Fish detection and species classification in underwater environments using deep learning with temporal information

Jalal, Ahsan; Ahmad, Sahrim; Mian, Ajmal; Shortis, Mark R.; Shafait, Faisal

doi:10.1016/j.ecoinf.2020.101088

Cited by 158 publications

(73 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Object detection and object tracking automate data collection, including gathering information about the type, location, and movement of objects of interest. Object detection algorithms can count and identify species of interest in underwater video footage (Christin et al, 2019) and have been applied to detect seals (Salberg, 2015), identify whale hotspots (Guirado et al, 2019), monitor fish populations (Ditria, Lopez-Marcano, et al, 2020;Jalal et al, 2020;Marini et al, 2018;Salman et al, 2016;Villon et al, 2016Villon et al, , 2018Villon et al, , 2020Xiu et al, 2015), and quantify floating debris on the ocean surface (Watanabe et al, 2019). On the other hand, object tracking can locate and output the movement direction and speed of objects between video frames.…”

Section: Introductionmentioning

confidence: 99%

Automatic detection of fish and tracking of movement for ecology

Lopez‐Marcano

Jinks

Buelow

et al. 2021

Ecology and Evolution

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Automatic detection of fish and tracking of movement for ecology

Lopez‐Marcano

Jinks

Buelow

et al. 2021

Ecology and Evolution

View full text Add to dashboard Cite

show abstract

“…To further evaluate the quality of the representations learned by A-LCFCN, we also test it on the FishLoc dataset for the counting and localization tasks. These tasks are essential for marine biologists, which have to assess and track changes in large fish populations 13,24 . Thus, having a model that automates the localization of these fishes can greatly reduce the cost of tracking large populations, thus helping marine scientist to do efficient monitoring.…”

Section: Counting and Localization Resultsmentioning

confidence: 99%

Weakly Supervised Underwater Fish Segmentation Using Affinity LCFCN

Saleh

Laradji

Rodríguez

et al. 2021

Preprint

View full text Add to dashboard Cite

Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time consuming and labour intensive. Others rely on fully-supervised segmentation models to automatically acquire these measurements but require collecting per-pixel labels which are also time consuming. It can take up to 2 minutes per fish to acquire accurate segmentation labels. To address this problem, we propose a segmentation model that can efficiently train on images labeled with point-level supervision, where each fish is annotated with a single click. This labeling scheme takes an average of only 1 second per fish. Our model uses a fully convolutional neural network with one branch that outputs per-pixel scores and another that outputs an affinity matrix. These two outputs are aggregated using a random walk to get the final, refined per-pixel output. The whole model is trained end-to-end using the LCFCN loss and thus we call our method Affinity-LCFCN (A-LCFCN). We conduct experiments on the DeepFish dataset, which contains several fish habitats from north-eastern Australia. The results show that A-LCFCN outperforms a fully-supervised segmentation model when the annotation budget is fixed. They also show that A-LCFCN achieves better segmentation results than LCFCN and a standard baseline.

show abstract

“…To further evaluate the quality of the representations learned by A-LCFCN, we also test it on the FishLoc dataset for the counting and localization tasks. These tasks are essential for marine biologists, which have to assess and track changes in large fish populations 61 , 62 . Thus, having a model that automates the localization of these fishes can greatly reduce the cost of tracking large populations, thus helping marine scientist to do efficient monitoring.…”

Section: Methodsmentioning

confidence: 99%

Weakly supervised underwater fish segmentation using affinity LCFCN

Laradji

Saleh

Rodríguez³

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Estimating fish body measurements like length, width, and mass has received considerable research due to its potential in boosting productivity in marine and aquaculture applications. Some methods are based on manual collection of these measurements using tools like a ruler which is time consuming and labour intensive. Others rely on fully-supervised segmentation models to automatically acquire these measurements but require collecting per-pixel labels which are also time consuming. It can take up to 2 minutes per fish to acquire accurate segmentation labels. To address this problem, we propose a segmentation model that can efficiently train on images labeled with point-level supervision, where each fish is annotated with a single click. This labeling scheme takes an average of only 1 second per fish. Our model uses a fully convolutional neural network with one branch that outputs per-pixel scores and another that outputs an affinity matrix. These two outputs are aggregated using a random walk to get the final, refined per-pixel output. The whole model is trained end-to-end using the localization-based counting fully convolutional neural network (LCFCN) loss and thus we call our method Affinity-LCFCN (A-LCFCN). We conduct experiments on the DeepFish dataset, which contains several fish habitats from north-eastern Australia. The results show that A-LCFCN outperforms a fully-supervised segmentation model when the annotation budget is fixed. They also show that A-LCFCN achieves better segmentation results than LCFCN and a standard baseline.

show abstract

Fish detection and species classification in underwater environments using deep learning with temporal information

Cited by 158 publications

References 13 publications

Automatic detection of fish and tracking of movement for ecology

Automatic detection of fish and tracking of movement for ecology

Weakly Supervised Underwater Fish Segmentation Using Affinity LCFCN

Weakly supervised underwater fish segmentation using affinity LCFCN

Contact Info

Product

Resources

About