering: methods, datasets and evaluation

Dynamic Downscaling Segmentation for Noisy, Low-Contrast in Situ Underwater Plankton Images

The ZooCAM, a new in-flow imaging system for fast onboard counting, sizing and classification of fish eggs and metazooplankton

Contact Info

Product

Resources

About