Fast Underwater Image Enhancement for Improved Visual Perception

Islam, Md Jahidul; Xia, Youya; Sattar, Junaed

doi:10.48550/arxiv.1903.09766

Cited by 4 publications

(26 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned in the previous section, this is mostly due to the lack of large-scale datasets (containing LR-HR pairs of images) that capture the distribution of the unique distortions prevalent in underwater imagery. The existing datasets are only suitable for underwater object detection [4] and image enhancement [5] tasks, as their image resolution is typically limited to 256 × 256, and they often contain synthetic images [12]. Consequently, the performance and applicability of existing and novel SISR models for underwater imagery have not been explored in depth.…”

Section: B Sisr For Underwater Imagerymentioning

confidence: 99%

“…We consider two standard metrics [61], [5] named Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) in order to quantitatively compare the SISR models' performances. The PSNR approximates the reconstruction quality of a generated image x compared to its ground truth y based on their Mean Squared Error (MSE) as follows:…”

Section: B Quantitative Evaluationmentioning

confidence: 99%

“…Autonomous Underwater Vehicles (AUVs) and Remotely Operated Vehicles (ROVs) are widely used in these applications, where they harness the synthesized images for visual attention modeling to make navigation decisions such as 'where to look or go next', 'which snapshots should be recorded', etc. However, despite often using high-end cameras, underwater images are often greatly affected [5] by poor visibility, absorption, and scattering. Consequently, Fig.…”

Section: Introductionmentioning

confidence: 99%

“…For instance, they tend to have a dominating green or blue hue because the red wavelengths get absorbed in deep water [12]. Other factors such as the lighting variations in different depths, amount of particles in the water, and scattering cause irregular non-linear distortions which result in low-contrast and blurry images [5]. Consequently, the off-the-shelf SISR models trained on arbitrary images fail to generate realistic higher resolution underwater images.…”

Section: Introductionmentioning

confidence: 99%

“…Secondly, the lack of large-scale underwater dataset restricts extensive research attempts for the training and performance evaluation of SISR models on underwater images. Because of the high costs and difficulties associated with acquiring real-world underwater data, the existing datasets (that were originally proposed for training object detection and image enhancement models) often contain synthetic images [12] and/or their resolution are typically limited to 256 × 256 [5]. Due to these challenges, designing SISR models for underwater imagery and investigating their applicability in real-world underwater robotic applications have not been explored in-depth in the literature.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Underwater Image Super-Resolution using Deep Residual Multipliers

Islam¹,

Enan²,

Luo³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

We present a deep residual network-based generative model for single image super-resolution (SISR) of underwater imagery for use by autonomous underwater robots. We also provide an adversarial training pipeline for learning SISR from paired data. In order to supervise the training, we formulate an objective function that evaluates the perceptual quality of an image based on its global content, color, and local style information. Additionally, we present USR-248, a large-scale dataset of three sets of underwater images of 'high' (640 × 480) and 'low' (80 × 60, 160 × 120, and 320 × 240) resolution. USR-248 contains over 7K paired instances in each set of data for supervised training of 2×, 4×, or 8× SISR models. Furthermore, we validate the effectiveness of our proposed model through qualitative and quantitative experiments and compare the results with several state-of-the-art models' performances.We also analyze its practical feasibility for applications such as scene understanding and attention modeling in noisy visual conditions.

show abstract

Section: B Sisr For Underwater Imagerymentioning

confidence: 99%

Section: B Quantitative Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Underwater Image Super-Resolution using Deep Residual Multipliers

Islam¹,

Enan²,

Luo³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Simultaneous Enhancement and Super-Resolution of Underwater Imagery for Improved Visual Perception

Islam

Luo

Sattar

2020

Robotics: Science and Systems XVI

View full text Add to dashboard Cite

In this paper, we introduce and tackle the simultaneous enhancement and super-resolution (SESR) problem for underwater robot vision and provide an efficient solution for near real-time applications. We present Deep SESR, a residualin-residual network-based generative model that can learn to restore perceptual image qualities at 2×, 3×, or 4× higher spatial resolution. We supervise its training by formulating a multi-modal objective function that addresses the chrominancespecific underwater color degradation, lack of image sharpness, and loss in high-level feature representation. It is also supervised to learn salient foreground regions in the image, which in turn guides the network to learn global contrast enhancement. We design an end-to-end training pipeline to jointly learn the saliency prediction and SESR on a shared hierarchical feature space for fast inference. Moreover, we present UFO-120, the first dataset to facilitate large-scale SESR learning; it contains over 1500 training samples and a benchmark test set of 120 samples. By thorough experimental evaluation on UFO-120 and several other standard datasets, we demonstrate that Deep SESR outperforms the existing solutions for underwater image enhancement and super-resolution. We also validate its generalization performance on several test cases that include underwater images with diverse spectral and spatial degradation levels, and also terrestrial images with unseen natural objects. Lastly, we analyze its computational feasibility for single-board deployments and demonstrate its operational benefits for visually-guided underwater robots.

show abstract

Machine Vision for Improved Human-Robot Cooperation in Adverse Underwater Conditions

Islam

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Visually-guided underwater robots are widely used in numerous autonomous exploration and surveillance applications alongside humans for cooperative task execution. However, underwater visual perception is challenging due to marine artifacts such as poor visibility, lighting variation, scattering, etc. Additionally, chromatic distortions and scarcity of salient visual features make it harder for an underwater robot to visually interpret its surroundings to effectively assist its companion diver during an underwater mission. In this paper, we delineate our attempts to address these challenges by designing novel and improved vision-based solutions. Specifically, we present robust methodologies for autonomous diver following, human-robot communication, automatic image enhancement, and image super-resolution. We depict their algorithmic details and describe relevant design choices to meet the real-time operating constraints on single-board embedded machines. Moreover, through extensive simulation and field experiments, we demonstrate how an autonomous robot can exploit these solutions to understand human motion and hand gesture-based instructions even in adverse visual conditions. As an immediate next step, we want to focus on relative pose estimation and visual attention modeling of an underwater robot based on its companion humans' body-pose and temporal activity recognition. We believe that these autonomous capabilities will facilitate a faster and better interpretation of visual scenes and enable more effective underwater human-robot cooperation.

show abstract

Fast Underwater Image Enhancement for Improved Visual Perception

Cited by 4 publications

References 0 publications

Underwater Image Super-Resolution using Deep Residual Multipliers

Underwater Image Super-Resolution using Deep Residual Multipliers

Simultaneous Enhancement and Super-Resolution of Underwater Imagery for Improved Visual Perception

Machine Vision for Improved Human-Robot Cooperation in Adverse Underwater Conditions

Contact Info

Product

Resources

About