Underwater image enhancement: a comprehensive review, recent trends, challenges and applications

Raveendran, Smitha; Patil, Mukesh D.; Birajdar, Gajanan K.

doi:10.1007/s10462-021-10025-z

Cited by 141 publications

(42 citation statements)

References 129 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering that single convolution will cause the number of channels to plummet and lose more information, the method of gradually reducing the number of feature channels by using three convolutions is used, while ref. [11,12] fully consider that the deep high-dimensional feature output retains more target and location information and ignores the target detail information, while the lowdimensional features focus more on local and boundary information. Therefore, the number of channels is reduced to 64 for each layer of VGG16Net backbone base network after three convolution operations and then combined with the saliency map output of the higher layer and then through convolution, deconvolution, and convolution operations to produce the saliency map of this layer.…”

Section: Methodsmentioning

confidence: 99%

“…Reference [11] proposes a cyclic CNN with a prediction map guided by a previous cyclic step. Reference [12] used a dropout technique to learn deep uncertain convolutional features in the network to enhance its generalization ability. However, since these methods only employ features extracted at the deeper layers of the CNN, they tend to miss details in salient objects captured mainly at the shallow layers.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Application and Analysis of RGB-D Salient Object Detection in Photographic Camera Vision Processing

Fu¹

2022

Journal of Sensors

View full text Add to dashboard Cite

To identify the most visually salient regions in a set of paired RGB and depth maps, in this paper, we propose a multimodal feature fusion supervised RGB-D image saliency detection network, which learns RGB and depth data by two independent streams separately, uses a dual-stream side-supervision module to obtain saliency maps based on RGB and depth features for each layer of the network separately, and then uses a multimodal feature fusion module to fuse the latter 3 layers of RGB and depth high-dimensional information to generate high-level significant prediction results. Experiments on three publicly available datasets show that the proposed network outperforms the current mainstream RGB-D saliency detection models with strong robustness due to the use of a dual-stream side-surveillance module and a multimodal feature fusion module. We use the proposed RGB-D SOD model for background defocusing in realistic scenes and achieve excellent visual results.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Application and Analysis of RGB-D Salient Object Detection in Photographic Camera Vision Processing

Fu¹

2022

Journal of Sensors

View full text Add to dashboard Cite

show abstract

“…Moghimi and Mohanna 1 reviewed the field of the UIE as hardware and software tools used for acquisition and enhancement, existing UIE methods, discussion on different parameters for real-time enhancement and image quality evaluation metrics. Raveendran et al 12 reviewed existing UIE methods, quality assessment metrics along with the applications and future directions. Hu et al 13 reviewed existing UIFM, UD, and evaluation metrics and discussed issues and challenges in enhancing underwater images.…”

Section: Motivationmentioning

confidence: 99%

Systematic review and analysis on underwater image enhancement methods, datasets, and evaluation metrics

Verma

Kumar

2022

J. Electron. Imag.

View full text Add to dashboard Cite

valuable resources that exist beneath Earth's surface. Underwater exploration requires enhanced images that are obtained using enhancement methods. So, it is important that underwater image enhancement (UIE) methods work well in terms of performance and accuracy. As a result, research in UIE has increased in the past few years. An extensive survey is conducted on existing UIE methods along with their broad classification, underwater datasets, and evaluation metrics, respectively. The experimental analysis is conducted to compare the existing UIE methods in terms of qualitative and quantitative evaluation metrics. The real-world applications and future scope of existing enhancement methods are highlighted and discussed.

show abstract

“…Underwater camera systems must deal with the challenges presented by the surrounding medium (i.e., water). The main technical challenge of underwater imaging lies in the limited viewing distances of objects, stemming from the effects of turbidity and light attenuation, which limit the coverage of a habitat compared to filming in air [11,12]. On land, medium sized animals such as birds can be viewed at distances of 100s of meters [13] while even in the clear waters of coral reefs, similar-sized fish can only be viewed from few 10s of meters at best [14,15].…”

Section: Underwater Computational Photographymentioning

confidence: 99%

Analysis of Larval Fish Feeding Behavior under Naturalistic Conditions

Bar

Levy

Avidan

et al. 2022

Preprint

View full text Add to dashboard Cite

Modern research efforts concerned with animal behavior rely heavily on image and video analysis. While such data are now quick to obtain, extracting and analyzing complex behaviors under naturalistic conditions is still a major challenge, specifically when the behavior of interest is sporadic and rare. In this study, we present an end-to-end system for capturing, detecting and analyzing larval fish feeding behavior in unconstrained naturalistic environments. We first constructed a specialized system for imaging these tiny, fast-moving creatures and deployed it in large aquaculture rearing pools. We then designed an analysis pipeline using several action classification backbones, and compare their performance. A natural feature of the data was the extremely low prevalence of feeding events, leading to low sample sizes and highly imbalanced datasets despite extensive annotation efforts. Nevertheless, our pipeline successfully detected and classified the sparsely-occurring feeding behavior of fish larvae in a curated experimental setting from videos featuring multiple animals. We introduce three new annotated datasets of underwater videography, in a curated and an uncurated setting. As challenges related to data imbalance and expert's annotation are common to the analysis of animal behavior under naturalistic conditions, we believe our findings can contribute to the growing field of computer vision for the study and understanding of animal behavior.

show abstract

Underwater image enhancement: a comprehensive review, recent trends, challenges and applications

Cited by 141 publications

References 129 publications

Application and Analysis of RGB-D Salient Object Detection in Photographic Camera Vision Processing

Application and Analysis of RGB-D Salient Object Detection in Photographic Camera Vision Processing

Systematic review and analysis on underwater image enhancement methods, datasets, and evaluation metrics

Analysis of Larval Fish Feeding Behavior under Naturalistic Conditions

Contact Info

Product

Resources

About