object detection
object detection on World Data Ocean: a running collection of 5 stories we have gathered and hand-picked because they are worth your time. Every post here touches on object detection in some way — the news, the analysis, the deep dives, and the occasional surprise find. A Global Hub for Ocean Intelligence 4 World Data Ocean is a centralized digital platform where researchers, scientists, and ocean enthusiasts converge to explore, analyze, and… New stories are added to this page as we find them, so check back if you want to keep up with what is happening around object detection, or subscribe to the RSS feed to get them as soon as they are published. Browse the collection below, or head back to the homepage to see everything World Data Ocean is covering right now.
RSADAF: a Rayleigh scattering-adaptive anisotropic diffusion filter for underwater image restoration in marine ecosystems
Underwater imaging presents a significant challenge for marine ecosystem monitoring due to pervasive degradation from haze, scattering, and noise. Addressing these limitations, researchers have developed RSADAF, a Rayleigh scattering-adaptive anisotropic diffusion filter. This physics-based framework leverages a second-order partial differential equation and atmospheric scattering models to achieve superior image restoration, preserving structural details and enhancing color fidelity. Experimental results demonstrate a PSNR of 42.85 dB and SSIM of 0.
Data-augmented vision system for maritime object detection
Accurate maritime vessel detection from aerial imagery remains a significant challenge, despite advances in convolutional neural networks. Our research addresses this by introducing a novel data-augmented vision system, demonstrating over a 10% improvement in cross-sensor vessel detection precision. This system integrates multiple sensors, advanced data augmentation techniques, and diverse CNN architectures to enhance resilience. Composed of six key subsystems—from image acquisition to system validation—it establishes a foundation for robust maritime applications.
Using semi-supervised learning to detect beluga whales from aerial image sequences
Precise beluga whale population monitoring is essential for Arctic conservation, yet manual annotation of aerial imagery presents significant logistical and cost barriers. This study investigates a solution: semi-supervised learning. We systematically evaluated SEMI-DETR, a novel detection transformer, against supervised methods, demonstrating a 20% mean Average Precision improvement with just 1% of labeled data. Notably, calf detection—critical for assessing reproductive health—benefited most. These findings establish an empirical benchmark and offer practical guidance for resource-constrained conservation programs, mirroring approaches explored in our related work on plankton monitoring.
Reliability-aware query restoration for embedded real-time object detection in degraded underwater vision systems
Real-time object detection in underwater vision systems faces significant challenges due to degraded visibility. Addressing this, our research introduces DQR-RTDETR, a novel framework that restores query reliability within RT-DETR detectors, demonstrably improving performance in turbid conditions. Empirical validation on the SeaClear dataset reveals a substantial increase in mean average precision (mAP@0.5:0.95) from 0.6998 to 0.7444, alongside improved localization. This compact intervention, adding only 0.05 million parameters, enables efficient edge deployment, achieving 18.7 FPS on an NVIDIA Jetson Orin NX
Improved Transformer-based detection of underwater plastic debris in complex environments
Accurate detection of underwater plastic debris presents a significant challenge due to image degradation, small debris size, and complex backgrounds. This study introduces an improved RF-DETR detector, leveraging frequency-aware feature reweighting and adaptive query strategies to enhance the representation of challenging features. Evaluations on the TrashCan and DeepTrash datasets demonstrate superior performance compared to established baselines, achieving notable precision and mAP scores.