Corpus ID: 204788831

A Deep Learning-based Framework for the Detection of Schools of Herring in Echograms

  title={A Deep Learning-based Framework for the Detection of Schools of Herring in Echograms},
  author={Alireza Rezvanifar and Tunai Porto Marques and Melissa Cote and Alexandra Branzan Albu and Alex Slonimer and Thomas M. Tolhurst and Kaan Ersahin and Todd Mudge and St{\'e}phane Gauthier},
Tracking the abundance of underwater species is crucial for understanding the effects of climate change on marine ecosystems. Biologists typically monitor underwater sites with echosounders and visualize data as 2D images (echograms); they interpret these data manually or semi-automatically, which is time-consuming and prone to inconsistencies. This paper proposes a deep learning framework for the automatic detection of schools of herring from echograms. Experiments demonstrated that our… Expand
Instance Segmentation-based Identification of Pelagic Species in Acoustic Backscatter Data
This paper proposes to detect pelagic species from echograms with a deep learning (DL) framework based on instance segmentation, allowing us to carefully study the acoustic properties of the targets and to address specific challenges such as close proximity between schools and varying size. Expand
Detecting Marine Species in Echograms via Traditional, Hybrid, and Deep Learning Frameworks
It is concluded that an end-to-end DL-based framework, that can be readily scaled to accommodate new species, is overall preferable to other learning approaches for echogram interpretation, even when only a limited number of annotated training samples is available. Expand
Size-invariant Detection of Marine Vessels From Visual Time Series
A detection approach that combines state-of-the-art object detectors and a novel Detector of Small Marine Vessels (DSMV) to identify boats of any size is proposed and Experimental results obtained show that the proposed approach comfortably outperforms five popular state of the art object detectors. Expand
EchoBERT: A Transformer-Based Approach for Behavior Detection in Echograms
EchoBERT - Echo Bidirectional Encoder Representation Transformer, a transformer-based approach for behavior detection in farmed Atlantic salmon using the spatiotemporal properties contained in echograms, has high potential for automatic behavior detection through unintrusive methods suitable for applications in aquaculture. Expand
Semi-supervised target classification in multi-frequency echosounder data
This work proposes a novel semi-supervised deep learning method leveraging a few annotated data samples together with vast amounts of unannotate data samples, all in a single model to optimize one shared convolutional neural network in an alternating manner. Expand


Discriminating fish species by an Echo sounder in a set-net using a CNN
Currently, the prediction of fish species and catches is based on the experience of fishermen. Echo sounders can support fisheries; however, they cannot identify fish species. A system for theExpand
Counting Fish in Sonar Images
The goal of this paper is to estimate the population of fishes in sonar images by building upon a recent local counts regression network and proposing two novel losses to regularize a modified $\ell_{1}$ loss with slack constraints. Expand
JellyMonitor: automated detection of jellyfish in sonar images using neural networks
JellyMonitor is an self-contained automated system that detects jellyfish blooms and reports their presence. It uses an embedded platform to analyse sonar imagery captured by a sonar imaging device.Expand
Artificial neural networks for fish-species identification
Techniques involving training and testing of artificial neural networks are applied for the automatic recognition and classification of digital echo recordings of schools in the Southwest Atlantic, proving the networks were able usually to recognize fish species based only on the intrinsic characteristics of the school. Expand
Classification of Southern Ocean krill and icefish echoes using random forests
Random forests were generated using acoustic and net sample data collected during surveys to classify krill, icefish, and mixed aggregations of weak scattering fish species with an overall estimated accuracy of 95%. Expand
Acoustic species identification in the Northwest Atlantic using digital image processing
Abstract Acoustic surveys for marine fish in coastal waters typically involve identification of several species groups. Incorrect classification can limit the usefulness of both distribution andExpand
Study on Echo Features and Classification Methods of Fish Species
  • Yue-quan Shang, Jian-long Li
  • Computer Science
  • 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP)
  • 2018
This work focuses on classification performance comparisons of three different features with four different methods, in order to find the optimal feature-method combination of twelve kinds of combinations. Expand
Rethinking the Inception Architecture for Computer Vision
This work is exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. Expand
A Joint Application of Fuzzy Logic Approximation and a Deep Learning Neural Network to Build Fish Concentration Maps Based on Sonar Data
This paper proposes an effective method for obtain topographic lake map with fish concentration based on the results of an intelligent sonar data processing using fuzzy logic and an algorithm for obtaining fish concentration maps. Expand
Multiple-frequency moored sonar for continuous observations of zooplankton and fish
Moored, internally-recording acoustic instruments can acquire continuous profiles of echoes throughout the water column, thus providing a low-cost method to study the behavior and abundance of fishExpand