Detecting, Tracking and Classifying Animals in Underwater Video

Edgington, D.R.; Cline, Danelle E.; Davis, D.; Kerkez, I.; Mariette, Jérôme

doi:10.1109/oceans.2006.306878

Cited by 48 publications

(36 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Firstly, the detection and tracking software described in Spampinato et al (2014b) is used to obtain the fish and mask images. Then the Grabcut algorithm (Rother et al 2004) is employed to segment fish from the background, similar to Edgington et al (2006), Cline and Edgington (2010)). Given prior information such as reference frame or pre-label foreground area, the graph cut solution gives each pixel a weight between foreground (source) and background (sink), and solves the segmentation problem with a minimum cost cut method to divide the source from the sink.…”

Section: Image Pre-processingmentioning

confidence: 99%

Hierarchical Classification System with Reject Option for Live Fish Recognition

Huang

2016

Fish4Knowledge: Collecting and Analyzing Massive Coral Reef Fish Video Data

View full text Add to dashboard Cite

This chapter presents a Balance-Guaranteed Optimized Tree with Reject option (BGOTR) for live fish recognition in a non-constrained environment. It recognizes the top 15 common species of fish and detects new species in an unrestricted natural environment recorded by underwater cameras. This system can assist ecological surveillance research, e.g., obtaining fish population statistics from the open sea. BGOTR is automatically constructed based on inter-class similarities. We apply a Gaussian Mixture Model (GMM) and Bayes rule as a reject option after hierarchical classification-we estimate the posterior probability of being a certain species and then filter out less confident decisions. The proposed BGOTR-based hierarchical classification method achieves significant improvements compared to state-of-theart techniques on a live fish image dataset of 24,150 manually labeled images from the south Taiwan sea. IntroductionLive fish recognition in the open sea has been investigated to help understand the marine ecosystem, which is vital for studying the marine environments and promoting commercial applications. This recognition task is fundamentally challenging because of its complex situation where the illumination changes frequently. Prior research is mainly restricted to constrained environments (fish in the tank or on a conveyor system) or dead fish, and these machine vision systems have only explored applications for a limited number of fish species. These methods perform worse when they deal with unconstrained fish in a real-world underwater environment, especially when the dataset is greatly imbalanced.

show abstract

Section: Image Pre-processingmentioning

confidence: 99%

Hierarchical Classification System with Reject Option for Live Fish Recognition

Huang

2016

Fish4Knowledge: Collecting and Analyzing Massive Coral Reef Fish Video Data

View full text Add to dashboard Cite

show abstract

“…There have been attempts to count marine species using stationary underwater cameras (Edgington et al 2006;Spampinato et al 2008). Background subtraction and shape detection (Williams et al 2006) has been used to count salmon.…”

Section: Vision-based Detection Of Marine Creaturesmentioning

confidence: 99%

Identifying sea scallops from benthic camera images

Kannappan

Walker

Trembanis

et al. 2014

Limnology & Ocean Methods

View full text Add to dashboard Cite

The article presents an algorithmic framework for the automated analysis of benthic imagery data. The data are collected by an autonomous underwater vehicle for the purpose of population assessment of epibenthic organisms, such as scallops. The architecture consists of three layers of processing: visual attention, graph-cut segmentation methods, and template matching. The visual attention layer filters the imagery input, focusing subsequent processing only on regions in the images that are likely to contain target objects. The segmentation layer prepares for subsequent template matching. Finally, template matching classifies filtered objects into targets and distractors. The significance of the proposed approach is in its modular nature and its ability to process imagery datasets of low resolution, brightness, and contrast.

show abstract

“…in real-life underwater systems, remains a challenge [5]. Fish detection and tracking is complicated by the variability of the underwater environment.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Fish Observation and Fish Category Database Construction

Shiau¹,

Lin²,

Lin³

et al. 2012

IJACSA

View full text Add to dashboard Cite

Abstract-This paper proposes a distributed real-time video stream system for underwater fish observation in the real world. The system, based on a three-tier architecture, includes capture devices unit, stream processor unit, and display devices unit. It supports variety of capture source devices, such as HDV, DV, WebCam, TV Card, Capture Card, and video compression formats, such as WMV, FLV/SWF, MJPEG, MPEG-2/4. The system has been demonstrated in Taiwan for long-term underwater fish observation. CCTV cameras and high-definition cameras are deployed on our system. Video compression methods and image processing methods are implemented to reduce network transfer flow and data storage space. Marine ecologists and end users can browse these real-time video streams via the Internet to understand the ecological changes immediately. These video data is preserved to form a resource base for marine ecologists. Based on the video data, fish detection is implemented. However, it is complicated in the unconstrained underwater environment, due to the water flow causes the water plants sway severely. In this paper, a bounding-surrounding boxes method is proposed to overcome the problem. It efficiently classifies moving fish as the foreground objects and the swaying water plants as the background objects. It enables to remove the irrelevant information (without fish) to reduce the massive amount of video data. Moreover, fish tracking is implemented to acquire multiple species of fish images with varied angles, sizes, shapes, and illumination to construct a fish category database.

show abstract

Detecting, Tracking and Classifying Animals in Underwater Video

Cited by 48 publications

References 13 publications

Hierarchical Classification System with Reject Option for Live Fish Recognition

Hierarchical Classification System with Reject Option for Live Fish Recognition

Identifying sea scallops from benthic camera images

Real-Time Fish Observation and Fish Category Database Construction

Contact Info

Product

Resources

About